su.diva-portal.orgsu.diva-portal.org/smash/get/diva2:1198320/fulltext01.pdf · receptor cells do...

HUMAN OLFACTION

Ingrid Ekström

Human olfaction

Associations with longitudinal assessment of episodic memory, dementia,and mortality risk

Ingrid Ekström

©Ingrid Ekström, Stockholm University 2018 ISBN print 978-91-7797-220-4ISBN PDF 978-91-7797-221-1 Printed in Sweden by Universitetsservice US-AB, Stockholm 2018Distributor: Department of Psychology, Stockholm University

In Erinnerung an meineGrosseltern, Ingrid & Ioan.

Abstract

A declining sense of smell is a common feature in older age. Above

and beyond diminished smelling capacity due to normal processes of

human aging, impairments in olfactory function have also been linked

to numerous ill-health related outcomes, such as cognitive dysfunc-

tions, dementia pathology and even an increased risk of death. Based

on population-based data from the Swedish Betula Prospective Cohort

Study, the aim of this thesis was to further our understanding regard-

ing the role of olfaction in long-term memory decline, dementia, and

mortality. Furthermore, this thesis investigated the predictive utility of

self-reported olfactory dysfunction for assessing the risk of conversion

to later dementia and to mortality, as well as the predictive utility of

long-term subjective olfactory decline for an actual long-term decline

in odor function. Study I explored associations of olfactory deficits

with memory decline and found that impairments in an odor identifi-

cation test were related to an ongoing and long-term decline in episod-

ic memory only in carriers of the �4 allele of the Apolipoprotein E, a

genetic risk factor for Alzheimer’s disease. Study II investigated the

predictive utility of olfactory ability for conversion to common forms

of dementia in participants with intact baseline cognition during a fol-

low-up time-span of 10 years. The results showed that lower odor

identification scores, as well as subjectively assessed odor impair-

ment, were associated with an increased risk for dementia conversion,

and that the effects of objective and subjective odor function were

cumulative. Study III investigated whether olfactory ability could

predict mortality and showed that lower odor identification scores, as

well as subjective odor impairments, were associated with an elevated

risk of death within a follow-up time-span of approximately 10 years.

Crucially, this effect could not be explained by dementia conversion

prior to death. Study IV showed that a subjectively assessed long-

term and ongoing olfactory decline was predictive of an objectively

assessed long-term and ongoing decline in odor function. Subjective

olfactory impairments might thus be indicative of an actual olfactory

decline in older adults. Overall, the findings of this thesis indicate that

sense of smell is closely related to processes of memory decline and

dementia as well as mortality in older adults. Furthermore, the results

of these investigations shed a new light on the role of subjectively

experienced olfactory decline, which might reflect an actual intra-

individual change in olfactory ability in older adults.

List of studies

This doctoral thesis is based on the following studies, each referred to

throughout the text as a “Study” with their corresponding Roman nu-

meral (Studies I-IV):

I. Olofsson, J.K., Josefsson, M., Ekström, I., Wilson, D.,

Nyberg, L., Nordin, S., Nordin Adolfsson, A., Adolfsson, R.,

Nilsson, L.G., & Larsson, M. (2016). Long-term episodic

memory decline is associated with olfactory deficits only in

carriers of ApoE-ε4. Neuropsychologia, 85, 1–9.

II. Stanciu, I., Larsson, M., Nordin, S., Adolfsson, R., Nilsson, L.

G., & Olofsson, J. K. (2014). Olfactory impairment and sub-

jective olfactory complaints independently predict conversion

to dementia: a longitudinal, population-based study. Journal of

the International Neuropsychological Society, 20(2), 209–217.

III. Ekström, I., Sjölund, S., Nordin, S., Nordin Adolfsson, A.,

Adolfsson, R., Nilsson, L. G., Larsson, M., & Olofsson, J. K.

(2017). Smell loss predicts mortality risk regardless of demen-

tia conversion. Journal of the American Geriatrics Society,

65(6), 1238–1243.

IV. Ekström, I., Josefsson, M., Larsson, M., Nordin, S., Nilsson,

L.-G., & Olofsson, J.K. (2018). Subjective olfactory loss cor-

responds to long-term odor identification decline in older

adults. (Submitted to journal).

Contents

Abstract .......................................................................................................... i

List of studies .............................................................................................. iii

List of Abbreviations ................................................................................... 8

Introduction .................................................................................................. 9

The olfactory system ................................................................................ 11 Basic neurobiological principles of the human olfactory experience .............. 11 Brain structures of human olfaction ..................................................................... 14

Primary and secondary olfactory structures ................................................. 15 Behavioral assessments of human olfaction ...................................................... 16

Olfactory detection and discrimination ability............................................... 17 Olfactory identification ...................................................................................... 18

Olfaction and normative aging ............................................................... 21 Consequences of olfactory deficits ....................................................................... 23

Olfaction and cognitive decline ............................................................... 25 Olfaction and episodic memory decline ............................................................... 28 The Apolipoprotein E (ApoE) gene ....................................................................... 31

Olfaction and dementia ............................................................................ 33 Clinical and pathophysiological progression of AD ............................................ 35 Olfaction as an early marker for AD .................................................................... 36

Olfactory deficits in AD ..................................................................................... 38 Perceptual versus cognitive olfactory deficits in AD .................................... 38 Olfactory deficits predict conversion from MCI to AD ................................. 39 Olfactory deficits in VaD ................................................................................... 41 Olfactory impairment as a predictor of dementia conversion .................... 42

Olfaction and risk of mortality ................................................................ 44

Subjective olfactory impairment ............................................................ 47 Sensitivity and specificity of subjective olfactory impairment ........................ 48 Subjective olfactory impairment as a predictor for dementia ......................... 49 Long-term subjective olfactory decline ............................................................... 50

Aim of the thesis ........................................................................................ 52

Methods ....................................................................................................... 55 The Betula Prospective Cohort Study .................................................................. 55

Ethics approval ................................................................................................... 55 Procedure ............................................................................................................ 56 Study sample ...................................................................................................... 57

Assessments of olfactory function ........................................................................ 57 Cognitive assessments ...................................................................................... 61 Health assessments ........................................................................................... 62

Overview of empirical studies ................................................................. 65 Study I....................................................................................................................... 65

Aim ....................................................................................................................... 65 Method ................................................................................................................. 65 Results ................................................................................................................. 67 Conclusion ........................................................................................................... 70

Study II ..................................................................................................................... 71 Aim ....................................................................................................................... 71 Method ................................................................................................................. 71 Results ................................................................................................................. 72 Conclusion ........................................................................................................... 74

Study III.................................................................................................................... 75 Aim ....................................................................................................................... 75 Method ................................................................................................................. 75 Results ................................................................................................................. 76 Conclusion ........................................................................................................... 77

Study IV .................................................................................................................... 78 Aim ....................................................................................................................... 78 Method ................................................................................................................. 79 Results ................................................................................................................. 80 Conclusion ........................................................................................................... 81

General discussion .................................................................................... 83 Olfaction in long-term cognitive decline and dementia .................................... 85

Olfactory deficits in cognitive decline ............................................................. 85 Olfactory deficits predict progression to dementia ...................................... 86 Proposed neuropathological mechanisms ...................................................... 90 The influence of the ApoE ɛ4............................................................................ 93 A proposed timeline for the development of olfactory and memory decline in cognitive aging and dementia ....................................................... 97

Olfactory dysfunction as a predictor of mortality ............................................ 100 Subjective olfactory dysfunction ......................................................................... 102

Subjective olfaction and dementia risk ........................................................ 102 Cognitive aspects of subjective olfactory complaints ................................ 103

Subjective olfaction in prospective settings ................................................ 104 Subjective and objective long-term olfactory decline ............................... 106

Sensory deficits and brain aging, dementia, and mortality ........................... 107 Sensory deficits and mortality ....................................................................... 110

Points of caution .................................................................................................... 110 Study design ..................................................................................................... 110 Olfactory assessments .................................................................................... 114 Dementia diagnoses ........................................................................................ 116

Effect sizes and clinical considerations .............................................................. 118 Ethical considerations ........................................................................................... 121 Suggestions for future studies of smell-based cognitive training................. 122 Concluding remarks .............................................................................................. 124

Summary in Swedish .............................................................................. 126

Acknowledgements ................................................................................. 128

References ................................................................................................ 130

8

List of Abbreviations

Abbreviation

Full reference

AD

Alzheimer’s disease

aMCI

Amnestic mild cognitive impairment

ApoE

Apolipoprotein E

CES-D Center for Epidemiological Studies Depres-sion Scale

CNS

Central nervous system

CSF

Cerebrospinal fluid

DSM The Diagnostic and Statistical Manual of Mental Disorders

ENT

Ear, nose, throat

fMRI

Functional magnetic resonance imaging

MCI

Mild cognitive impairment

MMSE

Mini Mental State Examination

PET

Positron-emission tomography

SCD

Subjective cognitive decline

SOIT

Scandinavian Odor Identification Test

SRB

Synonym Reasoning Block

VaD

Vascular dementia

WHO World Health Organization

9

Introduction

“A nose that can see is worth two that sniff.” Eugene Ionesco

Historically, the human sense of smell is a neglected one. Western

philosophy has a long tradition of considering olfaction as inferior to

more “intellectual” and “refined” visual and auditory senses, which

might explain the relative lack of psychological research on olfaction

as compared to other senses (Le Guérer, 2002). Similarly, olfaction

may be regarded as the least important among our senses (Toller,

1999), not being sufficiently appreciated until it is lost (Deems et al.,

1991; Miwa et al., 2001). Recent years have, however, witnessed an

emerging appraisal of olfactory function, both in scientific settings as

well as in wider society.

A growing body of literature suggests that olfactory performance

may be indicative of processes of human aging. As such, olfactory

deficits have been linked to various outcomes of ill-health, such as

cognitive decline (Graves et al., 1999; Schubert et al., 2008; Swan &

Carmelli, 2002), neurodegenerative diseases (Albers, Tabert, &

Devanand, 2006; Mesholam, Moberg, Mahr, & Doty, 1998), depres-

sion (Croy & Hummel, 2017; Kohli, Soler, Nguyen, Muus, &

Schlosser, 2016; Negoias et al., 2010), and even an increased risk of

mortality (Devanand et al., 2015; Gopinath, Kifley, & Mitchell, 2011;

Pinto, Wroblewski, Kern, Schumm, & McClintock, 2014). With aging

10

societies world-wide, prevalence rates of age-related diseases such as

dementia are soon reaching epidemic proportions and finding early

markers for the identification of an impending neurodegeneration are

among the top scientific priorities of this century (Winblad et al.,

2016). Prior research suggests that sense of smell may serve as a win-

dow into the process of brain aging and in certain situations even warn

of impending pathologies. Based on a longitudinal approach, the over-

all aim of this thesis was to wider our understanding of the role of

olfaction in processes of cognitive aging and dementia as well as mor-

tality.

11

The olfactory system

Basic neurobiological principles of the human olfactory experience Several reviews have thus far given a detailed summary of the elabo-

rate neuronal processes underlying olfactory sensations (Gottfried,

2010; Lodovichi & Belluscio, 2012; Nagayama, Homma, & Imamura,

2014; Wilson & Mainen, 2006; Zou, Chesler, & Firestein, 2009). An

increasing amount of evidence suggests that the types of neurons

found within olfactory structures are among the most diverse and that

the mechanisms underlying olfactory processing are by far more com-

plex than earlier anticipated (Nagayama et al., 2014; Shipley & Ennis,

1996). As a detailed description of these mechanisms is outside the

scope of this thesis, the following section will provide a brief, and

therefore simplified, overview of the most important neurobiological

principles underlying olfactory sensations.

To begin, odorants, in the form of chemical molecules, reach the ol-

factory epithelium at the top of each nasal cavity. The olfactory epi-

thelium consists of several million receptor cells that selectively react

to different odorants. While it is estimated that between 350 and 400

different functional receptor proteins are present in the cilia of the

receptor cells of the human olfactory epithelium (Doty & Kamath,

2014; Nagayama et al., 2014; Rouquier, Blancher, & Giorgi, 2000),

each ciliary membrane of a given receptor cell encompasses only one

12

type of receptor protein (Chess, Simon, Cedar, & Axel, 1994). Most

receptor cells do however not solemnly respond to one single odorous

ligand but can be activated by a range of different odor molecules

(Ressler, Sullivan, & Buck, 1994; Sicard & Holley, 1984). As a con-

sequence, each olfactory sensation is associated with a unique pattern

of activity in overlapping receptor cells (Cleland, Johnson, Leon, &

Linster, 2007; Johnson & Leon, 2007).

Once an action potential is provoked within the receptor cells of the

epithelium, the olfactory input is transmitted to the olfactory bulbs.

Histologically, the olfactory bulb is divided into multiple layers, com-

posed of morphologically distinct cells (Pinching & Powell, 1971;

Price & Powell, 1970a, 1970b). These layers form sophisticated net-

works to process olfactory input before forwarding it on to the cortical

structures. The neurons of the olfactory bulb are conventionally cate-

gorized based on the layer in which their nuclei are located. Tufted

cells of the glomerular layer receive olfactory input from the receptor

cells of the epithelium. They interact with mitral cells that are able to

directly forward input to the cortical areas of the brain. The axons of

the mitral and tufted cells of the olfactory bulbs combine on each side

and together form the olfactory tracts (Gottfried, 2010). In turn, the

olfactory tracts forward the olfactory input ipsilaterally (affecting the

same side of the body) to numerous areas of the brain within the

frontal and temporal lobes. Collectively, the projection sites of the

olfactory tracts make up the “primary olfactory cortex” (Gottfried,

2010). Thus, olfactory sensory activity is transferred directly from the

olfactory bulbs to the cortical areas, without a relay through the thal-

amus (Gottfried, 2010).

13

After being intercepted by the receptors in the nasal cavity, olfactory

input must thus only cross one synapse in the olfactory bulbs before

reaching emotional or cognitive areas in the brain (Wilson & Mainen,

2006). While there is a subsequent transmission from the olfactory

cortex to the thalamus, this transthalamic projection is, however, not

essential in order for olfactory sensory information to reach the corti-

cal areas. In this aspect, the processing of olfactory sensations differs

fundamentally from that of other sensations, which rely on projection

through the thalamus in order to reach the neocortices. As a conse-

quence, and in contrast to vision or hearing, olfaction provides a rela-

tively direct input to medial temporal areas of the brain that are essen-

tially involved in the processing of episodic recollection (Arshamian

et al., 2013; Olofsson et al., 2014). This may perhaps explain why

smells often evoke instant, vivid and emotional memories of previous

experiences (Willander & Larsson, 2006).

Another aspect that is fundamentally different from other sensory

modalities is the fact that receptor neurons of the olfactory epithelium

are the only neurons that are directly exposed to the environment. This

property may render the peripheral olfactory system especially sensi-

tive to damage. Previous studies have in fact shown that the olfactory

epithelium is vulnerable to the effects of viral or bacterial infection,

which in some cases may even cause irreparable deficits in olfactory

function (Doty & Mishra, 2001; Murphy, Doty, & Duncan, 2003).

Furthermore, olfactory sensations are the only sensory experiences

that rely on neural cell regeneration. The receptor neurons within the

olfactory epithelium have a life span of approximately 30 to 60 days

14

and are throughout the human lifespan replaced by new neurons (Doty

& Kamath, 2014). Apart from the olfactory epithelium, considerable

neural regeneration has also been observed for nerve cells within the

olfactory bulb (Bédard & Parent, 2004; LaMantia & Purves, 1989).

The unique plasticity of the olfactory system may also render it es-

pecially sensitive to the effects of aging. Age-related processes have

been shown to adversely affect the success of cell regeneration on

which the peripheral olfactory system relies (Watabe-Rudolph et al.,

2011). Furthermore, the plasticity of the olfactory system has been

shown to be influenced by the degree of cumulative damage from pri-

or confrontations with environmental agents. As the olfactory system

is directly exposed to the environment, air-borne agents such as virus-

es, bacteria, toxins, air-pollution and other xenobiotics can enter the

brain via the nasal cavities where they may damage the olfactory epi-

thelium. These environmental hazards accumulate with increasing age

and are thus likely to have more functional consequences in later years

(Doty, Petersen, Mensah, & Christensen, 2011; Hirai et al., 1996; Loo,

Youngentob, Kent, & Schwob, 1996).

Brain structures of human olfaction Based on advancements in functional neuroimaging methods such as

functional magnetic resonance imaging (fMRI) and positron-emission

tomography (PET), a constantly expanding body of research has been

able to identify neural correlates of olfactory experience in humans

(Gottfried, 2010, 2015; Sobel, Johnson, Mainland, & Yousem, 2003;

Zelano et al., 2005).

15

Primary and secondary olfactory structures

As mentioned above, the olfactory system is typically categorized into

peripheral structures outside the brain (e.g. the olfactory epithelium)

and central structures within the brain (e.g. the olfactory bulb and the

olfactory cortex; Doty & Kamath, 2014). Apart from this distinction,

the olfactory system is sometimes further classified into primary and

secondary brain structures. The term primary olfactory region refers

mainly to the olfactory bulb. The olfactory bulb represents the earliest

cortical stage of olfactory processing where the molecular features of

an odorant are most immediately represented.

Research using macaque monkeys indicates that the olfactory bulb

projects onto the following secondary cortical areas: the anterior ol-

factory nucleus, the olfactory tubercle, the periamygdaloid cortex, the

anterior cortical nucleus, the piriform cortex, and the anteromedial

part of the entorhinal cortex (Carmichael, Clugnet, & Price, 1994). In

humans, functional neuroimaging studies have shown that olfactory

tasks are associated with activity in all of these brain areas (Seubert,

Freiherr, Djordjevic, & Lundström, 2013) as well as with activation of

the orbitofrontal cortex (for a systematic review see Gottfried & Zald,

2005). In the case of emotionally meaningful or salient odors, the

amygdala may also be activated (Winston, Gottfried, Kilner, & Dolan,

2005).

Tasks relying on the interaction of olfactory stimuli with compo-

nents of language processing have shown to activate the temporal pole

in the anterior temporal lobe (Olofsson et al., 2014). This region, lying

between the orbital frontal cortex and the amygdala, has been pro-

16

posed to serve as an interface between olfaction and language

(Olofsson & Gottfried, 2015).

Behavioral assessments of human olfaction

Similar to other sensory or cognitive functions, performance in olfac-

tion can be behaviorally assessed. Today, numerous olfactory

measures are available, attempting to map different sub-functions of

the sense of smell. Distinctions are typically made between peripheral

measures of olfactory sensory functions (for example the ability to

detect weak odors) and assessments that also involve memory pro-

cesses and other higher cognitive components, such as the identifica-

tion or episodic recollection of odors. Despite this distinction, it is

important to emphasize that some degree of memory processing prob-

ably is involved in all olfactory tasks. First, consciousness itself can

be considered a form of memory, likely playing a role in all psycho-

logical assessments (Doty & Kamath, 2014). Second, even measures

of sensory olfactory tasks are dependent on memory performance as

they typically require participants to compare multiple odorous stimuli

with each other either in intensity or in quality. Hence, sensory olfac-

tory assessments also require intact recollection abilities of previously

presented stimuli.

The following section will summarize the basic principles of the

three standardized olfactory measures that are most commonly em-

ployed in both clinical and research environments and that are relevant

to the aims of this thesis: olfactory detection, discrimination and iden-

tification. A special focus is placed on measures of odor identification

17

because as it is the primary task used in empirical studies on olfaction

and human aging.

Olfactory detection and discrimination ability

Olfactory detection sensitivity, also commonly described as odor

thresholds, refers to the ability to correctly detect odorous concentra-

tions. Odor sensitivity is often measured through procedures that are

largely inspired by psychophysicist Gustav Fechner’s “method of lim-

its” (as described in detail in Gescheider, 1997). For example, the

standardized Sniffin’ Sticks battery test (Hummel, Kobal, Gudziol, &

Mackay-Sim, 2007; Hummel, Sekinger, Wolf, Pauli, & Kobal, 1997;

Kobal et al., 1996) employs a procedure in which the same odorant is

presented to participants in gradually stronger concentrations via felt-

tip pens. The task of the participant is to determine if the presented

stimuli contains an odor when compared to two blank stimuli that are

presented in random order during the same trial. To eliminate poten-

tial effects of chance, the participants’ olfactory threshold is deter-

mined based on the weakest odorous concentration that he or she

could reliably detect six times consecutively.

Olfactory discrimination is typically defined by the ability to cor-

rectly identify the odorous stimulus that slightly differs in quality from

other identical odorous stimuli (Doty & Kamath, 2014; Kobal et al.,

2000, p. 200). In the Sniffin’ Sticks battery test, discrimination ability

is, for example, measured using a forced-choice task involving three

alternatives where the participant is presented consecutively with two

identical and one different odor. In order to overcome potential con-

founding bias due to semantic association (e.g. the participant may

18

label the odors with a name and then recollect and compare the labels

of the stimuli; Jönsson, Møller, & Olsson, 2011), slightly unfamiliar

odorants that are difficult to label may be employed (Hummel et al.,

2007).

Olfactory identification

Tests of olfactory identification are the most widely used psychologi-

cal tests of olfactory ability (Doty & Kamath, 2014). Odor identifica-

tion ability is characterized by the correct pairing of an odor’s label to

an odorous stimulus. Odor identification can either be assessed in con-

texts were cues are presented to the participant (cued identification,

also called matching) or in contexts were no such cues are presented

(free identification, also called naming). Cued odor identification is

the most common olfactory method of choice in scientific studies and

can for example consist of presenting an odor followed by a sheet with

four written label or picture alternatives, of which one alternative is

correct (Doty, Marcus, & William Lee, 1996; Doty, Shaman, Kim-

melman, & Dann, 1984; Thomas Hummel et al., 1997; Kobal et al.,

1996; Krantz et al., 2009; Nordin, Brämerson, Liden, & Bende, 1998).

Odor identification tests typically involve odorants that are universally

recognized in a given cultural context. They are therefore often adjust-

ed to contain familiar everyday odorants and, if applicable, response

alternatives appropriate to the culture in which the testing is per-

formed.

A certain level of olfactory sensitivity is obviously needed in order

to successfully recognize the odors in an olfactory identification task.

However, while detection sensitivity mainly relies on low-level per-

19

ceptual processes (Hedner, Larsson, Arnold, Zucco, & Hummel,

2010), odor identification also engages high-level cognitive processes,

hence activating additional cerebral resources (Bohnen et al., 2010;

Dulay, Gesteland, Shear, Ritchey, & Frank, 2008; Larsson, Nilsson,

Olofsson, & Nordin, 2004; Royet, Koenig, Paugam-Moisy, Puzenat,

& Chasse, 2004). Odor identification represents a form of semantic

memory task (the general knowledge about the world that we acquire

through our lifetimes; Tulving 1972) in that it poses demands on the

participant’s prior knowledge of the presented odor (Hedner, Larsson,

et al., 2010; Richardson & Zucco, 1989). As such, it is correlated with

tests of general semantic knowledge and verbal fluency (Larsson,

Finkel, & Pedersen, 2000; Larsson et al., 2004; Larsson, Öberg, &

Bäckman, 2005). Executive functions have also been shown to influ-

ence olfactory identification performance, while such processes are

not required in tasks of olfactory sensitivity (Dulay et al., 2008;

Hedner, Larsson, et al., 2010).

Although different olfactory domains map on a common source of

variance (Doty, McKeown, Lee, & Shaman, 1995; Doty, Smith,

McKeown, & Raj, 1994), a previously conducted principal component

analysis suggests relative independence between measures of odor

identification on the one hand and tests of odor thresholds on the other

hand (Lötsch, Reichmann, & Hummel, 2008). Tests of odor identifica-

tion are therefore not interchangeable with tests of perceptual odor

function. As such, it has been posited that the integration of olfactory

information with processes of language and semantic memory might

impose a unique challenge for the brain, mapping on distinct neural

structures within the medial temporal lobe (Olofsson et al., 2014). As

20

will be outlined in later chapters, olfactory identification ability could

play a notable role in processes of cognitive aging and dementia.

21

Olfaction and normative aging

Aging is often accompanied by a general decline in sensory functions.

For example, deficits in hearing (e.g. Cruickshanks et al., 1998;

Dalton et al., 2003) and vision (e.g. Taylor et al., 2005) are common

in older adults. Likewise, olfactory performance declines as a function

of age. The unique features of the olfactory system regarding its anat-

omy, physiology, and neuroplasticity may render the sense of smell

especially sensitive to processes of aging (for a detailed overview

about the neurobiological correlates of age-related olfactory changes

see Doty & Kamath, 2014).

Quantifiable alterations in olfactory function may encompass par-

tial (hyposmia) or complete (anosmia) olfactory loss. Epidemiological

estimations of prevalence of olfactory dysfunction in aging popula-

tions range from between 30% to up to 70%, depending on the olfac-

tory measure that is employed and the age-range that is examined

(Doty, Shaman, Applebaum, et al., 1984; Landis, Konnerth, &

Hummel, 2004; Claire Murphy et al., 2002). Estimations regarding

alterations in odor identification in adults aged 53 and older suggest

that approximately 14% to 33% are affected (Brämerson, Johansson,

Ek, Nordin, & Bende, 2004; Claire Murphy et al., 2002; Wehling,

Nordin, Espeseth, Reinvang, & Lundervold, 2011). The effects of age

on olfactory ability have so far been mainly investigated by cross-

sectional comparisons, which consistently have found age-related dys-

22

functions in all olfactory domains such as in odor sensitivity (Cain &

Gent, 1991; Larsson et al., 2000; Schiffman, Moss, & Erickson, 1976;

Stevens & Cain, 1987; Strauss, 1970), odor quality discrimination

(e.g. De Wijk & Cain, 1994; Kaneda et al., 2000; Schemper, Voss, &

Cain, 1981), and odor identification and recognition (e.g. Doty, Sham-

an, Applebaum, et al., 1984; Larsson et al., 2004; Lehrner, Glück, &

Laska, 1999; Menon, Westervelt, Jahn, Dressel, & O’Bryant, 2013;

Claire Murphy et al., 2002; Claire Murphy, Cain, Gilmore, & Skinner,

1991; Olofsson et al., 2010). Likewise, olfactory abilities have also

been found to decline as a function of age in the few studies that have

followed the same adults longitudinally (Hedner, Nilsson, et al., 2010;

Schubert, Cruickshanks, Klein, Klein, & Nondahl, 2011).

As both peripheral and higher olfactory functions decline with in-

creasing age, an essential question is whether age also affects olfacto-

ry identification processes directly, or if deficits in the identification

of smells are mainly secondary products of a peripheral sensory loss.

Regarding this issue, it is important to highlight that olfactory sensi-

tivity has previously been shown to only explain a relatively small

part of the variation in odor identification in older adults, indicating

that age-related decline in odor identification may not be a secondary

consequence of deficits in perceptual olfactory function, but rather

evolve due to age-related difficulties in retrieving semantic odor

knowledge (Doty & Kamath, 2014; Larsson et al., 2005).

It is important to note at this point that not all olfactory alterations

found in older adults are exclusively attributable to advancing age.

Apart from normal aging and neurodegeneration, olfactory deficits are

often caused by sinonasal diseases, upper respiratory infections or

23

head traumas (Doty & Kamath, 2014). However, increasing age may

in some cases interact with other underlying pathological reasons for

olfactory impairment. For example, the olfactory system may be more

vulnerable to the accumulative effects of sinonasal or respiratory con-

ditions in older, as compared to younger, individuals (Doty &

Kamath, 2014).

Consequences of olfactory deficits Suffering from an olfactory dysfunction has been associated with a

number of detrimental outcomes. Many older persons complain that

food lacks flavor (Schiffman et al., 2000) resulting in a loss of appetite

(Boyce & Shone, 2006) and changes in dietary behaviors

(Aschenbrenner et al., 2008; Sergi, Bano, Pizzato, Veronese, &

Manzato, 2017). Changes in appetite related to olfaction may there-

fore lead to an increased risk of malnutrition in older persons. While

previous research could not find a direct link between olfactory defi-

cits and nutritional status, it has been suggested that olfactory altera-

tions may still be a risk factor for malnutrition as they often co-occur

with other mental and physical problems (Toussaint, de Roon, van

Campen, Kremer, & Boesveldt, 2015).

Notably, nutritional factors likely play an essential role for neu-

rocognitive development in older age. Dietary components such as

vitamins E and B, as well as omega-3 fatty acids, may be protective

against cognitive decline and different forms of dementia. In contrast,

a low intake of these nutritional elements may have adverse effects for

brain function in older age (for an overview see Morris, 2012).

24

Olfactory deficits have furthermore been linked to depressive symp-

toms (Croy et al., 2014; Negoias et al., 2010). At this point, it is how-

ever unknown if olfactory impairments may actually contribute to the

development of depression or if common causes, such as shared

pathological processes, may explain the emergence of both. In any

case, previous studies have shown that persons with an olfactory defi-

cit do in fact experience a diminished quality of life (Blomqvist,

Bramerson, Stjarne, & Nordin, 2004; Deems et al., 1991; Miwa et al.,

2001; Veldhuizen, Smeets, & Veldhuizen, 2009), indicating that an

intact sense of smell is important for well-being in older age.

25

Olfaction and cognitive decline

Apart from being a common feature of normal aging processes, olfac-

tory deficits are closely related to cognitive function in older age. To

date, several studies have demonstrated noteworthy associations of

olfactory impairment with poorer cognition in older age, indicating

that smell assessments may be indicative of early age-related brain

changes. Typically, previous studies investigating the sense of smell

in cognitive aging have assessed whether deficits in odor identifica-

tion at baseline were correlated with concurrent cognitive impairments

or prospective cognitive dysfunctions. Interestingly, these studies have

found that olfactory deficits often coincide with or even precede im-

pairments in non-olfactory cognitive tests (e.g. Djordjevic, Jones-

Gotman, De Sousa, & Chertkow, 2008; Dulay & Murphy, 2002;

Graves et al., 1999; Olofsson et al., 2009; Swan & Carmelli, 2002;

Robert S. Wilson, Arnold, Tang, & Bennett, 2006). Table 1 gives a

summary of previously published prospective studies on olfactory

deficits and age-related cognitive changes.

Previous research suggests that correlations between olfactory im-

pairments and cognitive deficits in older age are unlikely to arise

simply due to semantic memory components in the olfactory identifi-

cation task. Rather, olfactory impairments have been found to be asso-

ciated with a pronounced decline in general cognitive ability in older

adults (Conti et al., 2013; Graves et al., 1999; Olofsson et al., 2009).

26

Table 1. Description and main results of published longitudinal, prospective stud-ies, in chronological order, investigating olfactory functions and age-related cogni-tive changes (adapted from Study I). Study Participants

(appx. mean age)

Odor test (items/levels)

Study duration (years)

Main results

Bacon et al., 2008

70 memory impaired (75 years)

BOT (10) 1 Odor threshold deficit predict DAT

Graves et al., 1999

1604 adults (74 years)

B-SIT (12) 2 Odor id. deficit interacts with E4+ to predict general cognitive decline

Devanand et al., 2000

77 MCI, 45 C (65 years)

SIT (40) 2 DAT risk higher in MCI patients with unnoticed olfactory loss

Royall et al., 2002


SIT (40) 3 Odor id. deficit predicts general cognitive and executive decline

Swan & Camelli, 2002


B-SIT (12) 4.5 Odor id. deficits predict decline in episodic/verbal memory

Tabert et al., 2005

147 MCI, 100 DAT, 64 C (68 years)

SIT (10) ≤3.5 A reduced 10-item odor id. test predicts DAT conversion in MCI

Calhoun-Haney & Murphy, 2005


SDOIT 3 E4+ predicts odor id. decline, but not global cognitive or odor threshold decline

Wilson et al., 2006


B-SIT (12) 3 Odor id. deficit predicts decline in episodic memory and speed

27

Table 1 (continued) Study Participants

(appx. mean age)



Main results

Wilson et al., 2007


B-SIT (12) ≤5 Odor id. deficit predicts conversion to MCI, and rate of decline in episodic and semantic memory, and speed


148 MCI (67 years)

SIT (40) 3 A combination of lower odor id., cognitive function, hippocampus and entorhinal cortex vol-ume predicts DAT

Schubert et al., 2008

1920 (67 years)

SDOIT (8) 5 Odor id. deficit predicts cognitive im-pairment

Olofsson et al., 2009


SOIT (13) 5 E4+ interacts with odor id. deficit to predict general cognitive decline

Olofsson et al., 2010


SOIT (13) 5 Odor id. deficit predicts dementia conversion, but cannot account for effects of E4+ on odor id.

Finkel et al., 2011


NGOIT (6) ≤16 Odor id. deficit and E4+ independently predicts verbal, memory and speed decline

Sohrabi et al., 2012


SSTT (16), SSIT (16), SSDT (16)

3 years Odor discr. deficit predicts cognitive Impairment

Conti et al., 2013

88 MCI, 46 C (74 years)

CA-SIT (34) 2 years Odor id. deficit predicts dementia in MCI

28

Table 1 (continued) Study Participants

(appx. mean age)



Main results

Lipnicki et al., 2013


B-SIT (12) 2 years Odor id. deficit predicts decline in executive function, memory, dementia, and MCI and dementia

Stanciu et al., 2014


SOIT (13) 10 years Odor id. and subjective olfactory impairment independently predict dementia

Kjelvik et al., 2014

12 MCI, 6 early DAT, 30 C (70 years)

B-SIT (12), SSIT (16), SSDT (16)

0.5-1.5 years

Odor id. deficit pre-dicts dementia



SIT (40) 2-4 years

Odor id. deficit predicts general cognitive decline and dementia

Roberts et al., 2015

1630 C (80 years)

B-SIT (12) 1-6 years

Odor id. deficit predicts amnestic MCI and dementia

Abbreviations: DAT: Dementia of the Alzheimers' type; BOT: Butanol olfactory threshold test (Murphy, Gilmore, Seery, Salmon, & Lasker, 1990); CA-SIT: Cul-turally adapted version of the SIT (Parola & Liberini, 1999); E4+: Carrying at least one ApoE-ɛ4 allele; C: Control participants; MCI: Mild cognitive impairment (Petersen et al., 1999); SDOIT: San Diego odor identification test (Morgan & Murphy, 2002); SIT: (University of Pennsylvania) smell identification test (Doty, Shaman, Kimmelman, & Dann, 1984); SOIT: Scandinavian odor identification test (Nordin, Bramerson, Liden, & Bende, 1998); SSDT: Sniffin' sticks discrimination test; SSIT: Sniffin' sticks identification test; SSTT: Sniffin' Sticks threshold test (Hummel, Kobal, Gudziol, & Mackay-Sim, 2007).

Olfaction and episodic memory decline Apart from predicting a general cognitive decline, associations be-

tween olfactory deficits and cognitive abilities are especially common

in cognitive functions that typically decline in Alzheimer’s disease

29

(AD). For example, several studies have indicated that olfactory loss

might be particularly associated with future dysfunctions in tasks as-

sessing episodic memory, which are early symptoms of AD (Finkel,

Reynolds, Larsson, Gatz, & Pedersen, 2011; Roberts et al., 2015;

Swan & Carmelli, 2002; Wilson, Arnold, Tang, & Bennett, 2006).

Episodic memory is a form of declarative memory and has been con-

ceptualized as the neurocognitive system that underlies all remem-

brance of past experiences by humans. As such, episodic memory

function encompasses all conscious remembrance of events that are

situated in time and place (Tulving, 1972). The remembrance of epi-

sodes is a complex process. Each episodic memory is the result of

elaborate interactions between many pieces of information. Recall, for

example, your last vacation trip and you will likely remember the

place you visited, the foods you ate, the different smells you encoun-

tered, the persons you travelled with, including their names and faces,

amongst other details.

Episodic memory ability often declines as a function of age and the

change is typically more pronounced than age-related effects in se-

mantic memory (Lövdén et al., 2004; Naveh-Benjamin, Hussain,

Guez, & Bar-On, 2003; Lars Nyberg et al., 2003; Rönnlund, Nyberg,

Bäckman, & Nilsson, 2005). Notably, dysfunctions in episodic

memory are one of the first symptoms of dementia and characteristic

of AD (Bäckman, Small, & Fratiglioni, 2001; Greene, Baddeley, &

Hodges, 1996).

The exact mechanisms behind the interaction between episodic

memory and olfactory deficits in older age remain a matter of specula-

tion. Olfaction is intimately associated with episodic recollection, as

30

familiar smells often evoke vivid, emotional memories of childhood

experiences (Willander & Larsson, 2006). A mentioned earlier, the

olfactory system can forward odor information to neocortical areas of

the medial temporal lobes without having to rely on thalamic delay

and may thus represent a direct route to episodic memory systems.

Furthermore, olfactory and memory brain areas are intimately con-

nected in the medial temporal lobe, sharing many important neural

substrates (Carmichael et al., 1994; Eichenbaum, 2000; Kjelvik,

Evensmoen, Brezova, & Haberg, 2012; Claire Murphy, Jernigan, &

Fennema-Notestine, 2003; Segura et al., 2013). Olfactory identifica-

tion and episodic memory functions might thus activate a shared space

in the medial temporal lobes that may also be jointly affected by the

effects of brain aging and pathology (Wilson, Arnold, Schneider,

Tang, & Bennett, 2007). This mechanism would explain why the per-

formance levels on some olfactory-based and visual-based cognitive

tasks tend to decline simultaneously in old age (Dulay & Murphy,

2002).

In summary, previous research has consistently identified perfor-

mance in odor identification as a predictor of future cognitive dys-

function. However, an important aspect relating to the interaction be-

tween olfactory and cognitive deficits has so far received little atten-

tion. While previous studies have typically focused on olfactory defi-

cits as a predictor of either ongoing or future cognitive dysfunction, no

previous study has investigated cognitive changes prior to the olfacto-

ry assessment using longitudinal methods. Thus, little is known about

how long-term cognitive changes might develop prior to, in parallel

with, or after olfactory deficits. Individuals, especially older adults,

31

vary widely in their cognitive abilities as well as in their cognitive

decline trajectories (Wilson et al., 2002). However, it is critical to note

that these variables are not exchangeable. The rate of cognitive decline

is of key importance in order to understand age-related processes in

the brain (Persson et al., 2006). Cross-sectional cognitive assessments

from only one given point in time might confound cognitive decline

with poor, but stable, baseline cognitive performance. Thus, longitu-

dinal studies are needed in which cognitive function can be measured

at multiple testing occasions (Nyberg, Lövdén, Riklund, Linden-

berger, & Bäckman, 2012).

Mapping the temporal context in which olfactory and cognitive de-

cline develop in relation to one another may be of particular im-

portance regarding cognitive deficits that are indicative of an impend-

ing dementia disorder, such as disturbances in episodic memory. A

better understanding of olfactory deficits in the context of long-term

episodic memory decline might thus further our knowledge about the

relevance of olfaction as a potential marker for dementia risk.

The Apolipoprotein E (ApoE) gene

It is noteworthy that prior heritability analyses suggest that interac-

tions between olfactory and non-olfactory cognitive abilities might not

follow a universal pattern in all older people, but may be partially in-

fluenced by genetic risk factors for dementia (Doty et al., 2011;

Finkel, Pedersen, & Larsson, 2001). So far, a few reports suggest that

the ɛ4 allele of the Apolipoprotein E gene might play a notable role in

the interaction between olfactory and cognitive impairments (Bacon,

32

Bondi, Salmon, & Murphy, 1998; Finkel, Reynolds, Larsson, Gatz, &

Pedersen, 2011; Graves et al., 1999; Olofsson et al., 2009). The ɛ4

allele is present in approximately 10% to 15% of individuals and in-

creases the risk for dementia of the Alzheimer’s type while it also

lowers the age at onset of the disease. Carriers of one copy of the ɛ4

allele have a risk of AD conversion that is two to three times that of

non-carriers, while carriers of two alleles have a 12-fold increased risk

of conversion (Michaelson, 2014).

As of today, the ɛ4 allele is the most well established genetic risk

factor for cognitive dysfunction and AD among older adults (Bertram

& Tanzi, 2008; Corder et al., 1993; Davies et al., 2015). As such, cog-

nitive impairments in ɛ4-carriers may be interpreted as preclinical

symptoms of AD. As the ɛ4 allele, deficits in episodic memory, and

olfactory dysfunction all have been linked to dementia progression, it

is important to understand their relationship to one another. However,

little is known so far about how the ɛ4 allele might influence interac-

tions between ongoing memory decline and olfactory impairment in

aged adults.

33

Olfaction and dementia

As described in the previous chapter, olfactory deficits are closely

related to the process of brain aging and cognitive dysfunction. Nota-

bly, olfactory alterations may be indicative of an impending Alz-

heimer’s pathology. During the past thirty years, a growing body of

literature has investigated olfactory impairments in people with neu-

rodegenerative disorders and their possible role as an early warning

for dementia progression.

Amongst aging populations, the prevalence of common types of

dementia is increasing dramatically. Whereas around 46 million peo-

ple worldwide were estimated to be affected by a dementia ailment in

2015, this number is expected to reach 75 million by 2030 and 131

million by 2050 (Winblad et al., 2016). According to the World

Health Organization (WHO), these numbers reflect a rise of epidemic

proportions in the number of people afflicted with dementia (WHO,

2015). The Diagnostic and Statistical Manual of Mental Disorders

(DSM) classifies dementia as a major neurocognitive disorder, which

may be caused by many different neuropathological etiologies. On a

behavioral level, dementia manifests itself in both severe as well as

gradually worsening impairments in, amongst other areas, learning

and memory, language, executive function, attention, perception, and

social cognition (DSM). Given the extensive and substantial impact of

dementia on the daily lives of those affected, including family mem-

34

bers and wider society, finding a cure and increasing treatment possi-

bilities are regarded as one of the top public health priorities of this

century (WHO, 2012; Winblad et al., 2016).

By far the most common cause of dementia, and also the most re-

searched dementia type, is AD, which accounts for an estimated 60%

to 80% of diagnosed cases (Thies & Bleiler, 2013; Winblad et al.,

2016). Of particular interest are the neuropathologic abnormalities in

olfactory brain areas that are a common characteristic of AD (Braak &

Braak, 1995). The second most common cause of dementia, account-

ing for up to 25% of cases, is vascular dementia (VaD; Lobo et al.,

2000). A few previous studies have associated VaD with olfactory

deficits. These deficits may, however, be less pronounced than those

found in AD (Knupfer & Spiegel, 1986). Previous research studies

have also linked pronounced olfactory dysfunction to some of the less

common dementia forms, such as different subtypes of frontotemporal

dementia (Luzzi et al., 2007; McLaughlin & Westervelt, 2008;

Olofsson, Rogalski, Harrison, Mesulam, & Gottfried, 2013; Omar,

Mahoney, Buckley, & Warren, 2013; Orasji, Mulder, de Bruijn, &

Wirtz, 2016) and to dementia with Lewy Bodies (McShane et al.,

2001; Olichney et al., 2005; Williams et al., 2009).

The following review of olfactory impairment in the context of de-

mentia will however be restricted to AD and VaD as they are the most

relevant dementia types for the aims of this thesis. AD will receive

special attention as it is the most common form of dementia and also

the most researched dementia type in relation to olfactory function.

35

Clinical and pathophysiological progression of AD

Alzheimer’s disease is a progressive form of dementia characterized

by a steady worsening and accumulation of symptoms, reflecting the

gradual degeneration of affected brain areas (Dubois et al., 2007,

2010). Clinically, AD typically manifests in slow changes to episodic

memory function, which are often too subtle to be detected during the

earliest stages of the disease (Jessen, Wolfsgruber, et al., 2014). These

initial symptoms are gradually followed by more severe memory dis-

ruptions, changes in personality, language deficits, and motor impair-

ments (Thies & Bleiler, 2013; Winblad et al., 2016).

Although the exact cause of AD remains elusive until this day, it is

known that the disease is correlated with a number of progressive

changes in and between nerve cells. So far, AD has primarily been

associated with the accumulation of so-called amyloid plaques that

affect synaptic transmissions between neurons and with the develop-

ment of neurofibrillary tangles within nerve cells (e.g. Hardy &

Higgins, 1992; Theofilas et al., 2018). Other common features of AD

are neuronal loss, including loss of functional synapses that connect

different brain areas (e.g. Austin, Qiang, & Baas, 2017; Theofilas et

al., 2018), and inflammatory processes (Heneka & O’Banion, 2007;

Rios et al., 2013).

It is important to note that the patterns in which pathophysiological

processes spread within the brains of AD patients are far from ran-

dom. The stepwise progression of pathology in AD can be classified

into the following stages, with each stage relating to corresponding

neuropsychological dysfunction associated with the affected brain

areas: Braak I – transentorhinal region; Braak II – entorhinal region;

36

Braak III – temporo-occipital gyrus; Braak IV – temporal cortex;

Braak V – peristriatal cortex; and Braak VI – striatal cortex (Braak &

Braak, 1991; Nelson et al., 2012). Of particular interest are the regions

of the brain that are especially sensitive to early effects of AD-

pathology, specifically the cortical and sub-cortical medial temporal

lobe areas that receive and process olfactory input. Studies based on

post-mortem autopsies have shown that the olfactory bulb is one of

the first subcortical regions to be affected by AD-related tau neurofi-

brillary tangles (Attems, Lintner, & Jellinger, 2005; Attems, Thomas,

& Jellinger, 2012; Braak & Braak, 1995; Kovacs, Cairns, & Lantos,

1999). Furthermore, olfactory alterations appear to be correlated with

the amount of neurofibrillary tangles in primary and secondary olfac-

tory brain areas already in early stages of AD (Attems et al., 2012;

Franks, Chuah, King, & Vickers, 2015; Wilson et al., 2007).

Olfaction as an early marker for AD

Two recently developed and coexisting frameworks have been pro-

posed to divide the course of AD into three stages. According to the

International Working Group (IWG; Dubois et al., 2007, 2010), de-

mentia of the AD type is characterized first by an asymptomatic at-

risk stage of AD, which is evidenced by biomarkers but no cognitive

symptoms (1), followed by prodromal AD with episodic memory def-

icits and biomarkers (2) and lastly a diagnosis of AD dementia includ-

ing biomarkers (3). Similarly, the US National Institute on Aging-

Alzheimer’s Association (NIA-AA) group proposes a preclinical stage

of AD with evidence of biomarkers but no cognitive impairment (1),

37

then a second stage represented by MCI due to AD with cognitive

impairment and biomarkers (2), and lastly dementia due to AD with

biomarkers for AD (3) (Jack et al., 2011).

As of today there is no cure for AD. Available treatment possibili-

ties, predominantly in the form of anti-inflammatory medications, may

slow down disease progression if administered at an early stage

(Breitner et al., 1995; Etminan, Gill, & Samii, 2003; Kumar & Singh,

2015; McGeer, Schulzer, & McGeer, 2001; Vellas et al., 2011). Indi-

viduals at the prodromal stages of AD represent the most important

population for targeted dementia prevention trials (Jessen, Amariglio,

et al., 2014). However, given the long preclinical time span of AD,

neuronal loss and irreversible cognitive impairment may have already

occurred at these stages (Jessen, Amariglio, et al., 2014). Thus, a criti-

cal challenge is a better characterization of the preprodromal or pre-

MCI stage of AD, which, at the moment, is only defined by the occur-

rence of biomarkers. At this stage, neuronal damage may still be mild

and functional ability still sufficient to mask behavioral and cognitive

symptoms of impending MCI and episodic memory decline (Sperling

et al., 2011).

A growing body of research is therefore focusing on finding behav-

ioral markers that can identify individuals who are at this preclinical

stage. During recent years, olfactory deficits have received increased

attention as potentially being such early markers of preclinical AD

(Albers et al., 2006; Peters et al., 2003; Tabert et al., 2006). With re-

gards to the distribution of pathology in the early stages of Braak &

Braak, it seems reasonable that olfactory deficits may manifest very

early in the progression of dementia.

38

Olfactory deficits in AD

Although a declining sense of smell is a common feature of older age

and olfactory impairment prevalent even amongst cognitively healthy

older adults, previous studies have shown that olfactory function is

especially affected in patients with AD dementia. A systematic review

from 2012 (Sun, Raji, Maceachern, & Burke, 2012) identified 30 stud-

ies that had examined olfactory identification ability cross-sectionally

in AD patients. All of the included studies found that participants with

AD had statistically worse olfactory identification scores compared to

participants without a diagnosis of AD.

Perceptual versus cognitive olfactory deficits in AD

Not all domains of olfactory function may be equally impaired in in-

dividuals with AD. Most studies that have found correlations with

olfactory performance and preclinical stages of AD were based on

measurements of odor identification. In contrast, studies that have

used tests of peripheral olfactory ability, such as olfactory detection

sensitivity, have found mixed results regarding the role of olfaction in

AD and its preclinical stages. While some findings suggest that odor

threshold deficits may precede a dementia diagnosis and may be asso-

ciated with AD progression (Bacon et al., 1998; Djordjevic, Jones-

Gotman, De Sousa, & Chertkow, 2008; Mesholam et al., 1998;

Morgan, Nordin, & Murphy, 1995; Murphy, Gilmore, Seery, Salmon,

& Lasker, 1990; Nordin, Almkvist, Berglund, & Wahlund, 1997),

other studies suggest that preclinical AD-pathology is mainly related

to central rather than peripheral olfactory dysfunctions (Koss,

39

Weiffenbach, Haxby, & Friedland, 1988; Larsson et al., 1999; Serby,

Larson, & Kalkstein, 1991).

Likewise, findings from a recent meta-analysis of 53 studies sug-

gest that AD is primarily associated with olfactory loss in more cogni-

tively demanding olfactory tasks than in sensory olfactory tests

(Rahayel, Frasnelli, & Joubert, 2012). While all the studies found pro-

nounced olfactory deficits in AD patients when compared to controls

and the meta-analysis showed very large effect sizes across all olfacto-

ry domains, deficits in olfactory identification and recognition were

especially pronounced. In contrast, effect sizes for olfactory sensitivity

and discrimination were significantly smaller. The meta-analysis au-

thors proposed that the strong deficits in odor identification might be

attributable to AD affecting both sensory and cognitive aspects of ol-

factory function (Rayael et al., 2012).

Olfactory deficits predict conversion from MCI to AD

Given the susceptibility of olfactory brain areas to early AD-

pathology it is not difficult to assume that one of the earliest symp-

toms of preclinical AD might in fact present itself in the form of olfac-

tory deficits. If this assumption is correct, the prevalence of olfactory

impairments in AD patients may not only be relatively high when ex-

amined cross-sectionally, but may even predict later progression to

dementia in adults who have not yet received a dementia diagnosis.

Interestingly, findings from longitudinal prospective research studies

indicate that olfactory impairments may in fact predict progression to

AD in patient groups with mild cognitive impairment (MCI; see Table

1).

40

MCI is a syndrome characterized by cognitive decline that is greater

than expected based on a person’s age and education level. These

cognitive deficits are, however, not so severe as to notably interfere

with daily life (Gauthier et al., 2006). Incidence rates of MCI, estimat-

ed by prospective population-based studies, range from 12% to 18%

in persons over the age of 60 years (Petersen et al., 2009). While some

people with MCI remain stable or even return to their previous levels

of cognitive functioning (Manly et al., 2008), about 8% to 15% may

progress to dementia within one year, and about 50% within five years

(Gauthier et al., 2006). The concept of MCI may therefore be regarded

as a sort of “borderland” between normal cognitive aging, preclinical

stages of AD during which the person is cognitively normal but har-

bors the pathophysiological correlates of AD, and the very early stag-

es of the process that may ultimately lead to AD-dementia (Petersen et

al., 2014).

However, it is noteworthy that not all MCI conditions are caused by

Alzheimer’s pathology. The risk of converting from MCI to AD has

been shown to be higher in individuals with AD-related biomarkers

(Ewers et al., 2012) and with genetic risk factors for AD, such as the

ApoE �4 (Mosconi et al., 2004). The likelihood of conversion to AD

is also increased in subtypes of MCI that are characterized by deficits

in cognitive domains that likewise characterize early AD. For exam-

ple, the amnesic type of MCI (aMCI), which mainly manifests in defi-

cits in episodic memory function, is associated with a higher conver-

sion rate to AD than the non-amnesic type of MCI, in which episodic

memory remains relatively intact (Petersen et al., 2014).

41

To date a handful of prospective studies have investigated the role of

olfactory impairments for progression from MCI to AD (Table 1).

Notably, all studies but one (Bahar-Fuchs, Moss, Rowe, & Savage,

2011) found significant associations between baseline odor identifica-

tion ability and risk of dementia progression (Conti et al., 2013;

Devanand et al., 2008). Devanand et al. estimated odor identification

testing to have a sensitivity rate of almost 50% for predicting AD-

conversion within three years.

Olfactory deficits in VaD

After AD, vascular dementia (VaD) is the second main cause of de-

mentia. It is typically characterized typically by a stepwise and severe

decline in cognitive, motor and behavioral function (Román et al.,

1993, 2004). Dementia pathology in VaD occurs due to cerebro-

vascular disturbances in the CNS that eventually cause alterations in

the supply of blood to the brain. In comparison to AD, studies investi-

gating olfactory deficits in VaD are relatively sparse. Although olfac-

tory ability also appears to be impaired in individuals with VaD, find-

ings are mixed regarding the magnitudes of these deficits. While one

study found that patients with VaD had a similar degree of olfactory

impairment when compared to patients with a diagnosis of AD (Gray,

Staples, Murren, Dhariwal, & Bentham, 2001), two other studies re-

ported that olfactory deficits were notably more pronounced in AD

patients than in VaD patients and suggested that olfactory function

might even be used to differentiate between the two (Duff, McCaffrey,

& Solomon, 2002; Knupfer & Spiegel, 1986). As a possible explana-

42

tion for these mixed findings, it has been proposed that the extremity

of an olfactory deficit in VaD might depend on the location and extent

of vascular pathology in the brain, which can vary significantly be-

tween individuals (Alves, Petrosyan, & Magalhães, 2014).

Olfactory impairment as a predictor of dementia conversion

In summary, previous research suggests olfactory deficits to be pro-

nounced in patients suffering from the most common dementia diseas-

es. Furthermore, dementia-related neuropathological changes can be

observed early in brain regions of the olfactory system, and odor dys-

function is related to an increased risk of progression to dementia in

older adults with MCI. Associations between olfactory performance

and dementia are found particularly in contexts were assessments of

odor identification are employed, and where dementia of the Alz-

heimer’s type is investigated.

An intriguing question in light of these results is whether deficits in

olfactory identification can possibly indicate preclinical stages of de-

mentia and predict later progression in adults that do not yet exhibit

any non-olfactory cognitive deficits at all. As mentioned earlier, de-

mentia-related pathophysiological changes appear in olfactory brain

regions during the first stages of AD (e.g. Nyberg, Mclntosh, Houle,

Nilsson, & Tulving, 1996; Braak & Braak, 1995). Likewise, olfactory

impairment may not only predict progression in patients with MCI,

but may even be indicative, if monitored over several years, of in-

creased dementia risk in adults who are not exhibiting other cognitive

symptoms. However, no prospective study had at the onset of the

43

work of this thesis investigated the predictive utility of the sense of

smell for dementia conversion in cognitively healthy older adults.

44

Olfaction and risk of mortality

Apart from dementia, olfactory ability is also impaired in other dis-

eases and conditions. To date, marked olfactory deficits have been

found in Parkinson’s disease (e.g. Doty, 2012; Ponsen et al., 2004),

schizophrenia (Moberg et al., 1999) and major depression (Croy et al.,

2014). Recent prospective studies indicate that olfactory deficits may

also be related to an elevated risk of mortality in older adults.

Wilson, Yu, & Bennett, 2011 found that the risk of death within an

average time period of four years decreased by about 6% for each ad-

ditional correct choice on an odor identification task. Recently, Pinto

et al., 2014 reported that individuals with a lacking sense of smell

were more than three times as likely to die within five years compared

to those with intact olfactory ability, even after statistically adjusting

for malnutrition and numerous other important health variables that

could moderate the relationship between olfaction and death. Notably,

relevant demographic, social and cognitive confounders were also

controlled for.

While several mechanisms have been proposed to support the asso-

ciation between olfactory alterations and mortality, the exact causes

remain elusive. As the olfactory system could be particularly suscepti-

ble to the effects of aging and accumulative damage due to environ-

mental hazards (Doty, 2008), olfactory function might be indicative of

the state of the central nervous system (CNS) in general (Pinto et al.,

45

2014). As such, olfaction may be linked to numerous prospective ill-

health outcomes, even independent of dementia. This hypothesis was

recently supported by a study showing that the association between

odor impairment and death was dependent on cognitive impairment at

baseline (Gopinath et al., 2012). At the onset of the work of this the-

sis, only one study had been able to directly investigate if the associa-

tion between olfaction and mortality might in fact be caused by the

development of dementia after baseline and prior to death, or whether

it persisted even after statistically controlling for the effects of a de-

mentia diagnosis (Devanand et al., 2015). Notably, this study found a

persistent association between odor identification and future mortality

in older community-dwelling adults irrespective of a prospective de-

mentia diagnosis. These results suggested that the link between olfac-

tory function and mortality also exists independent of dementia.

In the context of previous findings regarding olfaction and mortali-

ty, some questions need further clarification. First, only four studies

had, at the onset of this thesis, investigated the association between

olfaction and mortality, of which only one study was able to control

for dementia conversion between baseline and death. Thus, there is a

need for confirmation of these findings. In this respect, the Swedish

sample that was used throughout the studies of this thesis would pro-

vide an especially valuable complement given that the results of pre-

vious studies might be influenced by the US health care system, which

differs from that of Sweden. Second, previous studies are based on

participants at a relatively high age, typically investigating partici-

pants at 70 years of age and older. It remained to be determined

whether the association between olfaction and mortality may also be

46

applicable to younger persons (Devanand et al., 2015). Third, previous

studies have followed their participants for an average time span of

about five years from baseline and it is yet unknown whether mortali-

ty might also be predicted by olfactory deficits under longer time in-

tervals.

47

Subjective olfactory impairment

Given that olfactory impairments are present in cognitive decline and

early-stage dementia and may even be linked to an increased risk of

mortality, a key question is whether such impairments can be self-

assessed. Compared with many other mammalian species, humans

rely far less on the use of olfactory sensations in everyday life. As a

consequence, we may pay less direct attention to our olfactory ability

than to other sensory functions.

As visual input is of primary importance for us to successfully nav-

igate our environment, an impending visual dysfunction is likely to be

noticed immediately. Similarly, we rely heavily on auditory infor-

mation for direct communication with other people. As a conse-

quence, hearing deficits may become rapidly apparent during the so-

cial interactions of our everyday lives. In contrast, the navigation of

modern societies is largely possible without even registering olfactory

cues. In addition, the role of the olfactory sense in social interaction is

of a comparatively subtle and indirect nature. An ongoing decline in

olfactory ability may therefore be more likely to go unnoticed than

changes in other sensory modalities.

48

Sensitivity and specificity of subjective olfactory impairment A small handful of studies have investigated self-assessed olfactory

alterations amongst older adults and they have typically found low

correlations between self-assessments and actual performance on

standardized tests of olfactory function. Individuals with poor olfacto-

ry ability are often unaware of their deficit (Murphy et al., 2002;

Nordin, Monsch, & Murphy, 1995; Shu et al., 2009; Wehling et al.,

2011). In patient-groups with MCI, a lack of awareness of olfactory

deficits combined with an actual alteration in olfactory identification

ability, may even be a risk factor for dementia progression (Devanand

et al., 2000). Conceivably, impaired meta-cognitive awareness may be

linked to a steeper cognitive decline trajectory in these individuals.

However, it is noteworthy that while correlations between subjec-

tive and olfactory measures are low on average and previous studies

have indicated poor sensitivity of subjective olfactory deficits amongst

study participants (i.e. the percentage of individuals with a deficit that

are correctly self-identified as having a deficit is small), self-rated

olfactory deficits seem nevertheless to have good specificity (i.e. the

percentage of individuals without a deficit that are correctly self-

identified as not having alteration deficit is also small). In a recent

population-based study, 87% of participants who reported subjective

olfactory complaints were also classified as having an olfactory im-

pairment based on their performance in a standardized test of odor

identification (Seubert et al., 2017). The actual specificity rate of self-

49

assessed olfaction might even be underestimated in this study given

that only one olfactory modality, odor identification, was assessed and

compared to the subjective estimate. While older adults also tend to

exhibit impairments in sensory olfactory domains (e.g. Schiffman et

al., 1976; Stevens & Cain, 1987).

Hence, while many individuals are unaware of an olfactory dys-

function, misclassifications due to false-positives are rare. Individuals

who do in fact report an olfactory alteration are also likely to be af-

fected by one.

Subjective olfactory impairment as a predictor for dementia

As mentioned earlier, a lack of awareness of olfactory dysfunction

may be linked to cognitive deficits (Devanand et al., 2000). In con-

trast, it is possible that individuals with intact cognitive abilities may

be more likely to detect an ongoing olfactory decline. In cases where

the cause for this decline is neurodegenerative, subjective olfactory

alterations may be predictive for later cognitive decline and progres-

sion to dementia. However, no previous prospective study had at the

onset of this thesis investigated the predictive utility of subjective ol-

factory assessments for dementia conversion in cognitively healthy

older adults.

The motivation to evaluate subjective olfactory impairment as a

marker of dementia progression is supported by recent developments

in neuropsychological assessments of memory. Typically, associations

between self-assessed memory performance and actual performance in

standardized memory tests are weak (Balash et al., 2013; Jorm et al.,

1994; O’Connor, Pollitt, Roth, Brook, & Reiss, 1990; Sunderland,

50

Watts, Baddeley, & Harris, 1986). Furthermore, impaired awareness

of memory deficits is common in older people, especially amongst

those with MCI or AD (Sevush & Leve, 1993; Vogel et al., 2004).

Nevertheless, longitudinal studies indicate that subjective memory

complaints might predict later cognitive decline and conversion to

dementia (Jessen, Amariglio, et al., 2014; Jessen, Wolfsgruber, et al.,

2014; Jorm, Christensen, Korten, Jacomb, & Henderson, 2001;

Schmand, Jonker, Hooijer, & Lindeboom, 1996; Schmand, Jonker,

Geerlings, & Lindeboom, 1997; Tobiansky, Blizard, Livingston, &

Mann, 1995; Waldorff, Siersma, Vogel, & Waldemar, 2012).

A recent study based on a large population-based sample from the

Swedish Betula study confirmed these earlier findings and showed

that subjective memory impairment predicted overall dementia inde-

pendent of baseline episodic memory performance within a follow-up

time span of up till 17 years (Rönnlund, Sundström, Adolfsson, &

Nilsson, 2015). Thus, subjective memory complaints might represent

longitudinal markers of conversion to dementia, despite the relatively

low sensitivity of capturing cognitive impairments in cross-sectional

settings. I hypothesize a similar pattern might also exist for subjective

olfactory complaints.

Long-term subjective olfactory decline To date, all previous studies that have examined correlations between

self-reported olfactory function and olfactory performance in stand-

ardized tests have been based on cross-sectional comparisons. In

summary, these studies have typically found weak associations be-

51

tween subjective olfactory ratings and distributions of scores on olfac-

tory tests (Murphy et al., 2002; Shu et al., 2009; Wehling et al., 2011).

As discussed above, the low sensitivity of self-evaluations for olfacto-

ry impairments might, in part, arise from humans paying relatively

little attention to their olfactory abilities. However, it may also stem

partially from the idea that individuals are often unaware of their own

olfactory abilities compared to those of other people.

Presumably, when assessing one’s own olfactory function, a re-

spondent may be more likely to compare his or her present level of

olfactory performance to his or her past performance levels rather than

comparing it to those of other people. From the perspective of identi-

fying the risk of future dementia conversion, an intra-individual

change may actually be of higher diagnostic utility than between indi-

vidual comparisons, especially given the large individual differences

in baseline olfactory sensitivity (Stevens, Cain, & Burke, 1988). Pre-

vious cross-sectional studies should therefore be complemented with

investigations of subjective olfactory assessments in longitudinal set-

tings. It is still unknown whether a subjective experience of an ongo-

ing long-term olfactory decline is in fact accompanied by an ongoing

decline in olfactory performance in standardized tests when measured

in the same individual over multiple test occasions.

52

Aim of the thesis

Disturbances in the sense of smell are a common feature in older age.

Based on a longitudinal approach, the overall aim of this thesis is to

advance the understanding of the role of sense of smell in memory

decline, dementia, and risk of mortality in older age. More specifical-

ly, the aim of the thesis is to address the following questions:

(1) Are olfactory deficits associated with an ongoing and long-term

decline in episodic memory and is this association influenced by vari-

ations in the �4 allele of the ApoE gene?

Over the past four decades, a growing body of research has linked

impairments in the sense of smell to memory dysfunction, as well as

to a variety of neurodegenerative conditions in older age. For exam-

ple, olfactory performance has been found to correlate with and in

some cases even precede impairments in cognitive function (see Table

1 of the introduction section). Notably, the association between cogni-

tive dysfunction and olfaction might be moderated by variations in the

ApoE gene, a genetic risk marker for AD. However, little is so far

known about how long-term changes in episodic memory function

might develop prior to, in parallel with, or after olfactory impairments.

Furthermore, it is unknown how variations in the �4 allele of the Ap-

oE gene might influence the interaction between olfactory impairment

and episodic memory decline. Such information might further our

53

understanding of the relevance of olfactory function as a marker of

dementia risk.

(2) Can olfactory deficits predict dementia conversion?

As noted earlier, olfactory deficits are highly prevalent in patients

with AD (e.g. Rahayel et al., 2012) and olfactory brain areas are early

affected by AD-related neuropathological changes (e.g. Braak et al.,

1991). It is therefore possible that olfactory impairments might predict

conversion to dementia not only in patients who already exhibit symp-

toms of MCI (e.g. Sun et al., 2012), but also in older adults with intact

cognitive baseline function if followed over a time span of several

years.

(3) Can olfactory deficits predict mortality when dementia is ac-

counted for?

The literature on the predictive utility of olfactory deficits for mortali-

ty is still sparse and only one previous study controlled for dementia

conversion when assessing the prospective relationship between olfac-

tion and mortality. Further studies are needed in order to confirm pre-

vious findings and also to investigate if they can be applied to younger

participants, as well as to longer follow-up time periods.

(4) How does self-assessed olfactory function contribute to the predic-

tion of dementia and mortality?

In the context of dementia and mortality prediction, also self-reported

olfactory dysfunction (i.e. subjective olfactory impairment) might be

of predictive value. This seems particularly relevant as self-assessed

54

olfactory function might be sensitive to intra-individual changes in

odor function that might not by captured by cross-sectional compari-

sons with other persons during objectively administered olfactory

tests. However, no previous study has investigated the predictive con-

tribution of subjective olfactory assessments in this context.

(5) Is a subjectively experienced olfactory decline indicative of an

actual ongoing decline in olfactory performance?

Given my interest in subjective olfactory dysfunction and its predic-

tive capacity, it is critical to establish the link between subjective and

objective olfactory loss over time. Previous studies looking at correla-

tions between subjective and olfactory assessments have all, however,

been based on cross-sectional comparisons. It is therefore unknown

whether an ongoing and long-term decline in subjective olfactory as-

sessments is indicative of an actual decline in standardized measures

of olfactory function over multiple test occasions in a longitudinal

setting.

55

Methods

The Betula Prospective Cohort Study The data from Studies I to IV in this thesis are derived from the Betula

study. Betula is an ongoing large scale prospective study following

community-dwelling adults in Umeå, a city of about 110,000 inhabit-

ants located in northern Sweden (Nilsson et al., 2004; Nilsson et al.,

1997). Participants are randomly sampled from the city's population

registry and contacted by mail. The Betula study has a narrow age-

cohort design in which only participants who fit into previously de-

fined age-groups (such as 25, 35, 40, 45, 50, 55, 60 etc.) are contacted

from the population. This design feature makes it possible to hold

chronological age constant in each cohort. Inclusion and exclusion

criteria for participation in the study are: no ongoing severe visual or

auditory handicap and no dementia diagnosis at inclusion, as well as

having Swedish as a first language. Considering that many partici-

pants are older, the study has so far had a relatively small attrition rate

of about 25% across each five-year testing interval. About 50% of the

drop-outs died (Nilsson et al., 2004).

Ethics approval

In the Betula study, written consent is obtained from all participants in

accordance with the Declaration of Helsinki (BMJ 1991; 302: 1194).

The Regional Ethical Vetting Board at Umeå University approved the

56

Betula study (approval no. 870303, 97–173, 221/97,97–173, 03–484,

01–008, 169/02, 02–164, 03–484, 05-082M, and 08-132M). This ap-

proval covers the four studies included in this thesis.

Procedure

Since the launch of the Betula study between 1988 and 1990, exten-

sive psychological testing and health assessments, largely following

the same testing procedures, are conducted every five years (for an

overview of the study design see Table 2). To date, six test waves

covering 25 years of data collection have been accomplished (T1;

1988-1990, T2; 1993-1995, T3; 1998-2000, T4; 2003-2005, T5; 2008-

2010, and T6; 2013-2015). Collection and entry of data in test wave

six (T6; 2013-2015) had, however, not been finalized when the studies

in this thesis were conducted.

Within each test wave, participants are tested individually in two

separate sessions, one week apart. The first testing session consists of

extensive health examinations performed by a trained research nurse,

including blood sampling for genetic analyses, and the filling out of

questionnaires related to health, social, and demographic variables.

These questionnaires also include a subjective assessment of olfactory

and other sensory functions. A few cognitive tests drawing on general

cognitive ability (e.g. the Mini Mental State Examination (MMSE))

are also performed during the first testing session. All second testing

sessions are performed by trained research assistants and comprise a

large battery of psychological tests regarding function in a variety of

cognitive domains, such as different aspects of memory. The second

testing session also includes assessments of olfactory performance.

57

Study sample

Tests of episodic memory were included into the psychological as-

sessment exams from the start of the Betula data collection. During

the third wave of data collection an olfactory identification test was

introduced for the first time in cohorts Sample 1 and Sample 3. The

data for Studies I to IV is therefore based on multiple collection occa-

sions in these cohorts from test wave three (T3; 2003-2005) to test

wave five (T5; 2008-2010).

Table 2. The Betula study design.

Time (years) Samples (age range)

T1 (1988-1990) S1 (35-80)

T2 (1993-1995) S1 (40-85)

S2 (40-85)

S3 (35-80)

S4 (35-95)

T3 (1998-2000) S1 (45-90)

S2 (45-90)

S3 (40-85)

T4 (2003-2005) S1 (50-95)

S3 (45-90)

S5 (35-95)

T5 (2008-2010) S1 (55-100)

S3 (50-95)

S6 (35-95)

Assessments of olfactory function

The Scandinavian Odor Identification Test (SOIT)

A modified version of the Scandinavian Odor-Identification Test

(SOIT) was introduced to the test battery of the Betula study in order

to measure the participant’s olfactory identification performance. The

SOIT has previously been established as a valid and reliable test of

58

olfactory identification and consists of odors that are commonly con-

sidered as familiar to people living in Scandinavian countries such as

Sweden (Nordin et al., 1998). The odorous stimuli used in the SOIT

are considered to represent a wide range of qualities such as sweet,

spicy, fruity, and floral (Larsson et al., 2004) and are perceptually fair-

ly strong in intensity (Nordin et al., 1998). All stimuli are natural

etheric oils (Stockholm Ether and Essence Manufactory, Stockholm,

Sweden) with the exception of tar, which is a synthetic product

(Bende & Nordin, 1997). The liquid odorants are injected into tam-

pons filled to saturation and placed in an opaque, 80 ml glass jar. Sub-

jects are provided with a written list of four response alternatives for

each odor (Bende & Nordin, 1997) from which they are instructed to

choose the response that best matches the odor. The stimuli are pre-

sented one to two centimeters under the participant’s nose for as long

as required to accomplish the task. There is a 30 second interval be-

tween stimuli in order to limit effects of adaptation. The order of the

stimuli is randomized between subjects.

The modified version of the SOIT, which is used in the Betula

study, consists of 13 odorous stimuli and performance in the test can

range from 0 to 13. In the modified version, the response alternatives

are perceptually more similar to the corresponding test odorant than in

the original version. This modification makes the test more difficult

than the original version, so that ceiling effects can be avoided (Lars-

son et al., 2004). Table 3 gives a summary of the target odors and dis-

tractors in the version of the SOIT that is used in the Betula study.

59

Table 3. The modified version of the SOIT used in the Betula study (Nordin et al.,

1998). Targets (bolded text) and distractors. The terms are translated from the origi-

nal Swedish version to English for demonstrative purposes.

Peppermint Almond (bitter) Acetone Ammonia

Rose Cherry Violet Vanilla

Anise Vinegar Honey Ginger

Apple Lily Violet Vanilla

Oregano Clove Almond (bitter) Cinnamon

Cinnamon Clove Ginger Leather

Pine Needle Gasoline Juniper berry Turpentine

Lilac Rose Lily Jasmine

Orange Tar Lemon Honey

Cherry Strawberry Orange Lemon

Vanilla Coconut Strawberry Jasmine

Eucalyptus Pine needle Camphor Oregano

Gasoline Tar Leather Turpentine

Subjectively assessed olfactory function

Subjective olfactory ability is measured during the first health assess-

ment, one week before performance in olfactory identification is test-

ed. The participants are asked the question, “How is your ability to

perceive weak odors?.” The response alternatives are “no ability,”

“worse than normal,” “normal,” and “better than normal.”

Based on the purposes of Studies II, III, and IV, and because the

self-assessed olfactory variable was not suitable for linear statistical

models, self-reported odor function was collapsed into a dichotomous

variable throughout all statistical analyses in these studies. The re-

sponse categories “no ability to perceive weak odors” and “worse than

normal” were collapsed into subjective reports of olfactory impair-

60

ment and considered subjective olfactory complaints, while the re-

sponse categories “normal” and “better than normal” were considered

as subjectively normal olfactory function and represented no subjec-

tive olfactory complaints.

The olfactory threshold test

During the fifth wave of the Betula study (T5; 2008–2010) a test of

olfactory detection sensitivity was introduced for the first time. In the

Betula, olfactory sensitivity is determined by presenting participants

with pairs of odorous pen-like sticks (Sniffin’ Sticks; Hummel et al.,

1997) with decreasing concentration levels of n-butanol (in total 16

concentrations, where 1 represents the strongest concentration and 16

represents the weakest).

The procedure begins at an intermediate concentration level (8).

Each stick is presented approximately two centimeters under the par-

ticipant’s nostrils every 15–20 seconds. One odorous and one odorless

stick are presented at each trial. The participant’s task is to report

which of the two sticks contained the odor. For those participants who

are able to correctly identify the stick that contained the odor four

times in a row at the intermediate level, the testing procedure contin-

ues at the lowest concentration level (16). When incorrect responses

are given at this level, the concentration of the next odor sample is

increased by one concentration level without any feedback until the

participants give correct responses at the same concentration level four

times in a row.

For participants who are unable to discriminate the odorous stick

from the blank at the intermediate concentration-level, the concentra-

tions of the odor are increased by one level until correct responses are

61

given four times in a row. Participants’ scores are based on the lowest

concentration level they can detect. Thus, an odor threshold score of

16 reflects high olfactory detection sensitivity whereas an odor

threshold score of 1 reflects low sensitivity.

Cognitive assessments

Episodic memory decline

In Study I, an episodic memory measure consisting of a composite of

the following five tasks was used to determine long-term episodic

memory decline: immediate free recall of 16 visually and orally pre-

sented short sentences, delayed cued recall of nouns from the previ-

ously presented sentences, immediate free recall of 16 enacted sen-

tences, delayed cued recall of nouns from the enacted sentences, and

immediate free recall of a list of 12 orally presented nouns. Two dif-

ferent sets of sentences and nouns are used alternately in the first four

measures, and four sets are used alternately in the fifth measure. The

episodic memory battery is administered during the second testing

session of each data collection occasion.

Composite memory scores can range from 0 (minimum) to 76

(maximum). Based on the usage of this composite measure in Study I,

participants were classified as having episodic memory decline if their

rates of change fell below one standard deviation of a model predic-

tion that accounted for their age and individual baseline in episodic

memory performance, and corrected for dropouts. All participants

who scored higher than the cutoff value for longitudinal memory de-

cline were included in the control group. The sample was thus divided

62

into two groups, one with maintaining and one with declining episodic

memory abilities.

Mini Mental State Examination (MMSE)

The Mini Mental State Examination (MMSE) was included in the

Betula study-battery as a general measure of cognitive function

(Folstein, Folstein, & McHugh, 1975). The MMSE is a brief 30-point

questionnaire that assesses general cognitive function and skills in,

amongst others, arithmetic and episodic memory, orientation, and lan-

guage. The test was used throughout Studies II to IV as a covariate in

the statistical regression models in order to control for general cogni-

tive ability.

Synonyms Reasoning Blocks (SRB)

Synonyms Reasoning Blocks (SRB; Dureman, Eriksson, Kebbon, &

Österberg, 1971) is a vocabulary test in which participants chose word

synonyms from four written alternatives in a forced-choice format

similar to the odor identification test. The test was used throughout

Studies I to IV as a covariate in the statistical regression models in

order to control for individual differences in non-olfactory semantic

memory.

Health assessments

Dementia diagnosis

In the Betula study, dementia diagnoses are established for

participants at each test wave through comprehensive reviews of

medical records supplemented with observations made at the health

and neuropsychological assessments, ultimately indicating either

63

normal performance or a decline in cognitive ability and global

functioning. Participants fulfilling one or several of the following

criteria are referred to a specialist in neuropsychiatry: (a) suspected

dementia signs observed by the staff conducting the Betula testing; (b)

MMSE performance below 24; (c) a decline of three or more points on

the MMSE from the previous testing occasion. The evaluation of

dementia status has so far been coordinated by the same senior

research geropsychiatrist throughout the study period and diagnoses

are based on the DSM IV criteria (American Psychiatric Association,

1991).

Determination of mortality

Mortality status and death dates are obtained yearly in conjunction

with Christmas cards being sent out to all Betula study participants

and also every fifth year when scheduling follow-up visits. Mortality

status and date of death are also systematically addressed in

conjunction with the dementia evaluation process at each test wave

and in that context, confirmed via the population registry. For those

participants in Study III who had died, the follow-up time was

measured from baseline (test wave three) to the date of death. For the

remaining participants, the follow-up time was defined from baseline

to the date of the last available follow-up assessment.

Health interviews Medical histories of the participants are examined based on face-to-

face interviews with a research nurse. A history of, for example, heart

disease, stroke, high blood pressure, and diabetes is determined at

inclusion by asking the participants if they had been diagnosed with,

64

sought medical help for, or been hospitalized for any of the above

conditions during their lifetime. At each follow-up, the same questions

are asked targeting the same event, only within the past five years.

Information about present depressive symptoms is obtained from

written answers on the Center for Epidemiologic Studies Depression

Scale (CES-D; Radloff, 1977).

Apolipoprotein E genotyping Blood samples are being collected from the participants during their

first health examination after inclusion to the Betula study. These

samples are amongst others used in order to determine if the partici-

pants are carriers of the ɛ4 allele of ApoE. For the purpose of the sta-

tistical analyses in Studies I to IV of this thesis, the ApoE variable was

dichotomized into one category of carriers of at least one copy of the

ɛ4 allele and one category of non-carriers of the ɛ4 allele.

65

Overview of empirical studies

Study I Olofsson, J.K., Josefsson, M., Ekström, I., Wilson, D., Nyberg, L.,

Nordin, S., Nordin Adolfsson, A., Adolfsson, R., Nilsson, L.G., &

Larsson, M. (2016). Long-term episodic memory decline is associated

with olfactory deficits only in carriers of ApoE-ε4. Neuropsychologia,

85, 1–9.

Aim

The main aims of Study I were to investigate whether odor identifica-

tion performance was associated with an ongoing and long-term epi-

sodic memory decline in older adults and to evaluate whether this as-

sociation was dependent on the �4 allele of the ApoE, a genetic risk

factor for dementia of the Alzheimer’s type.

Method

Participants

The study comprised 1087 participants (593 females and 494 males),

with a mean of 68.5 years of age at baseline. Participants were includ-

ed if they fulfilled the following inclusion criteria: (1) obtaining a

summated episodic memory score and memory classification in a pre-

vious study based on the Betula sample (Josefsson, Luna, Pudas,

Nilsson, & Nyberg, 2012); (2) having completed episodic memory as

66

well as olfactory identification assessments at the last wave of data

collection in which they participated (third, fourth, or fifth wave), and

(3) having participated in the ApoE-genotyping procedure.

Measurements

Odor identification scores in the SOIT from the last test wave in

which the participant participated were used. Long-term episodic

memory decline was determined based on an episodic memory com-

posite score of five tasks (Josefsson et al., 2012), assessed at multiple

test waves over a time span of at least ten and up to 20 years.

Depending on the change in episodic memory ability over the test

waves leading up to the olfactory assessment, the sample was divided

into two groups, one with stable episodic memory trajectories (“non-

decliners”) and one with episodic memory decline (“decliners”). Par-

ticipants were classified depending on whether or not their episodic

memory score declined more rapidly than the age-typical norm and

decline was operationalized as being steeper than one standard devia-

tion for the participant’s age cohort, which was based on a previously

established protocol (Josefsson et al., 2012). Thus, participants were

classified as “decliners” if their rates of change fell below one stand-

ard deviation of a model prediction that accounted for their age and

individual baseline in episodic memory.

This model prediction was also corrected for the effects of drop-

out. In contrast, participants were classified as “non-decliners” if their

rates of change were within one standard deviation of the same model

prediction.

67

Statistical analyses

A univariate analysis of variance (ANOVA) was conducted with pres-

ence of ɛ4 and episodic memory group as fixed factors and odor iden-

tification at the last test wave of participation as the dependent varia-

ble. In follow-up analyses, hierarchical regression models were used

to adjust for effects of possible confounders on the main interaction

effect.

Results

A total of 110 participants were classified as “decliners” and 977 as

“non-decliners”. Odor identification scores were lower in participants

with episodic memory decline as well as in participants with ɛ4. Of

particular interest was an interaction effect between these two varia-

bles (see Figure 1). This interaction effect was driven by poor odor

identification ability, especially in individuals who had both episodic

memory decline and ɛ4, relative to all other groups. In individuals

without ɛ4, odor identification was in fact unaffected by memory de-

cline status, with decliners and non-decliners performing equally well.

Similarly, in individuals with stable memory function, odor identifica-

tion was unaffected by the presence of the ɛ4 allele. Thus, olfaction

was only impaired in participants with both ɛ4 and an ongoing episod-

ic memory decline.

Notably, the interaction effect of the ɛ4 allele by memory decline

status on odor identification remained significant when adjusting for

age, gender and years of education as covariates, and was not caused

by a subset of 63 participants with suspected or confirmed dementia in

the sample.

68

Figure 1. Average episodic memory slopes across 20 years prior to odor

identification assessment. For visualization purposes, participants who par-

ticipated in the study for 10 to 15 years had their trajectories of episodic

memory change extrapolated for the remaining test waves. Note that assess-

ments of olfactory function, the vocabulary control task, and dementia eval-

uation were conducted at the end of the interval.

Follow-up analyses investigated the possibility that ɛ4-carrying in-

dividuals with episodic memory impairment simply had more pro-

nounced episodic memory deficits than non-carriers, which would

suggest that odor identification might only be a secondary feature of

an episodic memory impairment. However, memory in the declining

group was equally poor in carriers and non-carriers. Similarly, the

possibility was considered that the odor identification effect might be

secondary to an overall impairment in semantic memory. However,

69

although vocabulary was poorer overall in individuals with episodic

memory decline compared to non-decliners, vocabulary was not af-

fected by ɛ4 either as a main effect, or in interaction with episodic

memory decline (see Figure 2). Thus, the relationship between declin-

ing episodic memory and impaired odor identification in ɛ4 carriers

was specific to the olfactory nature of the SOIT-task and not driven by

a general decline in word knowledge.

Figure 2. Mean (±1SE) values of odor identification (left panel) and vocabu-

lary (right panel) as a function of ɛ4 and memory decline status. ** p<0.01;

*** p<0.001

70

Conclusion

The results of Study I suggest that olfactory impairments are accom-

panied by an episodic memory decline that has been ongoing for at

least over a decade prior to the olfactory assessment. Furthermore, the

results suggest that long-term episodic memory decline may only be

accompanied by an olfactory impairment in carriers of the �4 allele of

the ApoE, a genetic risk factor for dementia of the Alzheimer’s type.

The findings from Study I further suggest that the association be-

tween odor identification with memory decline and the �4 allele are

not secondary to an overall impairment in word knowledge, nor sec-

ondary to more pronounced episodic memory decline in ɛ4 carriers,

nor driven by individuals with diagnosed dementia. Long-term olfac-

tory and episodic memory assessments may provide complementary

information on dementia-related pathology prior to clinical diagnosis

of AD, but information about the presence of the ɛ4 allele should also

be evaluated.

As information from only one olfactory measurement occasion

could be considered in this study, and prior olfactory performance

levels were thus unknown, further longitudinal studies are needed in

order to determine when olfactory and cognitive changes develop in

relation to each other.

71

Study II Stanciu, I., Larsson, M., Nordin, S., Adolfsson, R., Nilsson, L. G., &

Olofsson, J. K. (2014). Olfactory impairment and subjective olfactory

complaints independently predict conversion to dementia: a longitudi-

nal, population-based study. Journal of the International Neuropsy-

chological Society, 20(2), 209–217.

Aim

The main aims of Study II were to investigate whether conversion to

common forms of dementia can be predicted from performance in

odor identification over a time span of ten years in older adults with

intact baseline cognition, and to determine the predictive utility of

subjective olfactory impairment in this context.

Method

Participants

The study comprised 1529 participants (851 females and 678 males),

with a mean of 61.2 years of age at baseline. Inclusion and exclusion

criteria were not having a dementia diagnosis as well as a MMSE

score of at least 25 points at baseline.

Measurements

Olfactory identification was tested with the modified version of the

SOIT during test wave three (T3). Subjective olfactory impairment

was also measured at this occasion, one week prior to the odor identi-

72

fication assessment. Dementia diagnoses were determined up to test

wave five (T5), ten years later.


The main analysis was a logistic regression model with olfactory iden-

tification performance and subjective olfactory impairment (yes or no)

as the independent variables and dementia conversion (yes or no) as

the outcome variable. The main model controlled for the possibly con-

founding effects of age, sex, years of education and overall cognitive

ability measured by the MMSE. Follow-up logistic regression models

also controlled for the effects of semantic memory (i.e. vocabulary) in

the SRB as well as for depressive symptoms assessed with the CES-D.

The potentially confounding effect of attrition was assessed by ex-

amining characteristics of the participants that had left the study be-

fore their dementia status could be determined by group comparisons

with the remaining participants. Follow-up analyses investigated de-

terminants of self-reported olfactory impairment cross-sectionally at

test wave five, as the olfactory threshold test was introduced at this

point for the first time.

Results

Over the ten-year follow-up time period, 159 participants received a

dementia diagnosis. Most of the dementia diagnoses (93%) were clas-

sified as AD (59.7%) or VaD (33.3%). Lower olfactory identification

ability was associated with an increased risk of dementia conversion.

Similarly, reporting a subjective olfactory impairment was linked to

higher risk of dementia conversion and this association was additive to

the relationship between odor identification and dementia. Notably,

73

the associations between olfaction and dementia remained statistically

significant after controlling for the effects of demographic and cogni-

tive variables (see Table 4).

Follow-up analyses showed that the association between the odor

identification test and dementia was not driven by baseline vocabulary

performance or depressive symptoms. Similar results were obtained

for subjective olfactory assessments. Follow-up analyses further sug-

gested that the association between olfactory identification and de-

mentia might have been driven by patients with AD dementia. Predic-

tion specifically of AD from odor identification scores was significant,

while prediction of VaD was not. In contrast, subjective olfactory im-

pairment was predictive of both AD and VaD.

Further follow-up logistic regression analyses based exclusively on

cross-sectional data from test wave five investigated which character-

istics were associated with self-reporting an olfactory deficit as com-

pared to reporting a normal sense of smell. These analyses were based

on a subsample of 976 cognitively healthy participants who had com-

pleted assessments in both the odor identification as well as the odor

threshold tests at test wave five (T5). The results showed that the like-

lihood of self-reporting an olfactory deficit was significantly increased

by higher age, male sex, lower olfactory identification scores and low-

er olfactory sensitivity threshold scores. The effects of odor identifica-

tion and odor thresholds were additive, indicating that older adults

consider both sensory and cognitive aspects of olfactory ability when

rating their sense of smell.

74

Conclusion

The results from Study II suggest that olfactory ability might predict

conversion to Alzheimer’s disease dementia in adults with intact base-

line function in the MMSE, a test that is sensitive to cognitive symp-

toms of MCI and dementia (Almkvist, 1996; Ford, Haley, Thrower,

West, & Harrell, 1996; Smith et al., 1996). The results strengthen ear-

lier claims about the role of olfaction as an early marker of dementia-

related pathology, prior to clinical diagnosis of dementia. Notably, the

study is the first to establish a direct link between olfactory impair-

ment and prospective dementia diagnosis in cognitively healthy older

adults. Apart from odor identification, also subjective olfactory as-

sessments might be of predictive value in the context of dementia pre-

diction, in addition to performance in odor identification.

CI = confidence interval for odds ratio (OR).

Table 4. Logistic regression analysis predicting conversion to dementia Predictor OR 95 % CI P

Age (+1 year) 1.16 [1.13, 1.19] < .001 Sex (female) 1.23 [0.82, 1.85] .315 Years of education (+1 year) 1.00 [0.95, 1.07] .892 MMSE (+1 point) 0.84 [0.73, 0.96] .013 Odor identification (+1 point) 0.89 [0.81, 0.98] .019 Subjective olfactory complaint (yes) 2.17 [1.40, 3.37] .001

75

Study III Ekström, I., Sjölund, S., Nordin, S., Nordin Adolfsson, A., Adolfs-

son, R., Nilsson, L. G., Larsson, M., & Olofsson, J. K. (2017). Smell

loss predicts mortality risk regardless of dementia conversion. Journal

of the American Geriatrics Society, 65(6), 1238–1243. (Submitted).

Aim

The main aims of Study III were to investigate whether performance

in odor identification is associated with an elevated risk of mortality in

middle-aged and older adults and to determine if this association was

independent from dementia conversion prior to death.

Method

Participants

The study comprised 1774 participants (54.7% females) with a mean

of 63.5 years of age at baseline. Inclusion and exclusion criteria were

not having a dementia diagnosis at baseline.

Measurements


SOIT during test wave three (T3). Subjective olfactory impairment

was also measured at this occasion, one week prior to the odor identi-

76

fication assessment. Dementia diagnoses were determined in a similar

manner as in Study II. Mortality data and dates of death were verified

with the Swedish population registry.


The association between the odor identification score and mortality

was analyzed using Cox proportional hazards models. The initial

model included a term for the SOIT score alone. In subsequent anal-

yses the regression model was adjusted for the following demograph-

ic, health and cognitive variables that were significantly associated

with mortality in the sample: age, sex, years of education, history of

heart disease, stroke, high blood pressure, diabetes mellitus, depres-

sion, cognitive performance in MMSE and SRB, the ApoE ɛ4 allele,

and dementia conversion. Dementia diagnoses were also adjusted for.

Follow-up analyses assessed the association between odor identifi-

cation score and mortality in subsamples including middle-aged par-

ticipants (40-70 years) and older adults (75-90 years). In addition, the

association between subjective olfactory impairment and mortality

was analyzed, and whether it was dependent on the SOIT score.

Results

Of the 1774 participants followed, 411 (23%) died within an average

of nine years after baseline. Performance in odor identification was

associated with an increased mortality risk, even after controlling for

demographic and cognitive variables as well as common morbidity.

Notably, controlling for dementia conversion before death did not

attenuate the association between olfaction and mortality. Subjective

olfactory deficits also emerged as a predictor of mortality risk, but this

77

effect was not overlapping with performance in odor identification.

Follow-up analyses showed that the association between olfaction and

mortality could also be applied to a subgroup of younger participants

aged between 40 and 70 at baseline.

Figure 3. Forest plot with hazards ratios (HRs) and 95% confidence intervals

(CIs) of Cox regression models predicting mortality from the Scandinavian

Odor Identification Test (SOIT). Data were adjusted step-wise for demo-

graphic, health, and cognitive variables; incident dementia; and ApoE �4.

Results are presented separately for SOIT in adults aged 40 to 70 and 75 to

90 and from subjective olfactory ability. All results are adjusted for the full

list of control variables.

Conclusion

The results from Study III confirm and extend findings from previous

studies that identified olfactory deficits as a marker of mortality risk in

community-dwelling older adults (Devanand et al., 2015; Gopinath et

al., 2012; Pinto et al., 2014; Wilson et al., 2011). Olfactory deficits

78

emerged as a predictor of mortality during a time span of approxi-

mately ten years, even independent of dementia conversion. Interest-

ingly, the association between olfactory deficits and mortality was

also present in adults that were younger than those investigated in

previous studies.

Subjective olfactory assessments were also predictive of an in-

creased mortality risk; however, this association was not independent

of performance in odor identification. Subjective olfactory measure-

ments might, however, be implemented in contexts not suited for ol-

factory testing routines.

Study IV Ekström, I., Josefsson, M., Larsson, M., Nordin, S., Nilsson, L.-G., &

Olofsson, J.K. (2018). Subjective olfactory loss corresponds to long-

term odor identification decline in older adults. (Submitted to journal).

Aim

Given that olfactory deficits are associated with a number of poor

health outcomes in older adults, an important question is whether a

decline in olfactory ability can be self-assessed. The main aim of

Study IV was to investigate whether an ongoing and long-term subjec-

tively measured decline in olfactory function was predictive of an on-

going and long-term objectively measured decline in olfactory identi-

fication in older adults.

79

Method

Participants

The study comprised 903 participants (56.8% females) with a mean of

60.5 years of age at baseline. Inclusion and exclusion criteria were not

having a dementia diagnosis at baseline.

Measurements


SOIT at test waves three (T3), four (T4) and five (T5). Subjective ol-

factory impairment was also measured at these occasions, each time

one week prior to the odor identification assessment.


Intercept and rate of change in odor identification performance were

calculated for each participant based on a linear regression model with

odor identification scores at baseline (test wave three; T3) plus a min-

imum of one additional test score from either test wave four (T4), five

(T5) or both as the outcome variable, and time as an exploratory vari-

able. Next, each participant was further categorized into one of the

following three groups based on repeated measures of subjective ol-

factory ability over the studied period: “Normal” long-term subjective

olfaction, “Decline” in long-term subjective olfaction, and “Impair-

ment” in long-term subjective olfaction.

Follow-up analyses investigated the effect of cognitive ability on

long-term associations between subjective odor function groups and

olfactory identification in subgroups of cognitive performance. For

this purpose, a cut-off score of 25 in baseline performance in the

MMSE was used. As a result of having too few cognitively impaired

80

participants (n=27) at baseline in our sample, we were not able not use

a cut-off of 23 or less, which has previously been established as a

cognitive impairment (Mungas, 1991).

Results

Participants who experienced olfactory decline over the study period

had had significantly steeper rates of decline in odor identification

(see Figure 4). The effect remained significant after adjusting for

demographic, cognitive and genetic factors that had previously been

associated with a decline in odor identification, as well as after

adjusting for previous olfactory testing experiences of the participants.

Additionally, participants who experienced an olfactory impairment

already at baseline and throughout the study period also had

significantly lower baseline odor identification scores.

81

Figure 4. Ten–year change in the odor identification score across the subjec-

tive olfactory function groups. Intercept and slopes are adjusted for all co-

variates.

Follow-up results confirmed the same pattern of association between

long-term subjective olfaction and odor identification in the subgroup

of cognitively healthy participants and the whole sample. In contrast,

subjectively experienced olfactory decline was not related to an actual

decline in odor identification in participants with low MMSE-scores at

baseline.

Conclusion

Subjective olfactory decline may be indicative of an intra-individual

long-term decline in olfaction. These changes might not be entirely

captured by cross-sectional comparisons in objective tests of odor

function. For example, older adults may rather compare themselves to

a previous level of olfactory ability than to performance levels of oth-

Years

OD

ID

0 5 10

56

78

NormalDeclineImpairment

82

er people. The results of Study IV of this thesis suggest that although

many older adults may be unaware of an ongoing olfactory deficit,

persons that do in fact report changes in odor function may also suffer

from an actual olfactory decline. However, it is important to note that

the examined sample consisted mainly of cognitively intact older

adults.

Notably, the association between subjective olfactory decline and

changes in odor identification was not significant in a subsample of

participants with low cognitive performance. This finding is consistent

with previous studies indicating that cognitive resources are important

for correctly assessing one’s olfactory function (Devanand et al.,

2000; Nordin et al, 1996).

83

General discussion

In summary, this thesis investigated associations of olfactory ability

with assessments of long-term memory decline, dementia conversion,

and risk of mortality in older age. In addition to the role of objectively

measured olfactory performance in a standardized odor identification

test, the role of olfactory self-assessments was investigated in the con-

text of dementia conversion and mortality. Furthermore, the relation-

ship between self-assessed long-term olfactory decline and objectively

measured long-term olfactory decline was evaluated in a longitudinal

setting. The main findings of the four studies in this thesis can be

summarized as follows:

I. Long-term episodic memory decline in older adults was asso-

ciated with odor identification impairment only in carriers of

the ε4 allele of the ApoE, a genetic risk factor for AD (Study

I).

II. Performance in olfactory identification predicted conversion to

common AD over a time span of ten years in older adults with

intact baseline cognition, independent of a number of known

demographic and cognitive risk factors for dementia (Study

II).

84

III. Performance in olfactory identification predicted mortality

over a time span of up to ten years in middle-aged and older

adults, independently from a number of known demographic,

cognitive, and health risk factors for mortality. Notably, the as-

sociation between olfaction and mortality risk was not depend-

ent on dementia conversion prior to death (Study III).

IV. Self-assessed olfactory ability predicted conversion to AD and

VaD over a time span of ten years in older adults with intact

baseline cognition independent of a number of known demo-

graphic and cognitive risk factors for dementia. The effect was

additive to the predictive utility of olfactory identification

(Study II).

V. Self-assessed olfactory ability predicted risk of mortality. In

contrast to dementia prediction, the association between sub-

jective olfaction and mortality was mainly due to overlap with

olfactory identification performance (Study III).

VI. Self-assessed long-term decline in olfactory function was as-

sociated with an actual long-term decline in objectively meas-

ured olfactory function in older adults (Study IV).

VII. Self-assessed long-term olfactory decline was not related to an

actual long-term odor identification decline in a subset of older

adults with low cognitive performance in the MMSE at base-

line (Study IV).

85

Below, some of the obtained results will be highlighted and discussed

in relation to previous research findings. I will also address methodo-

logical considerations together with some potential problems regard-

ing prospective longitudinal studies in aging research, and propose

ideas for future research.

Olfaction in long-term cognitive decline and dementia

Olfactory deficits in cognitive decline

Olfactory deficits have previously been suggested as a marker for AD.

Previous studies have therefore focused on whether baseline olfactory

performance may predict later cognitive decline or dementia conver-

sion in older adults (see Table 1 of the introduction section). The re-

sults of Study I complement this research by suggesting that in ε4 car-

riers, olfactory deficits may, in fact, also be accompanied by an epi-

sodic memory decline that has been going on for over a decade prior

to the odor assessment. Thus, the findings indicate that long-term

changes in episodic memory ability might concur with the develop-

ment of olfactory deficits in these individuals. Notably, the notion of a

co-occurrence of memory and olfactory decline in older adults may

strengthen earlier claims about using olfactory deficits as a diagnostic

tool for preclinical AD, especially when also considering genetic vari-

ations in the ApoE ɛ4 allele.

86

The odor identification task involves matching odors to word labels.

This feature of the test poses demands also on non-olfactory semantic

memory (Larsson et al., 2004; Olofsson et al., 2013). Study I, there-

fore, considered the possibility that the observed interaction effect of

ɛ4 by memory decline in odor identification might have been driven

by a general deficit in semantic memory, rather than in olfaction. For

this possibility to be true, the same interaction effect would have ap-

peared if the dependent olfactory identification variable had been re-

placed with a vocabulary control test.

In contrast, if the interaction effect would have been caused by me-

diotemporal changes in ɛ4 carriers, as was hypothesized in Study I, the

interaction effect would not be found for a vocabulary control test that

is less dependent on mediotemporal structures (Vargha-Khadem et al.,

1997). Notably, follow-up analyses in Study I showed that the effect

of the ε4 allele on the interaction between olfactory dysfunction and

long-term decline in episodic memory was not related to non-olfactory

verbal components of the memory task, but was instead driven by the

olfactory nature of the odor identification assessment. Thus, the co-

occurrence of episodic memory decline and olfactory deficits in older

age was unlikely to arise due to non-olfactory deficits in vocabulary.

Olfactory deficits predict progression to dementia

Olfactory deficits as a predictor of AD Study II of this thesis was the first to show a prospective association

between olfactory performance and dementia of the Alzheimer’s type

in older adults that were cognitively intact at baseline. The findings

support the notion that olfactory abilities could already be affected in

87

preclinical stages of AD. As such they may be sensitive to early neu-

ropathological changes that do not yet manifest in global measures of

non-olfactory cognitive function, such as the MMSE.

Since the publication of Study II in 2014, and up until today, three

additional longitudinal studies based on different large-scale samples,

have found a similar prospective association between olfactory identi-

fication ability and AD in cognitively intact older adults (for an over-

view see Table 5). Devanand et al. (2015) followed 757 older adults

for a period of two to four years of which 101 (13.3%) transitioned to

AD dementia. Lower baseline odor identification scores measured by

the University of Pennsylvania Smell Identification Test (UPSIT)

were significantly associated with conversion to AD after controlling

for the effects of age, gender, test language, education and cognitive

function.

Follow-up analysis with a subgroup of 259 MCI patients showed

that olfactory impairment similarly predicted transition to AD in this

group after controlling for the same covariates. Adams et al., (2018)

investigated conversion to AD in 2906 subjects with intact baseline

cognition of which 4.1% received a dementia diagnosis at a five-year

follow-up. After having controlled for the effects of age, gender, race,

ethnicity, education, comorbidities, as well as cognition at baseline,

participants with an olfactory impairment had more than twice the

odds of conversion to AD than those without an olfactory impairment.

The sensitivity of olfactory dysfunction for dementia diagnosis within

five years was 47% and the specificity was 79%.

Furthermore, the authors noted that olfactory dysfunction had a

predictive value for dementia conversion equivalent to aging 13 years,

88

in addition to baseline cognitive ability. Yaffe et al., (2017) studied

dysfunction in odor identification as a predictor of transition to AD in

2428 participants from a biracial (black and white) cohort of U.S. old-

er adults. All analyses were stratified by race. Their results showed

that after adjusting for age, gender, education, physical activity and

body mass index, participants in the poor or moderate odor identifica-

tion function tertile had a higher risk of conversion to AD amongst

both white and black participants.

In summary, and comprising Study II of this thesis, four longitudi-

nal prospective studies have thus so far shown that conversion to AD

can be predicted from olfactory identification in older persons without

a non-olfactory cognitive impairment at baseline. A total of 7620 pop-

ulation-based participants from different socioeconomic and ethnic

backgrounds were followed in these studies.

Table 5. Overview of prospective population-based studies using odor identifica-tion testing to predict conversion to AD. Study (Year)

N Country Follow-up (months)

Adj. odds or hazards ratio*

95% CI

Adams (2017)

2906 United States

60 2.13** 1.32–3.43

Devanand (2015)

757 United States

24 & 48 1.04*** 1.04–1.11

Stanciu (2014)

1529 Sweden 120 1.16*** 1.04–1.27

Yaffe (2017)

2428 United States

148 3.34** (white participants)

2.03** (black participants)

2.45–4.54

1.44–2.84

*Adjusted for age, sex and education at baseline as a minimum.

**Olfactory performance was treated as a dichotomous variable.

***Olfactory performance was treated as a continuous variable.

89

Olfactory deficits as a predictor of VaD In contrast to AD, follow-up analyses of Study II showed that olfacto-

ry identification was not significantly associated with an increased

risk of future VaD. Previous research on the role of olfaction in VaD

is sparse and findings regarding the magnitude of olfactory deficits in

this form of dementia are mixed (Duff et al., 2002; Gray et al., 2001;

Knupfer & Spiegel, 1986).

While pathophysiological changes in AD may spread between

Braak areas according to typical and largely predetermined patterns

without large interindividual variations (Braak & Braak, 1995), the

location of vascular neuropathology may differ significantly between

individuals (Alves et al., 2014). Furthermore, the extent of olfactory

identification deficits in VaD may be dependent on whether the loca-

tion of vascular neuropathology overlaps with relevant olfactory brain

areas (Alves et al., 2014). As a consequence, odor identification may

be markedly affected in some patients but not in others.

Study II also showed that while the odor identification score was

not related to an increased risk of VaD, a subjectively rated olfactory

impairment was predictive of VaD conversion. At this point in time,

the reasons behind the discrepancy between the predictive utility of

the subjective and objective olfactory measures for VaD remain a mat-

ter of speculation. It is possible that when assessing their sense of

smell, individuals may tend to reflect upon their olfactory function as

a whole, rather than assess only certain subdomains of olfactory abil-

ity; psychological distinctions that have little overlap with everyday

life. Subjective olfaction may thus represent a more global measure of

olfactory ability than odor identification and as such, also be correlat-

90

ed with pathophysiological changes in a wider array of brain struc-

tures.

Proposed neuropathological mechanisms

Despite their fundamental differences in study design, Studies I and II

both support previous research suggesting that olfactory deficits may

constitute an early clinical marker of AD. While the neurobiological

mechanisms behind this association still remain a matter of specula-

tion, a growing amount of recent research has been able to link per-

formance in olfactory tests to AD-related neuropathology in olfactory

brain areas. Olfactory structures such as the entorhinal cortex, the ol-

factory bulb, and the temporal pole are typically affected by neurofi-

brillary tangles (Attems et al., 2012; Attems et al., 2005; Braak &

Braak, 1995; Kovacs et al., 1999), and alterations in olfactory identifi-

cation have been shown to correlate with these tangles in olfactory

brain areas in early stages of AD (Attems et al., 2012; Franks et al.,

2015; Wilson et al., 2007). While these previous results were based on

postmortem examinations, recent work has been able to link odor

identification also to neurobiological markers of an ongoing AD-

pathology.

It has previously been shown that a lower concentration of cerebro-

spinal fluid (CSF) amyloid-β 1-42 (Aβ42) is reflective of amyloid

plaque deposition in the CNS and that a higher concentration of total

tau (t-tau) in the CSF is indicative of neuroaxonal degeneration that

can be linked to neurofibrillary tangles (Tapiola et al., 2009). A recent

study that examined these biomarkers in relation to olfactory perfor-

mance showed that odor identification scores were negatively corre-

91

lated with concentrations of t-tau in CSF (Reijs et al., 2017). In con-

trast, the same study did not find a significant association between

olfactory ability and cerebrospinal markers of amyloid plaque, which

is in line with previous research showing no relationship between ol-

factory deficits and measurements of in vivo amyloid imaging (Bahar-

Fuchs et al., 2010; Growdon et al., 2015). Interestingly, patients with

non-AD dementias exhibited lower odor identification scores. While

their olfactory deficits were not associated with t-tau in CSF, they may

have been caused by other pathologies (Rejis et al., 2017).

Notably, previous studies showed that the relationship between

psychological tests of olfactory identification and AD-neuropathology

could not be explained by interindividual variations in semantic

memory (Wilson et al., 2007, Reijs et al., 2017). The findings suggest

that pathology-driven interactions between odor identification and

dementia are independent of non-olfactory semantic memory. Hence,

the findings are consistent with the results found in Studies I and II of

this thesis, which suggest that interactions between olfactory identifi-

cation and long-term memory decline or dementia are not driven by

age-related changes in vocabulary. The process of odor identification

might likely involve distinct patterns of activity in brain areas that are

especially sensitive to early neuropathological changes. Interestingly,

neurofibrillary tangles can typically be found in the temporal pole

(Arnold, Hyman, Flory, Damasio, & Van Hoesen, 1991; Arnold,

Hyman, & Van Hoesen, 1994), an area that is important for the trans-

modal processing of object representation and that has also been sug-

gested as one of the neural correlates for odor identification (Olofsson

et al., 2014). The accumulation of neurofibrillary tangles in brain re-

92

gions important for processes of olfactory naming might perhaps ex-

plain why odor identification can predict future dementia conversion,

as was shown in Study II of this thesis and other recent studies (see

Table 5).

Apart from the temporal pole, the central olfactory system projects

to several other regions in the medial temporal lobe. These structures

are crucial for the processing of episodic memory function, which is

markedly impaired in AD. Study I of this thesis showed that olfactory

identification was associated with an ongoing decline in episodic

memory function only in carriers of the ɛ4 allele of the ApoE, which

is the most well established genetic risk-factor for AD (Corder et al.,

1993; Ward et al., 2012).

The exact neural substrates for these ɛ4-related effects are yet un-

known, but it is clear that with increasing age, the ɛ4 allele affects

brain structures that support olfactory and episodic memory functions.

Prior research suggests that many ɛ4 carriers might have increased

atrophy or sub-clinical levels of pathology in the hippocampus and

adjacent regions of the medial temporal lobes (Del Tredici & Braak,

2008; Hashimoto et al., 2001; Lind et al., 2006; Pievani et al., 2009).

Furthermore, in AD, the presence of the ɛ4 allele is specifically linked

to cortical atrophy in these mediotemporal regions (Wolk, Dickerson,

& Alzheimer’s Disease Neuroimaging Initiative, 2010).

As both olfactory and episodic memory functions rely on the integ-

rity of mediotemporal regions (Eichenbaum, 2000; Kjelvik et al.,

2012; Claire Murphy et al., 2003; Segura et al., 2013), these structures

might provide a shared space where atrophy develops to affect both

odor identification and episodic memory functions (Smith et al., 1998;

93

Wilson et al., 2007. Carriers of the ɛ4 allele might be especially sensi-

tive to these effects. Notably, previous research showed a faster pro-

gression from MCI to AD in ɛ4 carriers with high concentrations of

CSF t-tau, while amyloid markers did not interact with the ApoE in

these patients (Blom et al., 2009). Taken together, these findings sug-

gest that interactions between olfactory ability, episodic memory func-

tion, the ApoE and AD-dementia might partly arise due to a shared

relationship with pathological changes linked to neurofibrillary tan-

gles in mediotemporal lobe areas.

Classification of dementia risk based on cross-sectional olfactory

data has previously shown to be facilitated by cortical volume meas-

urements (Devanand et al., 2008). Such approaches could be extended

to data sets that use longitudinal olfactory and memory changes in the

prediction of dementia outcomes. In summary, multi-level longitudi-

nal approaches including neuropsychological, genetic, CSF and corti-

cal data might lead to an improved understanding of the role of olfac-

tory impairments in the prediction of age-related cognitive decline and

dementia.

The influence of the ApoE ɛ4

Differential effects of the ɛ4 allele on memory and olfactory decline As discussed above, carriers of the ɛ4 allele are more likely to have

olfactory deficits as well as poorer performance in episodic memory.

It has previously been proposed that the effect of the ApoE ɛ4 on cog-

nition may be restricted to individuals that are already at various stag-

es of a dementia disease. However, findings regarding the role of de-

94

mentia in the interaction between the ɛ4 allele and cognition in older

age are mixed. While some authors have found effects of the ɛ4 allele

on cognitive performance even after controlling for pre-diagnostic and

concurrent dementia (Olofsson et al., 2009), others have not (Bunce,

Fratiglioni, Small, Winblad, & Bäckman, 2004). It has been proposed

that effects of the ɛ4 allele on cognition could arise due to undetected

dementia pathology (Bunce et al., 2004). In the case of episodic

memory performance, two recent studies suggested that the effect of

the ɛ4 allele on memory assessments disappeared after controlling for

dementia (Larsson et al., 2016; Laukka et al., 2013).

By contrast, previous studies have suggested that effects of the Ap-

oE ɛ4 on olfactory identification performance might also exist inde-

pendent of concurrent and preclinical dementia (Murphy et al., 1998;

Olofsson et al., 2010). It is important to note in this context that the ɛ4

allele has previously been associated with deficits in promoting neurit-

ic outgrouwth in the olfactory epithelium (Hussain, Luong, Pooley, &

Nathan, 2013). As described earlier, the olfactory system depends on

processes of cellular regeneration throughout the human life span. It is

possible that the ApoE may affect neurobiological mechanisms under-

lying olfactory sensations also apart from interactions with dementia-

related neuropathological processes. As a consequence, ɛ4 carriers

that do not convert to dementia may be more likely to develop olfacto-

ry deficits in older age. However, it cannot be ruled out that parts of

the shared variance between the ApoE and olfactory performance

found in earlier studies may still have arisen due to dementia-related

pathological changes.

95

Previous research indicates that AD-related pathology may be more

common in carriers of the ɛ4 allele, even in the absence of a dementia

diagnosis (Jansen et al., 2015). As noted above, AD-related patholo-

gies may accumulate in regions that are engaged in olfactory pro-

cessing already during the earliest stages of the disease (Tabert et al.,

2005; Wesson et al., 2010; Wilson et al., 2007). These pathological

changes may lead to deficits in the performance in olfactory tests. At

the same time, these changes might not yet be sufficiently widespread

or severe enough to affect non-olfactory cognitive functions that are

captured by dementia assessments. Similarly, follow-up analyses of

Study I of this thesis indicated that the effect of the ɛ4 allele on the

interaction between olfactory dysfunction and episodic memory de-

cline was not restricted to persons with established dementia. The ob-

served ɛ4 effects may, however, have been driven by subclinical de-

mentia processes that could not be identified in Study I.

Taken together with previous research, the results from Study I in-

dicate that the effects of the ApoE ɛ4 on brain aging may manifest in

slightly different patterns on episodic memory and olfactory decline.

The ɛ4 allele may be associated with episodic memory performance in

older age predominantly due to interactions with dementia-related

pathology. By contrast, the effects of the ɛ4 allele on olfactory func-

tion may be at least twofold: 1) by interacting with dementia-related

neuropathology, and 2) by increasing the risk for impairment in cellu-

lar neurogeneration in the olfactory system.

The ɛ4 allele in the determination of prospective dementia risk

While the association between olfaction and episodic memory decline

in Study I was restricted to ɛ4 carriers of the ApoE, Study II showed

96

that olfactory impairments could predict progression to dementia in a

large sample of older adults. Thus, the results indicate that olfactory

impairment might signal an impending dementia also for non-carriers

of the ɛ4 allele. While the ɛ4 allele increases the risk for AD, not all

carriers will develop dementia (Smith, 2002). It is also notable that

non-carriers can convert to AD. Thus, while the ɛ4 allele is a well-

established risk factor for AD and may therefore be considered along-

side other variables in research settings and clinical evaluations, at

least in heterozygotes (i.e. carriers of only one ɛ4 allele), it only ex-

plains a relatively small proportion of the total risk for dementia con-

version (Michaelson, 2014).

Furthermore, it is important to note that the sample in Study II was

stratified in order to include only participants without a cognitive im-

pairment at baseline. As a consequence, a comparably large propor-

tion of ɛ4 carriers in the Betula sample might have been excluded

from the study, as they might be more likely to be affected by cogni-

tive dysfunction than non-carriers (Olofsson et al., 2009). While the

results of Study II, combined with those of later studies (see Table 5),

indicate that dementia conversion can be predicted from olfactory

identification in the general population of older adults, future studies

may clarify whether the prospective relationship between olfactory

impairment and dementia might be of a higher magnitude in ɛ4 carri-

ers than in non-carriers. To date, the results of Study I suggest that the

ɛ4 allele may be taken into account when evaluating the risk for

memory decline and dementia based on olfactory ability.

97

A proposed timeline for the development of olfactory and memory

decline in cognitive aging and dementia

An important remaining question regarding olfaction in cognitive ag-

ing and dementia concerns the temporal framework in which olfactory

and cognitive impairments develop in relationship to each other. Pre-

vious studies have typically predicted later cognitive decline from

baseline olfactory impairments. In addition, Study II of this thesis

showed that olfactory deficits preceded a dementia diagnosis within a

follow-up time span of up to ten years. Notably, baseline cognitive

symptoms as measured by the MMSE were absent in the participants

that were evaluated, indicating that the olfactory alterations that were

predictive of AD developed prior to cognitive deficits assessed by this

test. Thus, olfactory deficits may develop prior to pronounced deficits

in overall cognition.

By contrast, the results from Study I showed that, olfactory identifi-

cation scores were associated with a long-term episodic memory de-

cline that had been going on for a period of ten to twenty years prior

to the establishment of the olfactory deficit. These findings suggest

that certain memory impairments might develop in parallel with olfac-

tory deficits or even before. The results from Studies I and II may

seem contradictory at first glance. It is, however, important to note

that the studies differed fundamentally in the way in which cognition

was assessed. Study II excluded all participants according to a cut-off

of baseline MMSE-scores that might have been indicative of a cogni-

tive impairment (Mungas, 1991). Thus, in order for someone to score

below this cut-off point, cognitive dysfunction must already be quite

pronounced. In contrast, Study I employed a longitudinal composite

98

measure of memory decline based on repeated assessments in multiple

episodic memory domains. Decline was operationalized as being

steeper than one standard deviation for the participant’s age cohort.

Although more conservative definitions could have been used (e.g.

two standard deviations), this would have yielded a very small group

of ɛ4 carriers with memory decline. As a consequence, the episodic

memory decline that was measured in Study I likely represented a

much subtler cognitive decline than the more pronounced dysfunc-

tions within the MMSE that were used as a cut-off in Study II.

Previous studies have estimated that the onset of AD-related pa-

thology might be very early, several years prior to the manifestation of

the first clinical symptoms (Jack et al., 2013; Villemagne et al., 2013).

During this long subclinical time span, slow and subtle changes in

olfactory and memory function could develop in parallel given that

they are engaged by the same affected brain structures. Compared to

memory structures, the olfactory system has furthermore been shown

to be especially sensitive to non-dementia related pathology, environ-

mental agents and the aging process. The resultant hazardous effects

could accumulate in the olfactory system causing damage in olfactory

brain areas that are additive to the damage already caused by AD-

pathology. It is also possible that these effects could interact with AD-

pathology in olfactory brain regions. This might render olfactory abili-

ties more sensitive to early preclinical dysfunctions than non-olfactory

memory abilities that are less dependent on olfactory brain areas, and

thus only affected by dementia pathology. As a consequence, pro-

nounced olfactory deficits could manifest before pronounced deficits

in non-olfactory cognitive function in many older persons. One of the

99

biggest challenges for future research is to develop methods that can

differentiate between olfactory impairments that are caused by demen-

tia pathology and those that are related to other etiologies.

As the odor identification test was introduced to the Betula study

ten years after the episodic memory composite, the investigation of

the relationship between olfactory dysfunction and memory decline

was limited to olfactory measurements taken on one test occasion,

during Study I. To further the understanding about the temporal set-

ting in which olfactory and cognitive deficits develop, future studies

should also employ longitudinal measures of olfactory change. Nota-

bly, a recently published study, that was also based on participants

from the Betula sample, investigated longitudinal changes in odor

identification and non-olfactory cognitive abilities across three testing

occasions spanning an interval of up to ten years (Josefsson, Larsson,

Nordin, Adolfsson, & Olofsson, 2017). The results suggested that

odor identification ability could start to decline as early as middle age

and that carriers of the ɛ4 allele, who have the highest risk of develop-

ing dementia, could decline twice as fast. Thus, a decline in olfactory

identification may start earlier than previously hypothesized.

The study further showed that olfactory decline in middle-aged

adults was correlated with a decline in episodic memory in ApoE ɛ4

homozygotes (carriers of two ɛ4 alleles). Thus, the results are con-

sistent with the interpretation of findings from Study I of this thesis.

At least in ɛ4 carriers, odor identification and episodic memory could

jointly be affected in preclinical AD. As such, olfactory identification

and episodic memory could represent complementary assessments of

100

function within the mediotemporal lobes (Josefsson et al., 2017). Fur-

ther studies are needed in order to confirm these early findings.

Olfactory dysfunction as a predictor of mortality Study III of this thesis showed that odor identification scores were

associated with an increased risk for mortality within a follow-up time

span of approximately ten years. Notably, the effect was independent

of dementia conversion after baseline and prior to death, which sup-

ports the results of a previous study (Devanand et al., 2015). The find-

ing suggests that the relationship between olfaction and risk of mortal-

ity might follow a different mechanistic pathway than the ones that

have been proposed regarding the preclinical role of olfactory deficits

in dementia. However, the exact mechanisms that could account for

this relationship remain a matter of speculation. No study has yet fac-

tored in the cause of death of their participants, which would permit

an additional exploration of specific mechanisms.

As described in the introduction section of this thesis, olfactory def-

icits in older age have been linked to numerous detrimental outcomes.

One such outcome is loss of appetite with changes in dietary habits as

a consequence (Thomas Hummel & Nordin, 2005; Temmel et al.,

2002). It could be that alterations in dietary habits related to olfactory

loss lead to an increased risk of death. However, Pinto et al. (2014)

controlled for the possible effects of malnutrition and found that it

could not account for the relationship between olfactory deficits and

mortality. Notably, previous research has not been able to establish a

direct link between olfactory deficits and nutritional factors. Rather

101

than causing malnutrition per se, it has been proposed that olfactory

deficits may be overrepresented in mental and physical conditions,

which in turn elevate the risk for malnutrition (Toussaint et al., 2015).

Thus, malnutrition is unlikely to play a direct role in the link between

olfactory deficits and dementia.

Apart from changes in appetite and dietary habits, olfactory deficits

have been linked to an increased risk of accidents at home, such as gas

poisonings and the consumption of spoiled foods (Bonfils, Faulcon,

Tavernier, Bonfils, & Malinvaud, 2008; Doty & Kamath, 2014;

Santos, Reiter, DiNardo, & Costanzo, 2004). It is possible that such

accidents may lead to an increased risk of death in some cases. How-

ever, it is unlikely that accidents can explain the entire association

between olfactory function and risk of mortality. As mentioned in the

introduction of this thesis, the olfactory system might be particularly

exposed to the effects of aging as it relies on cellular regeneration

(Pagano et al., 2000), a process that has been associated with age-

related impairment (Laws et al., 2002; Striepens et al., 2011; Van der

Flier et al., 2008), (Watabe-Rudolph et al., 2011).

Furthermore, the olfactory system is unique amongst sensory sys-

tems in that it is directly exposed to the environment and may thus be

sensitive to accumulative damage caused by environmental hazards

(Doty, 2008). It has therefore been proposed that rather than causing

death, olfactory function may be indicative of fundamental physiolog-

ical processes of aging in the CNS and their interaction with environ-

mental exposure (Pinto et al., 2014). As such, olfaction may be linked

to numerous prospective ill-health outcomes, even independent of

dementia. Future studies with access to cause of death of the partici-

102

pants are needed to further our understanding of the mechanisms un-

derlying the association between olfactory dysfunction and mortality.

Subjective olfactory dysfunction

Subjective olfaction and dementia risk

Apart from an odor identification test, Studies II, III, and IV also in-

cluded self-assessments of olfactory ability. The studies in this thesis

were the first to employ subjective measurements of odor function in

longitudinal settings. Study II showed that subjective olfactory im-

pairment was associated with an increased risk for dementia conver-

sion. This effect was independent of scores in olfactory identification.

Interestingly, follow-up analyses showed that olfactory self-reports

could predict conversion to VaD, while performance in odor identifi-

cation could not. As discussed above, the neuropathological changes

that are associated with VaD may be less restricted to specific brain

areas than changes in AD-dementia (Alves et al., 2014). Thus, these

changes could affect olfactory brain areas in some VaD patients and

not in others.

When assessing one’s sense of smell, participants may be likely to

consider multiple olfactory domains meaning the subjective olfactory

measure may be represented by function in a wider array of brain are-

as than in the odor identification test. As a consequence, subjective

olfactory deficits may predict conversion to VaD, while odor identifi-

cation may not. It is important to note that the results of cross-

sectional follow-up analyses on determinants of subjective olfactory

assessments in Study II were consistent with this interpretation. Alt-

103

hough the item on the olfactory self-assessment questionnaire referred

specifically to the ‘‘ability to perceive weak odors,’’ odor identifica-

tion emerged as a significant predictor of subjective olfaction, in addi-

tion to olfactory thresholds. The results suggested that, when asked

about their ability to detect weak odors, participants’ responses re-

flected both sensory (detection) and cognitive (identification) olfacto-

ry abilities. Therefore, subjective olfactory assessments may relate to

a wider variety of olfactory domains than single tests of olfactory

function. As such, they may be more likely to predict conversion to

VaD than assessments in odor identification. Further studies are need-

ed in order to confirm this pattern.

Cognitive aspects of subjective olfactory complaints

Previous studies have suggested that awareness of an olfactory deficit

may be linked to cognitive ability (e.g. Devanand et al., 2000; Nordin

et al., 1996). Study IV therefore included follow-up analyses that in-

vestigated the possible effects of cognition on the association between

long-term subjectively and objectively established olfaction. For the

purpose of this analysis, the participants were separated into sub-

groups based on their baseline cognitive performance. The results

suggested that amongst participants with low scores in the MMSE,

subjectively experienced olfactory decline was not significantly relat-

ed to an actual decline in odor identification. By contrast, self-reported

olfactory decline was predictive of changes in olfactory identification

in the subgroup of cognitively intact individuals. While these results

may have arisen due to less statistical power in the analysis of the

subgroup with lower cognitive scores, it may also indicate that self-

104

assessments of olfactory ability may be less reliable in persons with

cognitive deficits. This interpretation is consistent with previous re-

search that has linked awareness of olfactory ability to cognitive re-

sources (Devanand et al., 2000; Wehling et al., 2011). It is likely that a

certain level of memory function and meta-cognitive awareness is

needed in order to accurately assess one’s olfactory performance level.

As a consequence, cognitive functioning levels may be considered

when evaluating the reliability of self-reported olfaction.

Subjective olfaction in prospective settings

In summary, Studies II to IV of this thesis shed new light on the pre-

dictive utility of subjective olfactory impairment in longitudinal pro-

spective settings, despite their unreliability in cross-sectional compari-

sons (Murphy et al., 2002; Nordin et al., 1995; Wehling et al., 2011).

Further research is needed to establish which olfactory assessment is

best suited to predict later conversion to dementia. The present results

suggest that a combination of olfactory test scores and subjective re-

ports would provide the best prediction.

The results of Study II mimor prior findings regarding subjective

memory complaints in the context of dementia prediction. As men-

tioned above, self-assessed memory function typically correlates

weakly with standardized tests of memory in cross-sectional studies

(e.g. Jorm et al., 1994). Nevertheless, a growing body of research has

shown that it is indicative of an impending AD dementia and can ex-

plain additional variance on top of baseline episodic memory assess-

ments in longitudinal prospective settings (e.g. Rönnlund et al., 2015).

According to the NIA-AA criteria for AD stages, the first (i.e. prepro-

105

domal) stage of preclinical AD may not only be characterized by man-

ifestations of AD-biomarkers in CSF, but also by a subtle cognitive

decline (Jessen, Amariglio, et al., 2014). Notably however, this subtle

decline is not severe enough to reach the same level as that of objec-

tive cognitive deficits that are required for a diagnosis of MCI or AD.

Furthermore, the decline may be hard to detect on standardized as-

sessments of cognitive function, partly because of insufficient sensi-

tivity and reliability of these measures and partly because of success-

ful cognitive compensation strategies in some individuals (Jessen,

Amariglio et al., 2014; Erk et al., 2011). However, the fact that stand-

ardized assessments might not be able to detect the subtle cognitive

decline of preprodromal AD does not preclude that individuals may

experience a subjective cognitive decline at this stage (Jessen, Amari-

glio et al., 2014).

Notably, preclinical subjective memory complaints are associated

with biomarkers for AD in CSF as well as with an increased prospec-

tive risk for AD diagnosis (Buckley et al., 2016). It has therefore been

proposed that subjective cognitive decline (SCD) may represent one

of the first symptomatic manifestations of AD (Jessen, Amariglio et

al., 2014).

Interestingly, SCD is overrepresented in �4 carriers of the ApoE

(Laws et al., 2002; Van der Flier et al., 2008; Striepens et al., 2011).

Furthermore, the �4 allele has been found to increase the risk of pro-

spective cognitive decline in individuals with SCD (Dik et al., 2001).

As mentioned earlier, the �4 allele has also been associated with olfac-

tory dysfunction (e.g. Bacon et al., 1998; Gilbert & Murphy,

2004).The results from Study I of this thesis suggest that variations in

106

the ApoE might moderate the relationship between olfactory deficits

and a long-term decline in episodic memory function. In light of these

previous findings, future studies may want to investigate the following

questions in order to elucidate the role of subjective olfactory decline

in the preclinical stages of AD: (1) Do objective and subjective olfac-

tory alterations co-occur with SCD? (2) Do objective and subjective

olfactory deficits overlap with SCD in their prediction of dementia?

(3) Are subjective odor dysfunctions associated with biomarkers of

AD in CSF? (4) Are subjective odor deficits overrepresented among

�4 carriers? (5) Are subjective odor deficits associated with a long-

term cognitive decline?

Subjective and objective long-term olfactory decline

Study IV of this thesis suggests that an ongoing subjective olfactory

decline might be indicative of an actual ongoing olfactory decline in

odor identification. These findings indicate that individuals who report

alterations in odor function might in fact also experience a deteriorat-

ing sense of smell. Previous cross-sectional studies have typically

found low correlations between self-reported olfaction and tests of

olfactory ability (e.g. Wehling et al., 2011). It has therefore been pos-

ited that subjective olfactory assessments are unreliable (Shu et al.,

2009). As a consequence, the results of Study IV may seem paradoxi-

cal in the context of previous research. However, I argue that our find-

ings should not be viewed as contradictory to the results of previous

studies that found weak associations between self-assessed and tested

odor performance, rather as complementary.

107

Similar to self-reported cognition, low correlations between subjective

and objective olfaction in cross-sectional research settings might, in

part, arise from comparisons with other persons, hence not represent-

ing a sufficient method for capturing subjective intra-individual

change. The results of Study IV indicate that while many older partic-

ipants might be unaware of an ongoing olfactory decline, those who

do report a deficit over time may also likely suffer from one. This in-

terpretation is consistent with a previous cross-sectional study sug-

gesting low sensitivity, but nevertheless high specificity, of olfactory

self-reports (Seubert et al., 2017).

Sensory deficits and brain aging, dementia, and mortality Aging is associated with an increased incidence of sensory loss not

only in olfactory function, but also in hearing and visual abilities

(Christ et al., 2014; Dalton et al., 2003). Furthermore, deficits in hear-

ing and vision have also been linked to an increased incidence and

prospective risk of cognitive dysfunction, AD, and all-cause dementia

in older adults (Lindenberger & Baltes, 1994; Uhlmann, Larson, Rees,

Koepsell, & Duckert, 1989; Valentijn et al., 2005). More than 20 years

ago, the common cause theory was introduced (Lindenberger & Bal-

tes, 1994). The theory suggests that common underlying factors such

as generalized aging effects, contribute to the development of both

sensory and cognitive impairment in older age. Consistent with this

theory, a few previous studies do suggest that sensory and cognitive

abilities may share a considerable proportion of age-related variance

(Anstey, Luszcz, & Sanchez, 2001; Lindenberger & Baltes, 1994). It

108

has therefore been proposed that sensory losses may represent early

signs of age-related neurological impairment, neuropathology or neu-

rodegeneration and that this decline manifests prior to cognitive symp-

toms (Anstey et al., 2001).

Despite suggestions regarding a common cause of sensory and cog-

nitive deficits, most previous studies have however mainly focused on

sensory specific relationships between sensory loss and cognitive ag-

ing. While a few cross-sectional investigations have considered dual

impairment in vision and hearing (Lin et al., 2004; Lindenberger &

Baltes, 1994; Tay et al., 2006), as yet, only one longitudinal study has

analyzed the interrelationships between olfactory, visual and hearing

deficits and the prospective risk of cognitive impairment (Fischer et

al., 2016). The objective of the study was to determine if baseline def-

icits in hearing, visual and olfactory function could independently

contribute to the prediction of cognitive impairment within ten years.

The results showed that all sensory modalities had significant effects

on the ten-year risk of cognitive dysfunction. Notably, the effects were

independent of one another and thus cumulative. A joint model

showed that an individual with deficits in all three sensory domains

was 15 times more likely to receive a cognitive impairment than an

individual without any sensory loss.

It is particularly interesting to note that effect sizes of the olfactory

impairment were notably higher than for visual and hearing deficits,

both in an unadjusted model and in the model that was adjusted for

demographic confounders. Olfactory deficits had also the highest

specificity rate for cognitive impairment (85.8%), while hearing defi-

cits had the highest sensitivity rate (72.9%). Combining all three sen-

109

sory modalities leads to the highest specificity, but to a lower sensitiv-

ity rate, than when considering any of the sensory modalities separate-

ly. In summary, these recently published results suggest that deficits in

each sensory modality have a unique effect on the risk of cognitive

impairment. Measures of olfactory, visual, and hearing function may

thus not be interchangeable in the context of cognitive aging, but

might be considered separately.

Hence, the findings suggest that in addition to a common cause, dif-

ferent mechanisms may be involved in the specific associations be-

tween each sensory function and overall cognitive function. Loss in

vision and hearing may lead to cognitive decline as they may cause an

increased cognitive load, sensory deprivation and possibly even social

isolation (Fischer et al., 2016; Lin et al., 2011). In contrast, associa-

tions between olfactory and cognitive impairment may mainly be a

product of shared neuropathological processes such as neurofibrillary

tangles (e.g. Wilson et al., 2007). Olfactory deficits may thus be of a

higher diagnostic value than other sensory deficits for clinical out-

comes in which cognitive dysfunctions are caused by these neuropa-

thological changes, such as AD. In order to shed light on this question,

studies are needed that examine the prospective utility of different

sensory deficits for common dementia forms. Future studies should

ideally also consider associations between domain-specific sensory

impairments and biomarkers of AD in CSF as well as their relation-

ship to genetic risk factors such as the ApoE �4.

110

Sensory deficits and mortality

Regarding mortality, a recent study suggests that olfactory deficits

may be unique among sensory systems in their association with an

increased risk of death (Schubert et al., 2017). This longitudinal popu-

lation-based study was the first to investigate the impact of multiple

sensory measures in hearing, vision and olfaction on mortality risk.

Notably, only olfactory identification deficits were associated with an

increased incidence of death over the 17 years of follow-up. The au-

thors suggest that olfactory deficits may be a marker of more ad-

vanced physiological aging than hearing or visual impairment. In con-

trast to hearing and vision, olfaction may indicate a decline in pro-

cesses of neurogenesis and loss of plasticity in the CNS (Schubert et

al., 2017).

Points of caution

In the following subsections, I will highlight features of the studies in

this thesis that deserve a more thorough or critical discussion. In par-

ticular, possible methodological and statistical limitations will be

mentioned.

Study design

All of the studies in this thesis used a longitudinal approach where

participants were followed over periods of time spanning several years

to several decades. The studies were observational in nature, without

any applications of external influence. Prospective observational ap-

proaches have shown to be particularly useful for evaluating the rela-

111

tionship between risk factors and the development of disease or other

detrimental outcomes over time (Caruana, Roman, Hernández-

Sánchez, & Solli, 2015). In contrast to cross-sectional designs, which

are static by their very nature, longitudinal studies might be especially

useful in research settings of cognitive and sensory aging, dementia,

and risk of mortality, as they provide useful information about the

influence of time on the measured variables (Caruana et al., 2015).

Longitudinal approaches can also consider intra-individual decline

trajectories. Given the large variability in levels of cognitive and sen-

sory function between older adults, intra-individual decline may re-

flect relevant age-related changes more accurately than cross-sectional

comparisons with other persons.

Furthermore, so-called “cohort effects” can be corrected for in lon-

gitudinal studies. Cohort effects are systematic differences between

participants in younger and older age groups that are related to other

factors than age. For example, level of education may differ largely

between generations.

Like all research designs, longitudinal approaches are also accom-

panied with certain challenges. The biggest challenge is the incom-

plete or interrupted follow-up of individuals, with attrition and drop-

out as a consequence. This might be especially problematic in settings

where the risk for attrition is correlated with the exposure and out-

come measures.

Attrition in longitudinal studies

Attrition is a common problem in prospective longitudinal research

studies, especially in those investigating older adults. A systematic

112

review from 2005 summarized all large population-based studies of

older adults that had investigated attrition in their samples and con-

cluded that increasing age, cognitive impairment, ill health and frailty

were consistently associated with a higher risk of dropping out from a

study (Chatfield, Brayne, & Matthews, 2005).

There is a risk that attrition might introduce bias into the findings of

a study, as dropping out might be associated with both the predictor as

well as the outcome variable. In the case of olfactory function,

measures of odor identification are correlated with known risk factors

for attrition, such as increasing age, decreasing cognitive performance

and various outcomes of ill health (Larsson, Finkel, & Pedersen, 2000;

Seubert et al., 2017). Rather than being directly linked to an increased

risk of attrition, it is likely that olfactory impairments might be associ-

ated with attrition through these confounding factors. As dementia is

most likely also related to an increased risk of attrition, studies about

the prospective association between olfaction and dementia might be

sensitive to bias. Study II therefore reported statistical control for the

potential effects of attrition on the main results. This analysis con-

cluded that olfactory impairment was linked to a higher drop-out rate,

but that this association could be explained by demographic and cog-

nitive differences, which were subsequently adjusted for.

In order to further examine the potential biasing effects of attrition

on the results of prospective studies investigating olfaction and de-

mentia, I constructed a Directed Acyclic Graph (DAG) with the aid of

the free software, Dagitty. With a DAG, accurate biasing paths can be

identified, given that the assumptions made about the relationships

between the predictor, outcome, and confounding variables in the

113

model are correct. Interestingly, the DAG showed that there might be

a low risk for bias due to lack of follow-up in studies investigating

associations between olfactory impairment and AD, as olfactory func-

tion is unlikely to be directly related to attrition (Figure 5). Bias might

be introduced through cognitive, demographic, and health variables,

but these effects are not dependent on attrition. The results of bias

analyses based on a DAG are always dependent on the premises on

which it was created. Although unlikely, it cannot be definitively ruled

out that olfactory function is directly linked to attrition. Ideally, future

studies should therefore also include reports of dropouts and attempt

to statistically account for the potential effects of attrition.

Figure 5. A DAG created with and visually adapted from Dagitty and based

on the assumed relationships between odor identification, dementia, attrition,

and confounding variables. Potential biasing paths are colored in grey while

non-biasing paths are colored in black. Note that no biasing path goes

through the variable “attrition” in this model.

114

Olfactory assessments

Further points of caution concern the psychometric properties of ol-

factory measures that were employed in the studies of this thesis, the

odor identification test and the subjective measure of olfactory func-

tion. These points are discussed below.

The odor identification test The odor identification test employed in the Betula study represents a

slightly modified version of the SOIT (Nordin et al., 1998). This mod-

ification has made the odor identification test in the Betula study more

difficult than the original version in order to avoid ceiling effects. Pre-

vious studies based on Betula data could show that the odor identifica-

tion scores are well distributed amongst participants and that they do

not show any ceiling or floor effects for any age group (Larsson et al.,

2004). It is important to note that the findings of Studies II and III of

this thesis have been confirmed in other studies using other standard-

ized olfactory identification measures such as the commonly em-

ployed University of Pennsylvania Smell Identification Test (UPSIT).

The subjective olfactory measure

In the Betula study, subjective olfactory status is measured with the

question “How is your ability to feel weak odors?” The response al-

ternatives are “no ability”, “worse than normal”, “normal” and “better

than normal”. In Studies II to IV, this self-reported olfactory status

was transformed into a dichotomous variable where the answer-

options “no ability to feel weak odors” and “worse than normal” rep-

resented an olfactory dysfunction and “normal” and “better than nor-

mal” were considered as a normal sense of smell.

115

The dichotomization was performed as none of the participants in

Study II had reported “no ability to feel weak odors” at baseline and

only a small percentage of 7.8 % rated their olfactory ability as better

than normal. Based on this response pattern, it was concluded that the

variable was more suited to be used as a dichotomous factor in the

regression models of Study II and onwards.

Although not reported in the studies of this thesis, it should be not-

ed that similar results were obtained for Studies II to IV when subjec-

tive olfactory ability was treated as a factor with three possible out-

comes and hence a distinction was made between a subjectively

“normal” and a “better” sense of smell.

It could be argued that the olfactory self-assessment in the Betula

study is ambiguous as it contains a reference to “normal” olfactory

sensitivity to which the individual contrasts his/her own ability. For

example, “normal” olfaction could either refer to other people’s abili-

ties or the participant’s own, prior olfactory abilities. While the exact

response format is indeed likely to influence responses, I argue that

subjective olfactory impairments must always be assessed with refer-

ence to “normal” olfactory functioning. The future development of

subjective olfactory assessments could include questionnaires map-

ping out everyday situations in which a person is likely to reflect upon

his or her olfactory acuity.

It is important to note at this point that the longitudinal nature of the

study-design in Study IV might introduce a certain risk of bias, as all

the participants had undergone repeated assessments in the standard-

ized odor identification test throughout their years of participation.

One previous study suggests that individuals who are asked to assess

116

their olfactory ability after participation in standardized olfactory tests

are more likely to be aware of an eventual deficit, even if no feedback

is given (Landis, Hummel, Hugentobler, Giger, & Lacroix, 2003). It

is, however, important to mention that the subjective assessment in the

Betula study was performed one week prior to the odor identification

test at each test wave. When the question about subjective olfactory

function was posed to the participants, their last occasion of participa-

tion in a standardized olfactory assessment was thus five years earlier.

Still, it cannot be ruled out that olfactory awareness might have been

higher in the Betula sample than in the general population. However,

the repeated measurements were necessary in order to examine long-

term decline, precluding alternative options for investigating correla-

tions between objective and subjective olfactory change longitudinal-

ly.

Given the prospective relationship of subjective olfactory com-

plaints for dementia and mortality found in Studies II and III as well

as negative consequences of odor deficits for overall quality of life

(Croy et al., 2014), future studies may want to focus on the develop-

ment of methods that can increase awareness of olfactory decline

amongst the older population in general.

Dementia diagnoses

Overlap between AD and VaD dementias

While AD and VaD dementias are considered to be different disease

entities, current diagnostic criteria (Fletcher, 2012; McKhann et al.,

2011; Román et al., 1993, 2004) often fail to discriminate between the

two (Llorens et al., 2017). In the Betula study, the subgroup definition

117

used to distinguish between AD and VaD was based on clinical crite-

ria and neuropyschological testing. Ideally, the diagnosis procedure of

dementia subtypes would have been complemented with biomarkers

(Andreasen & Blennow, 2005; Bibl, Esselmann, & Wiltfang, 2012). It

is, however, important to note that although biomarkers may be useful

in further distinguishing between AD and VaD, the currently available

CSF tests may exhibit relatively low specificity, which still impedes

absolute discrimination between dementia subtypes (Nägga, Gottfries,

Blennow, & Marcusson, 2002; Paraskevas et al., 2009).

For example, a recent study suggests that commonly employed AD

biomarkers, such as t-tau and Aβ42, cannot adequately distinguish

between AD and VaD when used independently of one another

(Llorens et al., 2017). While a combination of tau and amyloid mark-

ers may somewhat increase the likelihood of successfully differentiat-

ing AD from VaD, the findings suggest that the identification of fur-

ther biomarkers is required (Llorens et al., 2017). As a consequence,

even the establishment of dementia subtypes based on clinical assess-

ments and currently available biomarkers may not yield an absolute

discrimination between AD and VaD.

Regarding the results of Study II in this thesis, the abovementioned

findings regarding the possible overlap between AD and VaD diagno-

ses indicate that some dementia cases in the Betula sample may have

been falsely classified. Possible false classifications could have some

bearing on the differential pattern of results observed regarding per-

formance in olfactory identification in prospective AD and VaD. As

discussed earlier, Study II suggested that olfactory identification

might predict dementia conversion to AD, but not to VaD. The results

118

were consistent with some previous findings, suggesting that olfactory

deficits may be less pronounced in VaD than in AD and that they

might even differentiate between the two (Duff et al., 2002; Knupfer

& Spiegel, 1986). By contrast, another study has shown similar effect

sizes for olfactory impairment for AD and VaD (Gray et al., 2001). As

discussed above, the inconsistencies of prior research studies may be

explained by more heterogeneous patterns of neuropathological local-

ization in VaD, in some cases affecting olfactory brain structures and

in other cases not (Alves, et al., 2014). However, the mixed findings

might have also arisen due to a failure to correctly differentiate be-

tween AD and VaD in some studies. Until marked improvement of

current diagnosis procedures is achieved in future studies, discussions

of the potential role of differential patterns of olfactory function for

AD and VaD remain speculative.

Effect sizes and clinical considerations Although there still is no cure for common forms of dementia such as

AD, identifying individuals at risk of conversion before overt manifes-

tation of pronounced cognitive deficits would be associated with sev-

eral benefits. An early identification may for example allow for earlier

intervention, which may potentially postpone the age at onset or even

improve outcomes (Adams et al., 2018). During recent years, the iden-

tification of early risk factors for dementia and other ill-health related

outcomes in older age has made significant progress. However, the

predictions that can be drawn from these variables at an individual

level are still far from exact. For example, none of the candidate bi-

119

omarkers for AD have so far achieved appropriate levels of sensitivity

and specificity in order to be recommended to be used for screening

purposes in the general population of older adults. Similarly, screen-

ing for cognitive impairment is not recommended based on the current

evidence of balances between benefits and harms (Alves et al., 2014).

A combination of biomarkers and cognitive status may render

somewhat higher diagnostic screening characteristics. A systematic

Cochrane review suggests that in individuals who were already diag-

nosed with MCI, CSF amyloid-b may have a sensitivity of around

81% and a specificity of around 64% for predicting AD-conversion

(Ritchie et al., 2014). Crucially however, the implementation of inter-

vention methods may already be too late at this time point.

Interestingly, the recently published study by Adams et al. (2018;

see Table 5) found that a five-item odor identification test had a sensi-

tivity of 47% and a specificity of 79% for AD conversion, already five

years in advance of any evidence of cognitive deficits. While these

numbers may compare well in comparison with established bi-

omarkers when cognitive decline is already in process, they also mean

that about 90% of the participants with low olfactory scores did not

convert to dementia within five years. Likewise, the results of Study II

of this thesis suggested that only one out of six individuals with an

objectively established poor sense of smell actually converted to de-

mentia within ten years, even though the risk was elevated compared

to persons with intact olfaction. Study III obtained similar results for

the risk of mortality.

Similar to biomarkers and cognitive status, the effect sizes of olfac-

tory ability for the prediction of dementia are thus quite small when

120

considered on the individual level. Olfactory deficits are caused by a

wide array of health-related variables apart from dementia-related

pathology. Likewise, dementia is caused by a wide array of variables,

of which only some may be related to olfactory function. Crucially

however, and in contrast to non-olfactory cognitive markers, olfactory

deficits may manifest already several years prior to the development

of the first dementia symptoms.

Furthermore, the results of this thesis suggest for the first time that

also subjective olfactory deficits may be of predictive value for estab-

lishing the risk of dementia conversion in addition to the odor identifi-

cation test. A combination of objective and subjective olfactory

measures may hence improve the predictive accuracy. A long-term

goal of future research is to establish how clinical settings could bene-

fit from utilizing these olfactory measures in combination with other

established behavioral and biological risk factors, such as non-

olfactory neuropsychological tests, structural and functional neuroim-

aging of the hippocampal and related structures, measures of bi-

omarkers in CSF, and genetic markers (Devanand et al., 2008).

The usage of olfactory measures in clinical settings for dementia

prediction may have several advantages. For example, tests of odor

identification function and subjective olfactory questions are non-

invasive, easy and fast to conduct. Olfactory deficits should, however,

never be used as the sole basis for determining an individual’s risk for

dementia or mortality.

121

Ethical considerations While the implementation of olfactory deficits in clinical settings still

warrants careful evaluation, the findings of this thesis and similar

studies may be of theoretical importance for our understanding about

the mechanisms that are involved in cognitive aging, dementia and

processes associated with an increased risk of mortality. However,

when involved in and confronted with research studies concerning

specific risk factors for developing dementia and other outcomes of ill

health, many older persons may worry about their own vulnerability to

such conditions. These worries may impose serious strains on older

adults, especially as there is still no available cure for most forms of

dementia.

Olfactory deficits are very common among older adults, far more

common than dementia diagnoses. A majority of individuals with a

poor sense of smell may thus needlessly worry when presented with

research findings regarding prospective associations between olfaction

and dementia.

Taking into account the potential negative effects of communi-

cating the results of this thesis to the public is of special relevance in

light of the finding that subjective olfactory assessments may be relat-

ed to an increased risk of dementia and mortality. The studies of this

thesis are the first to point out that introspective evaluation of one’s

own olfactory abilities may have some diagnostic value. While many

older adults are unaware of an olfactory impairment, cognitively intact

persons who in fact recognize a change of olfactory function may in

122

reality also be affected by one. However, the results of this thesis also

suggested that only a small part of individuals with an objectively es-

tablished poor sense of smell actually converted to dementia or died

within ten years, even though the risk was elevated compared to per-

sons with intact olfaction.

While research on prospective risk factors of ill health may always

cause discomfort among affected persons, this is not a valid argument

to cease investigating these relationships, even when the obtained ef-

fect sizes are small at the individual level. It is important to note that

researchers have a responsibility in how they communicate their find-

ings to the general public ensuring that exaggerations and unsupported

generalizations of the found effects are avoided as well as clearly stat-

ing that no individual conclusions can be drawn with certainty.

Suggestions for future studies of smell-based cognitive training Specific suggestions for future studies based on the findings in Studies

I to IV of this thesis have been provided throughout the previous sec-

tions. In this section I will focus on the broader consequences of the

findings for future research of cognitive training in older adults.

Overall, the findings of this thesis suggest that olfaction is intimate-

ly linked to the process of brain aging. An important question in light

of these and similar results concerns the utility of the olfactory system

for brain training in older age. As the usefulness of visual-based cog-

nitive training interventions has been questioned over recent years

(Simons et al., 2016), new methods for increasing cognitive function

are warranted. Recent studies suggest a remarkable level of biological

123

plasticity in the olfactory system (Fletcher, 2012; Kass, Rosenthal,

Pottackal, & McGann, 2013; Li, Howard, Parrish, & Gottfried, 2008)

indicating that healthy adult humans might improve their olfactory

functions from repeated odor assessments (Mainland et al., 2002).

Olfactory experts (e.g. perfumers and wine tasters) perform better

than non-experts on olfactory assessments (Royet, Plailly, Saive,

Veyrac, & Delon-Martin, 2013) and their expertise is correlated with

structural and functional reorganizations in brain areas that are in-

volved in learning and memory (Delon-Martin, Plailly, Fonlupt,

Veyrac, & Royet, 2013; Plailly, Delon-Martin, & Royet, 2012). Fur-

thermore, daily odor exposure might improve olfactory abilities in

patients with olfactory loss (Damm et al., 2014; Hummel et al., 2009)

and increase quality of life in older adults (Wegener, Croy, Hähner, &

Hummel, 2018).

While the biological mechanisms underlying cognitive transfer ef-

fects are poorly understood, recent work suggests that in working

memory training, the overlapping neural substrates of the trained and

transfer tasks is a key determinant of transfer (Dahlin, Neely, Larsson,

Bäckman, & Nyberg, 2008). While olfaction is well integrated with

the neural systems supporting memory encoding, it is yet unknown

whether olfactory-based training programs may benefit non-olfactory

cognitive functions that are engaged by overlapping brain regions. The

olfactory system may not only be responsive to training, but also serve

as a vehicle for improving non-olfactory memory functions.

124

Concluding remarks In this thesis, performance in olfactory identification was investigated

in longitudinal settings of cognitive decline, dementia conversion, and

risk mortality. The main findings of Studies I to IV can be summa-

rized as follows: (1) Long-term episodic memory decline was associ-

ated with odor identification deficits only in carriers of the ɛ4 allele of

the ApoE, a genetic risk factor for AD. (2) The association between

odor identification and episodic memory decline in carriers of the ɛ4

allele was not driven by deficits in non-olfactory semantic memory

(vocabulary), but was due to the olfactory nature of the identification

task. (3) Scores in odor identification could predict conversion to AD

over a time span of ten years in older adults with intact baseline cogni-

tion. (4) Scores in odor identification could predict risk of mortality

over a time span of up to ten years independent of dementia conver-

sion prior to death. (5) Self-assessed olfactory impairment could pre-

dict conversion to AD and VaD independent of performance in odor

identification. (6) Self-assessed long-term decline in olfactory func-

tion was predictive of a long-term decline in standardized measures of

odor identification, indicating that subjective long-term olfactory de-

cline may reflect an actual decline in odor ability. (7) Self-reported

olfactory decline was not predictive of a decline in odor identification

in older adults with low cognitive performance scores, indicating that

awareness of olfactory decline is linked to cognitive resources.

While the sense of smell may be our most neglected sense, it may

warn of an ongoing cognitive decline, impending dementia pathology,

125

and prospective outcomes of ill health. It is, however, important to

point out that many different mechanisms may underlie these out-

comes, of which only some may be linked to olfactory function. Fur-

thermore, the olfactory system is sensitive to the general aging process

and olfactory deficits are thus also common amongst healthy older

adults. Olfactory measures should therefore not be used in isolation

when determining the risk for cognitive decline, dementia or mortali-

ty. Rather, further research is needed to determine if they may be uti-

lized alongside other risk factors such as genetic variations, bi-

omarkers, measures of cortical volume, and other behavioral, cogni-

tive, and sensory assessments.

The direct consequences of olfactory loss may be less dramatic to

our everyday lives than loss of hearing or vision. This might explain

why many older individuals are unaware of alterations in their sense

of smell, especially those who already suffer from a cognitive im-

pairment (Devanand et al., 2000). However, the results of this thesis

suggest that persons who do subjectively experience a decline in their

olfactory function may also suffer from one. Amongst cognitively

healthy adults at baseline, this decline may even be indicative of later

dementia and an increased risk of mortality.

Furthermore, while olfaction may have a more subtle impact on

everyday aspects of our modern lives than hearing or vision, deficits

in odor function are nevertheless linked to many detrimental out-

comes, such a decreased quality of life (Croy et al., 2014). Methods

that can increase awareness of an olfactory dysfunction amongst older

adults are therefore warranted. In summary, this thesis highlights the

importance of the sense of smell in different aspects of human aging.

126

Summary in Swedish

Äldre personer har ofta en betydande luktnedsättning. Det finns flera

anledningar till varför luktsinnet försämras med ökad ålder Luktsy-

stemet är till exempel känsligt för upprepade inflammationer i luftvä-

garna som så småningom kan skada luktslemhinnan. I och med att vi

blir äldre, utsätts vi också för fler inflammationer, vilket till viss del

kan förklara varför luktnedsättningar är mer vanligt förekommande

hos äldre personer än hos yngre.

Intressant nog påverkar även hjärnans sjukdomar luktsinnet. Lukt-

nedsättningar är därför vanligare bland äldre personer med kognitiv

svikt, demens och andra neurodegenerativa tillstånd. Ett fåtal tidigare

studier har också hittat att äldre personer med nedsatt luktsinne löper

större risk för att dö i förtid jämfört med jämnåriga med intakt lukt-

funktion. Det har därför föreslagits att luktsinnet skulle kunna vara en

spegel för hur kroppen mår i övrigt.

Målet med den föreliggande avhandlingen var att, på ett longitudi-

nellt sätt (över flera år) och genom fyra studier, undersöka sambanden

mellan förmågan att identifiera lukter (att kunna matcha en lukt till ett

namn), förändringar i minnesförmåga och demens hos äldre, samt att

undersöka huruvida luktsinnet är kopplat till en ökad risk för att dö i

förtid. Målet var även att undersöka huruvida självupplevda luktför-

sämringar, utöver det objektiva luktidentifieringstestet, kan förutsäga

senare utveckling av demens och risken för mortalitet, och huruvida

127

de återspeglar en faktisk luktnedgång.

Studierna är baserade på ett urval av över tusen deltagare från det

Svenska Betula projektet, en longitudinell studie om åldrande, hälsa

och kognition, som har pågått i Umeå sedan slutet på 1980-talet. Re-

sultaten av den första studien visade att ett försämrat luktsinne var

vanligare bland äldre personer med avtagande minnesförmåga över en

tio års period, och som var bärare av apolipoproteinet E4 (ApoE4).

Denna genvariant bärs av cirka 25 procent av den svenska befolkning-

en och ökar kraftigt risken för Alzheimers sjukdom. Resultaten kan

tolkas som att genvariantens inverkan på både luktsinnet och minnet

kan bero på tidig demensutveckling i hjärnan.

Den andra studien visade att ett försämrat luktsinne var förknippat

med en högre risk för att utveckla demens inom 10 år och att även

självupplevda luktsinnesförsämringar kunde predicera demens. Studie

tre visade att personer med nedsatt luktförmåga hade större risk för att

dö inom 10 år och att sambandet mellan luktfunktion och mortalitet

även gällde för äldre personer utan diagnos av demenssjukdom. Resul-

tatet bekräftar tidigare forskning som tyder på att luktsinnet skulle

kunna ge viktiga ledtrådar om hur kroppen mår.

Studie fyra visade att en självupplevd luktnedgång under en tidspe-

riod av upp till 10 år hängde ihop med en faktisk försämring av lukt-

förmågan mätt med det objektiva luktidentifieringstestet under samma

tidsperiod. Äldre personer som upplever en gradvis försämring av sin

luktfunktion drabbas alltså troligtvis också av en faktisk luktnedgång.

Sammantaget tyder resultaten av avhandlingen på att luktsinnet är

nära förknippat med processer av kognitivt åldrande, demens och även

mortalitet.

128

Acknowledgements

First and foremost, I am endlessly grateful to my main supervisor,

Jonas Olofsson. Thank you for giving me the opportunity to work with

you and for supporting me along the way. You always encouraged me

to take on every possible challenge, which has given me the confi-

dence to work and to grow as a researcher. Thank you for letting me

be part of a large variety of interesting olfactory research projects and,

most of all, for always believing and trusting in me.

My second supervisor, Maria Larsson, I am immensely grateful for

your warm support, for your interest in my projects, and for encourag-

ing me whenever I have felt lost along the way. Thank you for your

guidance, not only research-wise.

Boo Johansson and Timo Mäntylä; your valuable comments have

significantly improved the quality of this thesis. Thank you!

Clara Dayet, thank you for creating the beautiful artwork for this

thesis, it would be so much duller without you.

Lars-Göran Nilsson, Lars Nyberg, and the staff involved in the Bet-

ula study, thank you for letting me be part of this exciting project, for

sharing your data with me and for your support. Maria Josefsson, I

immensely enjoyed working with you, thank you for your valuable

input and all the nice chats over the phone. I am also grateful to my

academic grandfather, Steven Nordin. Without you, olfactory research

would be a lot less fun.

129

During my time as a PhD student, I was fortunate to meet many smart

and inspiring teachers. Mats Nilsson and Anders Sjöberg, I can hardly

express the value of your knowledge and enthusiasm in research

methods and statistics, and I am grateful for all the time you have tak-

en to answer my questions.

An incredibly big thank you goes to all my dear friends and col-

leagues at the Gösta Ekman laboratory and the Psychology department

at large. Thank you for all the interesting and challenging discussions,

the immense support, and laughter.

I want to thank my Finnish family for your loving welcome and

your support. My parents, Gabriela and Lucian, I am forever grateful

for your never-ending love, for always being there for me and for all

the opportunities you have given me.

At last, my dearest family, Einar and Vilhelm. I cannot even imag-

ine how life would be like without coming home to you at the end of

each day. Thank you for just being you. Jag älskar er.

130

References

Adams, D. R., Pinto, J. M., Kern, D. W., Wroblewski, K. E., McClintock, M. K., & Dale, W. (2018). Olfactory Dysfunction Predicts Subsequent Dementia in Older US Adults. Journal of the American Geriatrics Society, 66(1), 140–144.

Albers, M. W., Tabert, M. H., & Devanand, D. P. (2006). Olfactory dysfunction as a predictor of neurodegenerative disease. Current Neurology and Neuroscience Reports, 6(5), 379–386.

Almkvist, O. (1996). Neuropsychological features of early Alzheimer’s disease: preclinical and clinical stages. Acta Neurologica Scandinavica, 94(S165), 63–71.

Alves, J., Petrosyan, A., & Magalhães, R. (2014). Olfactory dysfunction in demen-tia. World Journal of Clinical Cases : WJCC, 2(11), 661–667.

American Psychiatric Association. Task Force on DSM-IV. (1991). DSM-IV options book: Work in progress (7/1/91). American Psychiatric Press.

Andreasen, N., & Blennow, K. (2005). CSF biomarkers for mild cognitive impair-ment and early Alzheimer’s disease. Clinical Neurology and Neurosurgery, 107(3), 165–173.

Anstey, K. J., Luszcz, M. A., & Sanchez, L. (2001). A reevaluation of the common factor theory of shared variance among age, sensory function, and cognitive function in older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 56(1), 3–11.

Arnold, S. E., Hyman, B. T., Flory, J., Damasio, A. R., & Van Hoesen, G. W. (1991). The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cerebral Cortex, 1(1), 103–116.

Arnold, S. E., Hyman, B. T., & Van Hoesen, G. W. (1994). Neuropathologic chang-es of the temporal pole in Alzheimer’s disease and Pick’s disease. Archives of Neurology, 51(2), 145–150.

Arshamian, A., Iannilli, E., Gerber, J. C., Willander, J., Persson, J., Seo, H.-S., … Larsson, M. (2013). The functional neuroanatomy of odor evoked autobiograph-ical memories cued by odors and words. Neuropsychologia, 51(1), 123–131.

Aschenbrenner, K., Hummel, C., Teszmer, K., Krone, F., Ishimaru, T., Seo, H.-S., & Hummel, T. (2008). The influence of olfactory loss on dietary behaviors. The Laryngoscope, 118(1), 135–144.

Attems, J., Lintner, F., & Jellinger, K. (2005). Olfactory involvement in aging and Alzheimer’s disease: an autopsy study. Journal of Alzheimer’s Disease, 7(2), 149–157.

Attems, J., Thomas, A., & Jellinger, K. (2012). Correlations between cortical and subcortical tau pathology. Neuropathology and Applied Neurobiology, 38(6), 582–590.

131

Austin, T. O., Qiang, L., & Baas, P. W. (2017). Mechanisms of Neuronal Microtu-bule Loss in Alzheimer’s Disease. In Neuroprotection in Alzheimer’s Disease (pp. 59–71). Elsevier.

Bäckman, L., Small, B. J., & Fratiglioni, L. (2001). Stability of the preclinical epi-sodic memory deficit in Alzheimer’s disease. Brain, 124(1), 96–102.

Bacon, A. W., Bondi, M. W., Salmon, D. P., & Murphy, C. (1998). Very early changes in olfactory functioning due to Alzheimer’s disease and the role of apolipoprotein E in olfaction. Annals of the New York Academy of Sciences, 855(1), 723–731.

Bahar-Fuchs, A., Chételat, G., Villemagne, V. L., Moss, S., Pike, K., Masters, C. L., … Savage, G. (2010). Olfactory deficits and amyloid-β burden in Alzheimer’s disease, mild cognitive impairment, and healthy aging: a PiB PET study. Jour-nal of Alzheimer’s Disease, 22(4), 1081–1087.

Bahar-Fuchs, A., Moss, S., Rowe, C., & Savage, G. (2011). Awareness of olfactory deficits in healthy aging, amnestic mild cognitive impairment and Alzheimer’s disease. International Psychogeriatrics, 23(7), 1097–1106.

Balash, Y., Mordechovich, M., Shabtai, H., Giladi, N., Gurevich, T., & Korczyn, A. D. (2013). Subjective memory complaints in elders: depression, anxiety, or cog-nitive decline? Acta Neurologica Scandinavica, 127(5), 344–350.

Bédard, A., & Parent, A. (2004). Evidence of newly generated neurons in the human olfactory bulb. Developmental Brain Research, 151(1–2), 159–168.

Bende, M., & Nordin, S. (1997). Perceptual learning in olfaction: professional wine tasters versus controls. Physiology & Behavior, 62(5), 1065–1070.

Bertram, L., & Tanzi, R. E. (2008). Thirty years of Alzheimer’s disease genetics: the implications of systematic meta-analyses. Nature Reviews Neuroscience, 9(10), 768.

Bibl, M., Esselmann, H., & Wiltfang, J. (2012). Neurochemical biomarkers in Alz-heimer’s disease and related disorders. Therapeutic Advances in Neurological Disorders, 5(6), 335–348.

Birte-Antina, W., Ilona, C., Antje, H., & Thomas, H. (2018). Olfactory training with older people. International Journal of Geriatric Psychiatry, 33(1), 212–220.

Blom, E. S., Giedraitis, V., Zetterberg, H., Fukumoto, H., Blennow, K., Hyman, B. T., … Ingelsson, M. (2009). Rapid progression from mild cognitive impairment to Alzheimer’s disease in subjects with elevated levels of tau in cerebrospinal fluid and the APOE ε4/ε4 genotype. Dementia and Geriatric Cognitive Disor-ders, 27(5), 458–464.

Blomqvist, E. H., Bramerson, A., Stjarne, P., & Nordin, S. (2004). Consequences of olfactory loss and adopted coping strategies. Rhinology, 42(4), 189–194.

Bohnen, N. I., Müller, M. L., Kotagal, V., Koeppe, R. A., Kilbourn, M. A., Albin, R. L., & Frey, K. A. (2010). Olfactory dysfunction, central cholinergic integrity and cognitive impairment in Parkinson’s disease. Brain, 133(6), 1747–1754.

Bonfils, P., Faulcon, P., Tavernier, L., Bonfils, N. A., & Malinvaud, D. (2008). Home accidents associated with anosmia. Presse Medicale (Paris, France: 1983), 37(5 Pt 1), 742–745.

Boyce, J. M., & Shone, G. R. (2006). Effects of ageing on smell and taste. Post-graduate Medical Journal, 82(966), 239–241.

Braak, H., & Braak, E. (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathologica, 82(4), 239–259.

Braak, H., & Braak, E. V. A. (1995). Staging of Alzheimer’s disease-related neuro-fibrillary changes. Neurobiology of Aging, 16(3), 271–278.

132

Brämerson, A., Johansson, L., Ek, L., Nordin, S., & Bende, M. (2004). Prevalence of Olfactory Dysfunction: The Skövde Population-Based Study. The Laryngo-scope, 114(4), 733–737.

Breitner, J. C., Welsh, K. A., Helms, M. J., Gaskell, P. C., Gau, B. A., Roses, A. D., … Saunders, A. M. (1995). Delayed onset of Alzheimer’s disease with non-steroidal anti-inflammatory and histamine H2 blocking drugs. Neurobiology of Aging, 16(4), 523–530.

Buckley, R. F., Maruff, P., Ames, D., Bourgeat, P., Martins, R. N., Masters, C. L., … Savage, G. (2016). Subjective memory decline predicts greater rates of clini-cal progression in preclinical Alzheimer’s disease. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, 12(7), 796–804.

Bunce, D., Fratiglioni, L., Small, B. J., Winblad, B., & Bäckman, L. (2004). APOE and cognitive decline in preclinical Alzheimer disease and non-demented aging. Neurology, 63(5), 816–821.

Cain, W. S., & Gent, J. F. (1991). Olfactory sensitivity: Reliability, generality, and association with aging. Journal of Experimental Psychology: Human Perception and Performance, 17(2), 382.

Carmichael, S. T., Clugnet, M.-C., & Price, J. L. (1994). Central olfactory connec-tions in the macaque monkey. Journal of Comparative Neurology, 346(3), 403–434.

Caruana, E. J., Roman, M., Hernández-Sánchez, J., & Solli, P. (2015). Longitudinal studies. Journal of Thoracic Disease, 7(11), E537.

Chatfield, M. D., Brayne, C. E., & Matthews, F. E. (2005). A systematic literature review of attrition between waves in longitudinal studies in the elderly shows a consistent pattern of dropout between differing studies. Journal of Clinical Epi-demiology, 58(1), 13–19.

Chess, A., Simon, I., Cedar, H., & Axel, R. (1994). Allelic inactivation regulates olfactory receptor gene expression. Cell, 78(5), 823–834.

Christ, S. L., Zheng, D. D., Swenor, B. K., Lam, B. L., West, S. K., Tannenbaum, S. L., … Lee, D. J. (2014). Longitudinal relationships among visual acuity, daily functional status, and mortality: the Salisbury Eye Evaluation Study. JAMA Ophthalmology, 132(12), 1400–1406.

Cleland, T. A., Johnson, B. A., Leon, M., & Linster, C. (2007). Relational represen-tation in the olfactory system. Proceedings of the National Academy of Sciences, 104(6), 1953–1958.

Conti, M. Z., Vicini-Chilovi, B., Riva, M., Zanetti, M., Liberini, P., Padovani, A., & Rozzini, L. (2013). Odor identification deficit predicts clinical conversion from mild cognitive impairment to dementia due to Alzheimer’s disease. Archives of Clinical Neuropsychology, 28(5), 391–399.

Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell, P. C., Small, G., ... & Pericak-Vance, M. A. (1993). Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science, 261(5123), 921-923.

Croy, I., & Hummel, T. (2017). Olfaction as a marker for depression. Journal of Neurology, 264(4), 631–638.

Croy, I., Symmank, A., Schellong, J., Hummel, C., Gerber, J., Joraschky, P., & Hummel, T. (2014). Olfaction as a marker for depression in humans. Journal of Affective Disorders, 160, 80–86.

Cruickshanks, K. J., Wiley, T. L., Tweed, T. S., Klein, B. E., Klein, R., Mares-Perlman, J. A., & Nondahl, D. M. (1998). Prevalence of hearing loss in older adults in Beaver Dam, Wisconsin: The epidemiology of hearing loss study. American Journal of Epidemiology, 148(9), 879–886.

133

Dahlin, E., Neely, A. S., Larsson, A., Bäckman, L., & Nyberg, L. (2008). Transfer of learning after updating training mediated by the striatum. Science, 320(5882), 1510–1512.

Dalton, D. S., Cruickshanks, K. J., Klein, B. E., Klein, R., Wiley, T. L., & Nondahl, D. M. (2003). The impact of hearing loss on quality of life in older adults. The Gerontologist, 43(5), 661–668.

Damm, M., Pikart, L. K., Reimann, H., Burkert, S., Göktas, Ö., Haxel, B., … Ren-ner, B. (2014). Olfactory training is helpful in postinfectious olfactory loss: a randomized, controlled, multicenter study. The Laryngoscope, 124(4), 826–831.

Davies, G., Armstrong, N., Bis, J. C., Bressler, J., Chouraki, V., Giddaluru, S., … Lahti, J. (2015). Genetic contributions to variation in general cognitive function: a meta-analysis of genome-wide association studies in the CHARGE consortium (N= 53 949). Molecular Psychiatry, 20(2), 183.

De Wijk, R. A., & Cain, W. S. (1994). Odor quality: discrimination versus free and cued identification. Perception & Psychophysics, 56(1), 12–18.

Deems, D. A., Doty, R. L., Settle, R. G., Moore-Gillon, V., Shaman, P., Mester, A. F., … Snow, J. B. (1991). Smell and taste disorders, a study of 750 patients from the University of Pennsylvania Smell and Taste Center. Archives of Oto-laryngology–head & Neck Surgery, 117(5), 519–528.

Del Tredici, K., & Braak, H. (2008). Neurofibrillary changes of the Alzheimer type in very elderly individuals: neither inevitable nor benign: Commentary on “No disease in the brain of a 115-year-old woman.” Neurobiology of Aging, 29(8), 1133–1136.

Delon-Martin, C., Plailly, J., Fonlupt, P., Veyrac, A., & Royet, J.-P. (2013). Perfum-ers’ expertise induces structural reorganization in olfactory brain regions. Neu-roimage, 68, 55–62.

Devanand, D. P., Lee, S., Manly, J., Andrews, H., Schupf, N., Doty, R. L., ... & Mayeux, R. (2015). Olfactory deficits predict cognitive decline and Alzheimer dementia in an urban community. Neurology, 84(2), 182-189.

Devanand, D. P., Lee, S., Manly, J., Andrews, H., Schupf, N., Masurkar, A., … Doty, R. L. (2015). Olfactory identification deficits and increased mortality in the community. Annals of Neurology, 78(3), 401–411.

Devanand, D. P., Liu, X., Tabert, M. H., Pradhaban, G., Cuasay, K., Bell, K., … Pelton, G. H. (2008). Combining early markers strongly predicts conversion from mild cognitive impairment to Alzheimer’s disease. Biological Psychiatry, 64(10), 871–879.

Devanand, D. P., Michaels-Marston, K. S., Liu, X., Pelton, G. H., Padilla, M., Marder, K., … Mayeux, R. (2000). Olfactory deficits in patients with mild cog-nitive impairment predict Alzheimer’s disease at follow-up. American Journal of Psychiatry, 157(9), 1399–1405.

Dik, M. G., Jonker, C., Comijs, H. C., Bouter, L. M., Twisk, J. W. R., Van Kamp, G. J., & Deeg, D. J. H. (2001). Memory complaints and APOE-ε4 accelerate cognitive decline in cognitively normal elderly. Neurology, 57(12), 2217–2222.

Djordjevic, J., Jones-Gotman, M., De Sousa, K., & Chertkow, H. (2008). Olfaction in patients with mild cognitive impairment and Alzheimer’s disease. Neurobiol-ogy of Aging, 29(5), 693–706.

Doty, R. L. (2012). Olfactory dysfunction in Parkinson disease. Nature Reviews Neurology, 8(6), 329.

Doty, R. L., & Kamath, V. (2014). The influences of age on olfaction: a review. Frontiers in Psychology, 5, 20.

134

Doty, R. L., Marcus, A., & William Lee, W. (1996). Development of the 12-item cross-cultural smell identification test (CC-SIT). The Laryngoscope, 106(3), 353–356.

Doty, R. L., McKeown, D. A., Lee, W. W., & Shaman, P. (1995). A study of the test-retest reliability of ten olfactory tests. Chemical Senses, 20(6), 645–656.

Doty, R. L., & Mishra, A. (2001). Olfaction and its alteration by nasal obstruction, rhinitis, and rhinosinusitis. The Laryngoscope, 111(3), 409–423.

Doty, R. L., Petersen, I., Mensah, N., & Christensen, K. (2011). Genetic and envi-ronmental influences on odor identification ability in the very old. Psychology and Aging, 26(4), 864.

Doty, R. L., Shaman, P., Applebaum, S. L., Giberson, R., Siksorski, L., & Rosen-berg, L. (1984). Smell identification ability: changes with age. Science, 226(4681), 1441–1443.

Doty, R. L., Shaman, P., Kimmelman, C. P., & Dann, M. S. (1984). University of Pennsylvania Smell Identification Test: a rapid quantitative olfactory function test for the clinic. The Laryngoscope, 94(2), 176–178.

Doty, R. L., Smith, R., McKeown, D. A., & Raj, J. (1994). Tests of human olfactory function: principal components analysis suggests that most measure a common source of variance. Perception & Psychophysics, 56(6), 701–707.

Dubois, B., Feldman, H. H., Jacova, C., Cummings, J. L., DeKosky, S. T., Bar-berger-Gateau, P., … Galasko, D. (2010). Revising the definition of Alz-heimer’s disease: a new lexicon. The Lancet Neurology, 9(11), 1118–1127.

Dubois, B., Feldman, H. H., Jacova, C., DeKosky, S. T., Barberger-Gateau, P., Cummings, J., … Jicha, G. (2007). Research criteria for the diagnosis of Alz-heimer’s disease: revising the NINCDS–ADRDA criteria. The Lancet Neurolo-gy, 6(8), 734–746.

Duff, K., McCaffrey, R. J., & Solomon, G. S. (2002). The Pocket Smell Test: suc-cessfully discriminating probable Alzheimer’s dementia from vascular dementia and major depression. The Journal of Neuropsychiatry and Clinical Neurosci-ences, 14(2), 197–201.

Dulay, M. F., Gesteland, R. C., Shear, P. K., Ritchey, P. N., & Frank, R. A. (2008). Assessment of the influence of cognition and cognitive processing speed on three tests of olfaction. Journal of Clinical and Experimental Neuropsychology, 30(3), 327–337.

Dulay, M. F., & Murphy, C. (2002). Olfactory acuity and cognitive function con-verge in older adulthood: Support for the common cause hypothesis. Psychology and Aging, 17(3), 392.

Dureman, I., Eriksson, U.-B., Kebbon, L., & Österberg, E. (1971). Manual till DS-batteriet. Skandinaviska testförlaget.

Eichenbaum, H. (2000). A cortical–hippocampal system for declarative memory. Nature Reviews Neuroscience, 1(1), 41.

Etminan, M., Gill, S., & Samii, A. (2003). Effect of non-steroidal anti-inflammatory drugs on risk of Alzheimer’s disease: systematic review and meta-analysis of observational studies. BMJ, 327(7407), 128.

Ewers, M., Walsh, C., Trojanowski, J. Q., Shaw, L. M., Petersen, R. C., Jack, C. R., … Scheltens, P. (2012). Prediction of conversion from mild cognitive impair-ment to Alzheimer’s disease dementia based upon biomarkers and neuropsycho-logical test performance. Neurobiology of Aging, 33(7), 1203–1214.

Finkel, D., Pedersen, N. L., & Larsson, M. (2001). Olfactory functioning and cogni-tive abilities: a twin study. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 56(4), P226–P233.

135

Finkel, D., Reynolds, C. A., Larsson, M., Gatz, M., & Pedersen, N. L. (2011). Both odor identification and ApoE-ε4 contribute to normative cognitive aging. Psy-chology and Aging, 26(4), 872.

Fischer, M. E., Cruickshanks, K. J., Schubert, C. R., Pinto, A. A., Carlsson, C. M., Klein, B. E., … Tweed, T. S. (2016). Age-Related Sensory Impairments and Risk of Cognitive Impairment. Journal of the American Geriatrics Society, 64(10), 1981–1987.

Fletcher, M. L. (2012). Olfactory aversive conditioning alters olfactory bulb mi-tral/tufted cell glomerular odor responses. Frontiers in Systems Neuroscience, 6, 16.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198.

Ford, G. R., Haley, W. E., Thrower, S. L., West, C. A., & Harrell, L. E. (1996). Utility of Mini-Mental State Exam scores in predicting functional impairment among white and African American dementia patients. The Journals of Geron-tology Series A: Biological Sciences and Medical Sciences, 51(4), M185–M188.

Franks, K. H., Chuah, M. I., King, A. E., & Vickers, J. C. (2015). Connectivity of pathology: The olfactory system as a model for network-driven mechanisms of Alzheimer’s disease pathogenesis. Frontiers in Aging Neuroscience, 7, 234.

Gauthier, S., Reisberg, B., Zaudig, M., Petersen, R. C., Ritchie, K., Broich, K., … Chertkow, H. (2006). Mild cognitive impairment. The Lancet, 367(9518), 1262–1270.

Gescheider, G. (1997). Chapter 3: The Classical Psychophysical Methods. Psycho-physics: The Fundamentals. 3rd Ed. Mahwah: Lawrence Erlbaum Associates.

Gilbert, P. E., & Murphy, C. (2004). The effect of the ApoE ε4 allele on recognition memory for olfactory and visual stimuli in patients with pathologically con-firmed Alzheimer’s disease, probable Alzheimer’s disease, and healthy elderly controls. Journal of Clinical and Experimental Neuropsychology, 26(6), 779–794.

Gopinath, B., Sue, C. M., Kifley, A., & Mitchell, P. (2011). The association between olfactory impairment and total mortality in older adults. Journals of Gerontolo-gy Series A: Biomedical Sciences and Medical Sciences, 67(2), 204–209.

Gottfried, J. A. (2010). Central mechanisms of odour object perception. Nature Reviews Neuroscience, 11(9), 628–641. https://doi.org/10.1038/nrn2883

Gottfried, J. A. (2015). Structural and functional imaging of the human olfactory system. Handbook of Olfaction and Gustation, 279–303.

Gottfried, J. A., & Zald, D. H. (2005). On the scent of human olfactory orbitofrontal cortex: meta-analysis and comparison to non-human primates. Brain Research Reviews, 50(2), 287–304.

Graves, A. B., Bowen, J. D., Rajaram, L., McCormick, W. C., McCurry, S. M., Schellenberg, G. D., & Larson, E. B. (1999). Impaired olfaction as a marker for cognitive decline Interaction with apolipoprotein E ε4 status. Neurology, 53(7), 1480–1480.

Gray, A. J., Staples, V., Murren, K., Dhariwal, A., & Bentham, P. (2001). Olfactory identification is impaired in clinic-based patients with vascular dementia and senile dementia of Alzheimer type. International Journal of Geriatric Psychia-try, 16(5), 513–517.

Greene, J. D., Baddeley, A. D., & Hodges, J. R. (1996). Analysis of the episodic memory deficit in early Alzheimer’s disease: evidence from the doors and peo-ple test. Neuropsychologia, 34(6), 537–551.

136

Growdon, M. E., Schultz, A. P., Dagley, A. S., Amariglio, R. E., Hedden, T., Rentz, D. M., … Marshall, G. A. (2015). Odor identification and Alzheimer disease bi-omarkers in clinically normal elderly. Neurology, 84(21), 2153–2160.

Hardy, J. A., & Higgins, G. A. (1992). Alzheimer’s disease: the amyloid cascade hypothesis. Science, 256(5054), 184.

Hashimoto, M., Yasuda, M., Tanimukai, S., Matsui, M., Hirono, N., Kazui, H., & Mori, E. (2001). Apolipoprotein E ε4 and the pattern of regional brain atrophy in Alzheimer’s disease. Neurology, 57(8), 1461–1466.

Hedner, M., Larsson, M., Arnold, N., Zucco, G. M., & Hummel, T. (2010). Cogni-tive factors in odor detection, odor discrimination, and odor identification tasks. Journal of Clinical and Experimental Neuropsychology, 32(10), 1062–1067.

Hedner, M., Nilsson, L., Olofsson, J. K., Bergman, O., Eriksson, E., Nyberg, L., & Larsson, M. (2010). Age-related olfactory decline is associated with the BDNF val66met polymorphism: evidence from a population-based study. Frontiers in aging neuroscience, 2, 24.

Heneka, M. T., & O’Banion, M. K. (2007). Inflammatory processes in Alzheimer’s disease. Journal of Neuroimmunology, 184(1), 69–91.

Hirai, T., Kojima, S., Shimada, A., Umemura, T., Sakai, M., & Itaku-rat, C. (1996). Age related changes in the olfactory system of dogs. Neuropathology and ap-plied neurobiology, 22(6), 531–539.

Hummel, T., Kobal, G., Gudziol, H., & Mackay-Sim, A. (2007). Normative data for the “Sniffin’Sticks” including tests of odor identification, odor discrimination, and olfactory thresholds: an upgrade based on a group of more than 3,000 sub-jects. European Archives of Oto-Rhino-Laryngology, 264(3), 237–243.

Hummel, T., & Nordin, S. (2005). Olfactory disorders and their consequences for quality of life. Acta Oto-Laryngologica, 125(2), 116–121.

Hummel, T., Rissom, K., Reden, J., Hähner, A., Weidenbecher, M., & Hüttenbrink, K.-B. (2009). Effects of olfactory training in patients with olfactory loss. The Laryngoscope, 119(3), 496–499.

Hummel, T., Sekinger, B., Wolf, S. R., Pauli, E., & Kobal, G. (1997). ‘Sniffin’sticks’: olfactory performance assessed by the combined testing of odor identification, odor discrimination and olfactory threshold. Chemical Senses, 22(1), 39–52.

Hussain, A., Luong, M., Pooley, A., & Nathan, B. P. (2013). Isoform-specific ef-fects of ApoE on neurite outgrowth in Olfactory Epithelium culture. Journal of Biomedical Science, 20(1), 49.

Jack, C. R., Albert, M. S., Knopman, D. S., McKhann, G. M., Sperling, R. A., Car-rillo, M. C., … Phelps, C. H. (2011). Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diag-nostic guidelines for Alzheimer’s disease. Alzheimer’s & Dementia: The Jour-nal of the Alzheimer’s Association, 7(3), 257–262.

Jack Jr, C. R., Knopman, D. S., Jagust, W. J., Petersen, R. C., Weiner, M. W., Aisen, P. S., … Weigand, S. D. (2013). Tracking pathophysiological processes in Alz-heimer’s disease: an updated hypothetical model of dynamic biomarkers. The Lancet Neurology, 12(2), 207–216.

Jansen, W. J., Ossenkoppele, R., Knol, D. L., Tijms, B. M., Scheltens, P., Verhey, F. R. J., … Zetterberg, H. (2015). Prevalence of Cerebral Amyloid Pathology in Persons Without Dementia: A Meta-analysis. JAMA, 313(19), 1924–1938.

Jessen, F., Amariglio, R. E., Van Boxtel, M., Breteler, M., Ceccaldi, M., Chételat, G., … Van Der Flier, W. M. (2014). A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, 10(6), 844–852.

137

Jessen, F., Wolfsgruber, S., Wiese, B., Bickel, H., Mösch, E., Kaduszkiewicz, H., … Fuchs, A. (2014). AD dementia risk in late MCI, in early MCI, and in subjective memory impairment. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, 10(1), 76–83.

Johnson, B. A., & Leon, M. (2007). Chemotopic odorant coding in a mammalian olfactory system. Journal of Comparative Neurology, 503(1), 1–34.

Jönsson, F. U., Møller, P., & Olsson, M. J. (2011). Olfactory working memory: effects of verbalization on the 2-back task. Memory & Cognition, 39(6), 1023–1032.

Jorm, A. F., Christensen, H., Henderson, A. S., Korten, A. E., Mackinnon, A. J., & Scott, R. (1994). Complaints of cognitive decline in the elderly: a comparison of reports by subjects and informants in a community survey. Psychological Medi-cine, 24(2), 365–374.

Jorm, A. F., Christensen, H., Korten, A. E., Jacomb, P. A., & Henderson, A. S. (2001). Memory complaints as a precursor of memory impairment in older peo-ple: a longitudinal analysis over 7–8 years. Psychological Medicine, 31(3), 441–449.

Josefsson, M., Larsson, M., Nordin, S., Adolfsson, R., & Olofsson, J. (2017). AP-OE-ɛ4 effects on longitudinal decline in olfactory and non-olfactory cognitive abilities in middle-aged and old adults. Scientific Reports, 7(1), 1286.

Josefsson, M., Luna, X., Pudas, S., Nilsson, L.-G., & Nyberg, L. (2012). Genetic and lifestyle predictors of 15-year longitudinal change in episodic memory. Journal of the American Geriatrics Society, 60(12), 2308–2312.

Kaneda, H., Maeshima, K., Goto, N., Kobayakawa, T., Ayabe-Kanamura, S., & Saito, S. (2000). Decline in taste and odor discrimination abilities with age, and relationship between gustation and olfaction. Chemical Senses, 25(3), 331–337.

Kass, M. D., Rosenthal, M. C., Pottackal, J., & McGann, J. P. (2013). Fear learning enhances neural responses to threat-predictive sensory stimuli. Science, 342(6164), 1389–1392.

Kjelvik, G., Evensmoen, H. R., Brezova, V., & Håberg, A. K. (2012). The human brain representation of odor identification. Journal of neurophysiology, 108(2), 645-657.

Knupfer, L., & Spiegel, R. (1986). Differences in olfactory test performance be-tween normal aged, Alzheimer and vascular type dementia individuals. Interna-tional Journal of Geriatric Psychiatry, 1(1), 3–14.

Kobal, G., Hummel, T. H., Sekinger, B., Barz, S., Roscher, S., & Wolf, S. (1996). “ Sniffin’sticks”: screening of olfactory performance. Rhinology, 34(4), 222–226.

Kobal, G., Klimek, L., Wolfensberger, M., Gudziol, H., Temmel, A., Owen, C. M., … Hummel, T. (2000). Multicenter investigation of 1,036 subjects using a standardized method for the assessment of olfactory function combining tests of odor identification, odor discrimination, and olfactory thresholds. European Ar-chives of Oto-Rhino-Laryngology, 257(4), 205–211.

Kohli, P., Soler, Z. M., Nguyen, S. A., Muus, J. S., & Schlosser, R. J. (2016). The association between olfaction and depression: a systematic review. Chemical Senses, 41(6), 479–486.

Koss, E., Weiffenbach, J. M., Haxby, J. V., & Friedland, R. P. (1988). Olfactory detection and identification performance are dissociated in early Alzheimer’s disease. Neurology, 38(8), 1228–1228.

Kovacs, T., Cairns, N. J., & Lantos, P. L. (1999). Beta-amyloid deposition and neu-rofibrillary tangle formation in the olfactory bulb in ageing and Alzheimer’s disease. Neuropathology and Applied Neurobiology, 25(6), 481–491.

138

Krantz, E. M., Schubert, C. R., Dalton, D. S., Zhong, W., Huang, G. H., Klein, B. E. K., … Cruickshanks, K. J. (2009). Test–retest reliability of the San Diego Odor Identification Test and comparison with the brief smell identification test. Chemical Senses, 34(5), 435–440.

Kumar, A., & Singh, A. (2015). A review on Alzheimer’s disease pathophysiology and its management: an update. Pharmacological Reports, 67(2), 195–203.

LaMantia, A.-S., & Purves, D. (1989). Development of glomerular pattern visual-ized in the olfactory bulbs of living mice. Nature, 341(6243), 646.

Landis, B. N., Hummel, T., Hugentobler, M., Giger, R., & Lacroix, J. S. (2003). Ratings of overall olfactory function. Chemical Senses, 28(8), 691–694.

Landis, B. N., Konnerth, C. G., & Hummel, T. (2004). A study on the frequency of olfactory dysfunction. The Laryngoscope, 114(10), 1764–1769.

Larsson, M., Finkel, D., & Pedersen, N. L. (2000). Odor identification: influences of age, gender, cognition, and personality. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 55(5), P304–P310.

Larsson, M., Hedner, M., Papenberg, G., Seubert, J., Bäckman, L., & Laukka, E. J. (2016). Olfactory memory in the old and very old: relations to episodic and se-mantic memory and APOE genotype. Neurobiology of Aging, 38, 118–126.

Larsson, M., Nilsson, L.-G., Olofsson, J. K., & Nordin, S. (2004). Demographic and cognitive predictors of cued odor identification: evidence from a population-based study. Chemical Senses, 29(6), 547–554.

Larsson, M., Öberg, C., & Bäckman, L. (2005). Odor identification in old age: De-mographic, sensory and cognitive correlates. Aging, Neuropsychology, and Cognition, 12(3), 231–244.

Larsson, M., Semb, H., Winblad, B., Amberla, K., Wahlund, L.-O., & Bäckman, L. (1999). Odor identification in normal aging and early Alzheimer’s disease: Ef-fects of retrieval support. Neuropsychology, 13(1), 47.

Laukka, E. J., Lövdén, M., Herlitz, A., Karlsson, S., Ferencz, B., Pantzar, A., … Bäckman, L. (2013). Genetic effects on old-age cognitive functioning: a popula-tion-based study. Psychology and Aging, 28(1), 262–274.

Laws, S. M., Clarnette, R. M., Taddei, K., Martins, G., Paton, A., Hallmayer, J., … Förstl, H. (2002). APOE-ε4 and APOE- 491A polymorphisms in individuals with subjective memory loss. Molecular Psychiatry, 7(7), 768.

Le Guérer, A. (2002). Olfaction and cognition: A philosophical and psychoanalytic view. Olfaction, Taste, and Cognition, 3–15.

Lehrner, J. P., Glück, J., & Laska, M. (1999). Odor identification, consistency of label use, olfactory threshold and their relationships to odor memory over the human lifespan. Chemical Senses, 24(3), 337–346.

Li, W., Howard, J. D., Parrish, T. B., & Gottfried, J. A. (2008). Aversive learning enhances perceptual and cortical discrimination of indiscriminable odor cues. Science, 319(5871), 1842-1845.

Lin, F. R., Metter, E. J., O’Brien, R. J., Resnick, S. M., Zonderman, A. B., & Fer-rucci, L. (2011). Hearing loss and incident dementia. Archives of Neurology, 68(2), 214–220.

Lin, M. Y., Gutierrez, P. R., Stone, K. L., Yaffe, K., Ensrud, K. E., Fink, H. A., … Mangione, C. M. (2004). Vision impairment and combined vision and hearing impairment predict cognitive and functional decline in older women. Journal of the American Geriatrics Society, 52(12), 1996–2002.

Lind, J., Larsson, A., Persson, J., Ingvar, M., Nilsson, L.-G., Bäckman, L., … Van Broeckhoven, C. (2006). Reduced hippocampal volume in non-demented carri-ers of the apolipoprotein E ɛ4: relation to chronological age and recognition memory. Neuroscience Letters, 396(1), 23–27.

139

Lindenberger, U., & Baltes, P. B. (1994). Sensory functioning and intelligence in old age: a strong connection. Psychology and Aging, 9(3), 339.

Lipnicki, D. M., Sachdev, P. S., Crawford, J., Reppermund, S., Kochan, N. A., Trol-lor, J. N., ... & Mather, K. A. (2013). Risk factors for late-life cognitive decline and variation with age and sex in the Sydney Memory and Ageing Study. PloS one, 8(6), e65841.

Llorens, F., Schmitz, M., Knipper, T., Schmidt, C., Lange, P., Fischer, A., … Zerr, I. (2017). Cerebrospinal Fluid Biomarkers of Alzheimer’s Disease Show Different but Partially Overlapping Profile Compared to Vascular Dementia. Frontiers in Aging Neuroscience, 9.

Lobo, A., Launer, L. J., Fratiglioni, L., Andersen, K., Di Carlo, A., Breteler, M. M. B., … Martinez-Lage, J. (2000). Prevalence of dementia and major subtypes in Europe: a collaborative study of population-based cohorts. Neurology, 54(5), S4.

Lodovichi, C., & Belluscio, L. (2012). Odorant receptors in the formation of the olfactory bulb circuitry. Physiology, 27(4), 200–212.

Loo, A. T., Youngentob, S. L., Kent, P. F., & Schwob, J. E. (1996). The aging olfac-tory epithelium: neurogenesis, response to damage, and odorant-induced activi-ty. International Journal of Developmental Neuroscience, 14(7–8), 881–900.

Lötsch, J., Reichmann, H., & Hummel, T. (2008). Different Odor Tests Contribute Differently to the Evaluation of Olfactory Loss. Chemical Senses, 33(1), 17–21.

Lövdén, M., Rönnlund, M., Wahlin, Å., Bäckman, L., Nyberg, L., & Nilsson, L. G. (2004). The extent of stability and change in episodic and semantic memory in old age: Demographic predictors of level and change. The Journals of Geron-tology Series B: Psychological Sciences and Social Sciences, 59(3), P130-P134.

Luzzi, S., Snowden, J. S., Neary, D., Coccia, M., Provinciali, L., & Ralph, M. A. L. (2007). Distinct patterns of olfactory impairment in Alzheimer’s disease, seman-tic dementia, frontotemporal dementia, and corticobasal degeneration. Neuro-psychologia, 45(8), 1823–1831.

Mainland, J. D., Bremner, E. A., Young, N., Johnson, B. N., Khan, R. M., Bensafi, M., & Sobel, N. (2002). Olfactory plasticity: one nostril knows what the other learns. Nature, 419(6909), 802.

Manly, J. J., Tang, M.-X., Schupf, N., Stern, Y., Vonsattel, J.-P. G., & Mayeux, R. (2008). Frequency and course of mild cognitive impairment in a multiethnic community. Annals of Neurology, 63(4), 494–506.

McGeer, P. L., Schulzer, M., & McGeer, E. G. (2001). Anti-inflammatory agents as possible protective factors for Alzheimer’s disease: Analysis of relevant epide-miological studies. In Neuroinflammatory Mechanisms in Alzheimer’s Disease Basic and Clinical Research (pp. 53–64). Springer.

McKhann, G. M., Knopman, D. S., Chertkow, H., Hyman, B. T., Jack, C. R., Ka-was, C. H., … Phelps, C. H. (2011). The diagnosis of dementia due to Alz-heimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, 7(3), 263–269.

McLaughlin, N. C., & Westervelt, H. J. (2008). Odor identification deficits in fron-totemporal dementia: a preliminary study. Archives of Clinical Neuropsycholo-gy, 23(1), 119–123.

McShane, R. H., Nagy, Z., Esiri, M. M., King, E., Joachim, C., Sullivan, N., & Smith, A. D. (2001). Anosmia in dementia is associated with Lewy bodies ra-ther than Alzheimer’s pathology. Journal of Neurology, Neurosurgery & Psy-chiatry, 70(6), 739–743.

140

Menon, C., Westervelt, H. J., Jahn, D. R., Dressel, J. A., & O’Bryant, S. E. (2013). Normative performance on the Brief Smell Identification Test (BSIT) in a multi-ethnic bilingual cohort: A Project FRONTIER study. The Clinical Neuropsy-chologist, 27(6), 946–961.

Mesholam, R. I., Moberg, P. J., Mahr, R. N., & Doty, R. L. (1998). Olfaction in neurodegenerative disease: a meta-analysis of olfactory functioning in Alz-heimer’s and Parkinson’s diseases. Archives of Neurology, 55(1), 84–90.

Michaelson, D. M. (2014). APOE ε4: The most prevalent yet understudied risk fac-tor for Alzheimer’s disease. Alzheimer’s & Dementia: The Journal of the Alz-heimer’s Association, 10(6), 861–868.

Miwa, T., Furukawa, M., Tsukatani, T., Costanzo, R. M., DiNardo, L. J., & Reiter, E. R. (2001). Impact of olfactory impairment on quality of life and disability. Archives of Otolaryngology–Head & Neck Surgery, 127(5), 497–503.

Moberg, P. J., Agrin, R., Gur, R. E., Gur, R. C., Turetsky, B. I., & Doty, R. L. (1999). Olfactory dysfunction in schizophrenia: a qualitative and quantitative review. Neuropsychopharmacology, 21(3), 325–340.

Morgan, C. D., Nordin, S., & Murphy, C. (1995). Odor identification as an early marker for Alzheimer’s disease: impact of lexical functioning and detection sen-sitivity. Journal of Clinical and Experimental Neuropsychology, 17(5), 793–803.

Morris, M. C. (2012). Nutritional determinants of cognitive aging and dementia. Proceedings of the Nutrition Society, 71(1), 1–13.

Mosconi, L., Perani, D., Sorbi, S., Herholz, K., Nacmias, B., Holthoff, V., … Pado-vani, A. (2004). MCI conversion to dementia and the APOE genotype A predic-tion study with FDG-PET. Neurology, 63(12), 2332–2340.

Mungas, D. (1991). In-office mental status testing: a practical guide. Geriatrics, 46(7), 54–58, 63, 66.

Murphy, C., Cain, W. S., Gilmore, M. M., & Skinner, R. B. (1991). Sensory and semantic factors in recognition memory for odors and graphic stimuli: elderly versus young persons. The American Journal of Psychology, 161–192.

Murphy, C., Doty, R. L., & Duncan, H. J. (2003). Clinical disorders of olfaction. Neurological Disease and Therapy, 57, 461–478.

Murphy, C., Gilmore, M. M., Seery, C. S., Salmon, D. P., & Lasker, B. R. (1990). Olfactory thresholds are associated with degree of dementia in Alzheimer’s dis-ease. Neurobiology of Aging, 11(4), 465–469.

Murphy, C., Jernigan, T. L., & Fennema-Notestine, C. (2003). Left hippocampal volume loss in Alzheimer’s disease is reflected in performance on odor identifi-cation: a structural MRI study. Journal of the International Neuropsychological Society, 9(3), 459–471.

Murphy, C., Schubert, C. R., Cruickshanks, K. J., Klein, B. E., Klein, R., & Non-dahl, D. M. (2002). Prevalence of olfactory impairment in older adults. JAMA, 288(18), 2307–2312.

Nagayama, S., Homma, R., & Imamura, F. (2014). Neuronal organization of olfacto-ry bulb circuits. Frontiers in Neural Circuits, 8, 98.

Nägga, K., Gottfries, J., Blennow, K., & Marcusson, J. (2002). Cerebrospinal Fluid Phospho-Tau, Total Tau and β-Amyloid1–42 in the Differentiation between Alzheimer’s Disease and Vascular Dementia. Dementia and Geriatric Cognitive Disorders, 14(4), 183–190.

Naveh-Benjamin, M., Hussain, Z., Guez, J., & Bar-On, M. (2003). Adult age differ-ences in episodic memory: further support for an associative-deficit hypothesis. Journal of Experimental Psychology: Learning, Memory, and Cognition, 29(5), 826.

141

Negoias, S., Croy, I., Gerber, J., Puschmann, S., Petrowski, K., Joraschky, P., & Hummel, T. (2010). Reduced olfactory bulb volume and olfactory sensitivity in patients with acute major depression. Neuroscience, 169(1), 415–421.

Nelson, P. T., Alafuzoff, I., Bigio, E. H., Bouras, C., Braak, H., Cairns, N. J., … Tredici, K. D. (2012). Correlation of Alzheimer disease neuropathologic chang-es with cognitive status: a review of the literature. Journal of Neuropathology & Experimental Neurology, 71(5), 362–381.

Nilsson, L.-G., Adolfsson, R., Bäckman, L., de Frias, C. M., Molander, B., & Nyberg, L. (2004). Betula: A prospective cohort study on memory, health and aging. Aging Neuropsychology and Cognition, 11(2–3), 134–148.

Nilsson, L. G., Bäckman, L., Erngrund, K., Nyberg, L., Adolfsson, R., Bucht, G., ... & Winblad, B. (1997). The Betula prospective cohort study: Memory, health, and aging. Aging, Neuropsychology, and Cognition, 4(1), 1–32.

Nordin, S., Almkvist, O., Berglund, B., & Wahlund, L.-O. (1997). Olfactory dys-function for pyridine and dementia progression in Alzheimer disease. Archives of Neurology, 54(8), 993–998.

Nordin, S., Brämerson, A., Liden, E., & Bende, M. (1998). The Scandinavian Odor-Identification Test: development, reliability, validity and normative data. Acta Oto-Laryngologica, 118(2), 226–234.

Nordin, S., Monsch, A. U., & Murphy, C. (1995). Unawareness of smell loss in normal aging and Alzheimer’s disease: discrepancy between self-reported and diagnosed smell sensitivity. The Journals of Gerontology Series B: Psychologi-cal Sciences and Social Sciences, 50(4), P187–P192.

Nyberg, L., Lövdén, M., Riklund, K., Lindenberger, U., & Bäckman, L. (2012). Memory aging and brain maintenance. Trends in Cognitive Sciences, 16(5), 292–305.

Nyberg, L., Maitland, S. B., Rönnlund, M., Bäckman, L., Dixon, R. A., Wahlin, Å., & Nilsson, L.-G. (2003). Selective adult age differences in an age-invariant mul-tifactor model of declarative memory. Psychology and Aging, 18(1), 149.

Nyberg, L., Mclntosh, A. R., Houle, S., Nilsson, L.-G., & Tulving, E. (1996). Acti-vation of medial temporal structures during episodic memory retrieval. Nature, 380(6576), 715.

O’Connor, D. W., Pollitt, P. A., Roth, M., Brook, C. P. B., & Reiss, B. B. (1990). Memory complaints and impairment in normal, depressed, and demented elderly persons identified in a community survey. Archives of General Psychiatry, 47(3), 224–227.

Olichney, J. M., Murphy, C., Hofstetter, C. R., Foster, K., Hansen, L. A., Thal, L. J., & Katzman, R. (2005). Anosmia is very common in the Lewy body variant of Alzheimer’s disease. Journal of Neurology, Neurosurgery & Psychiatry, 76(10), 1342–1347.

Olofsson, J. K., & Gottfried, J. A. (2015). The muted sense: neurocognitive limita-tions of olfactory language. Trends in Cognitive Sciences, 19(6), 314–321.

Olofsson, J. K., Hurley, R. S., Bowman, N. E., Bao, X., Mesulam, M.-M., & Gott-fried, J. A. (2014). A designated odor–language integration system in the human brain. Journal of Neuroscience, 34(45), 14864–14873.

Olofsson, J. K., Nordin, S., Wiens, S., Hedner, M., Nilsson, L.-G., & Larsson, M. (2010). Odor identification impairment in carriers of ApoE-ɛ4 is independent of clinical dementia. Neurobiology of Aging, 31(4), 567–577.

Olofsson, J. K., Rogalski, E., Harrison, T., Mesulam, M.-M., & Gottfried, J. A. (2013). A cortical pathway to olfactory naming: evidence from primary progres-sive aphasia. Brain, 136(4), 1245–1259.

142

Olofsson, J. K., Rönnlund, M., Nordin, S., Nyberg, L., Nilsson, L.-G., & Larsson, M. (2009). Odor identification deficit as a predictor of five-year global cognitive change: interactive effects with age and ApoE-ε4. Behavior Genetics, 39(5), 496–503.

Omar, R., Mahoney, C. J., Buckley, A. H., & Warren, J. D. (2013). Flavour identifi-cation in frontotemporal lobar degeneration. Journal of Neurology, Neurosur-gery & Psychiatry, 84(1), 88–93.

Orasji, S. S. S., Mulder, J. L., de Bruijn, S., & Wirtz, P. W. (2016). Olfactory dys-function in behavioral variant frontotemporal dementia. Clinical Neurology and Neurosurgery, 141, 106–110.

Pagano, S. F., Impagnatiello, F., Girelli, M., Cova, L., Grioni, E., Onofri, M., … Giombini, S. (2000). Isolation and characterization of neural stem cells from the adult human olfactory bulb. Stem Cells, 18(4), 295–300.

Paraskevas, G. P., Kapaki, E., Papageorgiou, S. G., Kalfakis, N., Andreadou, E., Zalonis, I., & Vassilopoulos, D. (2009). CSF biomarker profile and diagnostic value in vascular dementia. European Journal of Neurology, 16(2), 205–211.

Persson, J., Lind, J., Larsson, A., Ingvar, M., Cruts, M., Van Broeckhoven, C., … Nyberg, L. (2006). Altered brain white matter integrity in healthy carriers of the APOE ε4 allele A risk for AD? Neurology, 66(7), 1029–1033.

Peters, J. M., Hummel, T., Kratzsch, T., Lötsch, J., Skarke, C., & Frölich, L. (2003). Olfactory function in mild cognitive impairment and Alzheimer’s disease: an investigation using psychophysical and electrophysiological techniques. Ameri-can Journal of Psychiatry, 160(11), 1995–2002.

Petersen, R. C., Caracciolo, B., Brayne, C., Gauthier, S., Jelic, V., & Fratiglioni, L. (2014). Mild cognitive impairment: a concept in evolution. Journal of Internal Medicine, 275(3), 214–228.

Petersen, R. C., Roberts, R. O., Knopman, D. S., Boeve, B. F., Geda, Y. E., Ivnik, R. J., … Jack, C. R. (2009). Mild cognitive impairment: ten years later. Archives of Neurology, 66(12), 1447–1455.

Pievani, M., Rasser, P. E., Galluzzi, S., Benussi, L., Ghidoni, R., Sabattoli, F., … Frisoni, G. B. (2009). Mapping the effect of APOE ε4 on gray matter loss in Alzheimer’s disease in vivo. Neuroimage, 45(4), 1090–1098.

Pinching, A. J., & Powell, T. P. S. (1971). The neuron types of the glomerular layer of the olfactory bulb. Journal of Cell Science, 9(2), 305–345.

Pinto, J. M., Wroblewski, K. E., Kern, D. W., Schumm, L. P., & McClintock, M. K. (2014). Olfactory dysfunction predicts 5-year mortality in older adults. PloS One, 9(10), e107541.

Plailly, J., Delon-Martin, C., & Royet, J.-P. (2012). Experience induces functional reorganization in brain regions involved in odor imagery in perfumers. Human Brain Mapping, 33(1), 224–234.

Ponsen, M. M., Stoffers, D., Booij, J., van Eck-Smit, B. L., Wolters, E. C., & Ber-endse, H. W. (2004). Idiopathic hyposmia as a preclinical sign of Parkinson’s disease. Annals of Neurology, 56(2), 173–181.

Price, J. L., & Powell, T. P. S. (1970a). The mitral and short axon cells of the olfac-tory bulb. Journal of Cell Science, 7(3), 631–651.

Price, J. L., & Powell, T. S. (1970b). The synaptology of the granule cells of the olfactory bulb. Journal of Cell Science, 7(1), 125–155.

Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401.

Rahayel, S., Frasnelli, J., & Joubert, S. (2012). The effect of Alzheimer’s disease and Parkinson’s disease on olfaction: a meta-analysis. Behavioural Brain Re-search, 231(1), 60–74.

143

Reijs, B. L., Ramakers, I. H., Elias-Sonnenschein, L., Teunissen, C. E., Koel-Simmelink, M., Tsolaki, M., … Johannsen, P. (2017). Relation of Odor Identifi-cation with Alzheimer’s Disease Markers in Cerebrospinal Fluid and Cognition. Journal of Alzheimer’s Disease, 60(3), 1025–1034.

Ressler, K. J., Sullivan, S. L., & Buck, L. B. (1994). Information coding in the olfac-tory system: evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell, 79(7), 1245–1255.

Richardson, J. T., & Zucco, G. M. (1989). Cognition and olfaction: a review. Psychological Bulletin, 105(3), 352.

Rios, M., Antonio, M., Toral-Rios, D., Franco-Bocanegra, D., Villeda-Hernández, J., & Campos-Peña, V. (2013). Inflammatory process in Alzheimer’s Disease. Frontiers in Integrative Neuroscience, 7.

Ritchie, C., Smailagic, N., Noel Storr, A. H., Takwoingi, Y., Flicker, L., Mason, S. E., & McShane, R. (2014). Plasma and cerebrospinal fluid amyloid beta for the diagnosis of Alzheimer's disease dementia and other dementias in people with mild cognitive impairment (MCI). The Cochrane Library.

Roberts, R. O., Christianson, T. J., Kremers, W. K., Mielke, M. M., Machulda, M. M., Vassilaki, M., ... & Petersen, R. C. (2016). Association between olfactory dysfunction and amnestic mild cognitive impairment and Alzheimer disease dementia. JAMA neurology, 73(1), 93–101.

Román, G. C., Sachdev, P., Royall, D. R., Bullock, R. A., Orgogozo, J.-M., López-Pousa, S., … Wallin, A. (2004). Vascular cognitive disorder: a new diagnostic category updating vascular cognitive impairment and vascular dementia. Jour-nal of the Neurological Sciences, 226(1), 81–87.

Román, G. C., Tatemichi, T. K., Erkinjuntti, T., Cummings, J. L., Masdeu, J. C., Garcia, J. A., ... & Moody, D. M. (1993). Vascular dementia Diagnostic criteria for research studies: report of the NINDS AIREN International Workshop. Neu-rology, 43(2), 250-250.

Rönnlund, M., Sundström, A., Adolfsson, R., & Nilsson, L.-G. (2015). Subjective memory impairment in older adults predicts future dementia independent of baseline memory performance: Evidence from the Betula prospective cohort study. Alzheimer’s and Dementia, 11(11), 1385–1392.

Rönnlund, M., Nyberg, L., Bäckman, L., & Nilsson, L.-G. (2005). Stability, growth, and decline in adult life span development of declarative memory: cross-sectional and longitudinal data from a population-based study. Psychology and Aging, 20(1), 3.

Rouquier, S., Blancher, A., & Giorgi, D. (2000). The Olfactory Receptor Gene Rep-ertoire in Primates and Mouse: Evidence for Reduction of the Functional Frac-tion in Primates. Proceedings of the National Academy of Sciences of the United States of America, (6), 2870.

Royet, J. P., Plailly, J., Saive, A. L., Veyrac, A., & Delon-Martin, C. (2013). The impact of expertise in olfaction. Frontiers in psychology, 4, 928.

Royet, J.-P., Koenig, O., Paugam-Moisy, H., Puzenat, D., & Chasse, J.-L. (2004). Levels-of-processing effects on a task of olfactory naming. Perceptual and Mo-tor Skills, 98(1), 197–213.

Santos, D., Reiter, E., DiNardo, L., & Costanzo, R. (2004). Hazardous Events Asso-ciated with Impaired Olfactory Function. Archives of Otolaryngology - Head and Neck Surgery, 130(3), 317–319.

Schemper, T., Voss, S., & Cain, W. s. (1981). Odor identification in young and elderly persons: Sensory and cognitive limitations. Journals of Gerontology, 36(4), 446–452.

144

Schiffman, S., Zervakis, J., Westall, H., Graham, B., Metz, A., Bennett, J., & Heald, A. (2000). Effect of antimicrobial and anti-inflammatory medications on the sense of taste. Physiology and Behavior, 69(4–5), 413–424.

Schiffman, S. S., Moss, J., & Erickson, R. P. (1976). Thresholds of food odors in the elderly. Experimental Aging Research, 2(5), 389.

Schmand, B., Jonker, C., Hooijer, C., & Lindeboom, J. (1996). Subjective memory complaints may announce dementia. Neurology, 46(1), 121–125.

Schmand, B., Jonker, C., Geerlings, M. I., & Lindeboom, J. (1997). Subjective memory complaints in the elderly: depressive symptoms and future dementia. The British Journal of Psychiatry, 171(4), 373-376.

Schubert, C., Cruickshanks, K., Klein, B., Klein, R., & Nondahl, D. (2011). Olfacto-ry impairment in older adults: Five-year incidence and risk factors. Laryngo-scope, 121(4), 873–878.

Schubert, C., Carmichael, L., Klein, B., Klein, R., Cruickshanks, K., & Murphy, C. (2008). Olfaction and the 5-year incidence of cognitive impairment in an epi-demiological study of older adults. Journal of the American Geriatrics Society, 56(8), 1517–1521.

Schubert, C. R., Fischer, M. E., Pinto, A. A., Klein, B. E. K., Klein, R., Tweed, T. S., & Cruickshanks, K. J. (2017). Sensory Impairments and Risk of Mortality in Older Adults. Journals of Gerontology Series A-Biological Sciences and Medi-cal Sciences, 72(5), 710–715.

Segura, B., Baggio, H., Solana, E., Palacios, E., Vendrell, P., Bargalló, N., & Junqué, C. (2013). Neuroanatomical correlates of olfactory loss in normal aged subjects. Behavioural Brain Research, 246, 148–153.

Serby, M., Larson, P., & Kalkstein, D. (1991). The nature and course of olfactory deficits in Alzheimer’s disease. American Journal of Psychiatry, 148(3), 357–360.

Sergi, G., Bano, G., Pizzato, S., Veronese, N., & Manzato, E. (2017). Taste loss in the elderly: Possible implications for dietary habits. Critical Reviews in Food Science and Nutrition, 57(17), 3684–3689.

Seubert, J., Freiherr, J., Djordjevic, J., & Lundström, J. N. (2013). Statistical locali-zation of human olfactory cortex. NeuroImage, 66, 333–342.

Seubert, J., Laukka, E. J., Rizzuto, D., Hummel, T., Fratiglioni, L., Bäckman, L., & Larsson, M. (2017). Prevalence and Correlates of Olfactory Dysfunction in Old Age: A Population-Based Study. The Journals Of Gerontology. Series A, Bio-logical Sciences And Medical Sciences, 72(8), 1072–1079.

Sevush, S., & Leve, N. (1993). Denial of memory deficit in Alzheimer’s disease. The American Journal Of Psychiatry, 150(5), 748–751.

Shipley, M. T., & Ennis, M. (1996). Functional organization of olfactory system. Developmental Neurobiology, 30(1), 123–176.

Shu, C.-H., Hummel, T., Lee, P.-L., Chiu, C.-H., Lin, S.-H., & Yuan, B.-C. (2009). The proportion of self-rated olfactory dysfunction does not change across the life span. American Journal of Rhinology & Allergy, 23(4), 413–416.

Sicard, G., & Holley, A. (1984). Receptor cell responses to odorants: Similarities and differences among odorants. Brain Research, 292(2), 283–296.

Simons, D., Boot, W., Charness, N., Gathercole, S., Chabris, C., Hambrick, D., & Stine-Morrow, E. (2016). Do “Brain-Training” Programs Work? Psychological Science in the Public Interest, Supplement, 17(3), 103–186.

Smith, G., Ivnik, R., Petersen, R., Kokmen, E., Parisi, J., Tangalos, E., & Waring, S. (1996). Definition, course, and outcome of mild cognitive impairment. Aging, Neuropsychology, and Cognition, 3(2), 141–147.

145

Sobel, N., Johnson, B. N., Mainland, J., & Yousem, D. M. (2003). Functional Neu-roimaging of Human Olfaction. Neurological Disease and Therapy, 251.

Sperling, R. A., Aisen, P. S., Beckett, L. A., Bennett, D. A., Craft, S., Fagan, A. M., … Phelps, C. H. (2011). Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alz-heimer’s & Dementia: The Journal of the Alzheimer’s Association, 7(3), 280–292.

Stevens, J. C., & Cain, W. S. (1987). Old-age deficits in the sense of smell as gauged by thresholds, magnitude matching, and odor identification. Psychology and Aging, 2(1), 36–42.

Stevens, J. C., Cain, W. S., & Burke, R. J. (1988). Variability of olfactory thresh-olds. Chemical Senses, 13(4), 643–653.

Strauss, E. (1970). A Study on Olfactory Acuity. Annals of Otology Rhinology and Laryngology, 79(1), 95.

Striepens, N., Scheef, L., Wind, A., Meiberth, D., Popp, J., Spottke, A., … Jessen, F. (2011). Interaction effects of subjective memory impairment and ApoE4 geno-type on episodic memory and hippocampal volume. Psychological Medicine, 41(9), 1997–2006.

Sun, G. H., Raji, C. A., Maceachern, M. P., & Burke, J. F. (2012). Olfactory identi-fication testing as a predictor of the development of Alzheimer’s dementia: a systematic review. The Laryngoscope, 122(7), 1455–1462.

Sunderland, A., Watts, K., Baddeley, A. D., & Harris, J. E. (1986). Subjective memory assessment and test performance in elderly adults. Journal of Geron-tology, 41(3), 376–384.

Swan, G. E., & Carmelli, D. (2002). Impaired olfaction predicts cognitive decline in nondemented older adults. Neuroepidemiology, 21(2), 58–67.

Rouquier, S., Blancher, A., & Giorgi, D. (2000). The Olfactory Receptor Gene Rep-ertoire in Primates and Mouse: Evidence for Reduction of the Functional Frac-tion in Primates. Proceedings of the National Academy of Sciences of the United States of America, (6), 2870.

Tabert, M. H., Manly, J. J., Liu, X., Pelton, G. H., Rosenblum, S., Jacobs, M., … Devanand, D. P. (2006). Neuropsychological prediction of conversion to Alz-heimer disease in patients with mild cognitive impairment. Archives of General Psychiatry, 63(8), 916–924.

Tapiola, T., Alafuzoff, I., Herukka, S.-K., Parkkinen, L., Hartikainen, P., Soininen, H., & Pirttilä, T. (2009). Cerebrospinal fluid β-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Archives of Neu-rology, 66(3), 382–389.

Tay, T., Wang, J. J., Kifley, A., Lindley, R., Newall, P., & Mitchell, P. (2006). Sen-sory and cognitive association in older persons: findings from an older Australi-an population. Gerontology, 52(6), 386–394.

Taylor, H. R., Keeffe, J. E., Vu, H. T. V., Wang, J. J., Rochtchina, E., Pezzullo, M. L., & Mitchell, P. (2005). Vision loss in Australia. The Medical Journal of Aus-tralia, 182(11), 565–568.

Temmel, A. F. P., Quint, C., Schickinger-Fischer, B., Klimek, L., Stoller, E., & Hummel, T. (2002). Characteristics of olfactory disorders in relation to major causes of olfactory loss. Archives of Otolaryngology-Head & Neck Surgery, 128(6), 635–641.

Theofilas, P., Ehrenberg, A. J., Nguy, A., Thackrey, J. M., Dunlop, S., Mejia, M. B., … Grinberg, L. T. (2018). Probing the correlation of neuronal loss, neurofibril-

146

lary tangles, and cell death markers across the Alzheimer’s disease Braak stag-es: a quantitative study in humans. Neurobiology of Aging, 61, 1–12.

Thies, W., & Bleiler, L. (2013). Alzheimer’s disease facts and figures. Alzheimers Dement., 9(2), 208–45.

Tobiansky, R., Blizard, R., Livingston, G., & Mann, A. (1995). The Gospel Oak Study Stage-IV - the Clinical Relevance of Subjective Memory Impairment in Older-People. Psychological Medicine, 25(4), 779–786.

Toller, S. V. (1999). Assessing the impact of anosmia: review of a questionnaire’s findings. Chemical Senses, 24(6), 705–712.

Toussaint, N., de Roon, M., van Campen, J. P. C. M., Kremer, S., & Boesveldt, S. (2015). Loss of Olfactory Function and Nutritional Status in Vital Older Adults and Geriatric Patients. Chemical Senses, 40(3), 197–203.

Uhlmann, R., Larson, E., Rees, T., Koepsell, T., & Duckert, L. (1989). Relationship of Hearing Impairment to Dementia and Cognitive Dysfunction. Jama-Journal of the American Medical Association, 261(13), 1916–1919.

Valentijn, S. a. M., van Boxtel, M. P. J., van Hooren, S. a. H., Bosma, H., Beckers, H. J. M., Ponds, R., & Jolles, J. (2005). Change in sensory functioning predicts change in cognitive functioning: Results from a 6-year follow-up in the Maas-tricht Aging Study. Journal of the American Geriatrics Society, 53(3), 374–380.

Van der Flier, W. M., Pijnenburg, Y. a. L., Schoonenboom, N. M., Dik, M. G., Blankenstein, M. A., & Scheltens, P. (2008). Distribution of APOE genotypes in a memory clinic cohort. Dementia and Geriatric Cognitive Disorders, 25(5), 433–438.

Vargha-Khadem, F., Gadian, D. G., Watkins, K. E., Connelly, A., Van Paesschen, W., & Mishkin, M. (1997). Differential Effects of Early Hippocampal Pathology on Episodic and Semantic Memory. Science, 277(5324), 376.

Veldhuizen, M., Smeets, M. A. M., & Veldhuizen, M. G. (2009). Sense of Smell Disorder and Health-Related Quality of Life. Rehabilitation Psychology, 54(4), 404–412.

Vellas, B., Aisen, P. S., Sampaio, C., Carrillo, M., Scheltens, P., Scherrer, B., … Hampel, H. (2011). Prevention trials in Alzheimer’s disease: An EU-US task force report. Progress in Neurobiology, 95(4), 594–600.

Villemagne, V. L., Burnham, S., Bourgeat, P., Brown, B., Ellis, K. A., Salvado, O., … Australian Imaging Biomarkers and Lifestyle (AIBL) Research Group. (2013). Amyloid β deposition, neurodegeneration, and cognitive decline in spo-radic Alzheimer’s disease: a prospective cohort study. The Lancet. Neurology, 12(4), 357–367.

Vogel, A., Stokholm, J., Gade, A., Andersen, B. B., Hejl, A. M., & Waldemar, G. (2004). Awareness of deficits in mild cognitive impairment and Alzheimer’s disease: Do MCI patients have impaired insight? Dementia and Geriatric Cog-nitive Disorders, 17(3), 181–187.

Waldorff, F. B., Siersma, V., Vogel, A., & Waldemar, G. (2012). Subjective memory complaints in general practice predicts future dementia: a 4-year fol-low-up study. International Journal of Geriatric Psychiatry, 27(11), 1180–1188.

Ward, A., Crean, S., Mercaldi, C. J., Collins, J. M., Boyd, D., Cook, M. N., & Arrighi, H. M. (2012). Prevalence of Apolipoprotein E4 Genotype and Homo-zygotes (APOE e4/4) among Patients Diagnosed with Alzheimer’s Disease: A Systematic Review and Meta-Analysis. Neuroepidemiology, 38(1), 1–17.

Watabe-Rudolph, M., Begus-Nahrmann, Y., Lechel, A., Rolyan, H., Scheithauer, M.-O., Rettinger, G., … Rudolph, K. L. (2011). Telomere Shortening Impairs

147

Regeneration of the Olfactory Epithelium in Response to Injury but Not Under Homeostatic Conditions. PLOS ONE, 6(11), e27801.

Wehling, E., Nordin, S., Espeseth, T., Reinvang, I., & Lundervold, A. J. (2011). Unawareness of Olfactory Dysfunction and its Association with Cognitive Functioning in Middle Aged and Old Adults. Archives of Clinical Neuropsy-chology, 26(3), 260–269.

WHO | Dementia: a public health priority. (n.d.). Retrieved March 3, 2018, from http://www.who.int/mental_health/publications/dementia_report_2012/en/

Willander, J., & Larsson, M. (2006). Smell your way back to childhood: Autobio-graphical odor memory. Psychonomic Bulletin & Review, 13(2), 240–244.

Williams, S. S., Williams, J., Combrinck, M., Christie, S., Smith, A. D., & McShane, R. (2009). Olfactory impairment is more marked in patients with mild dementia with Lewy bodies than those with mild Alzheimer disease. Journal of Neurology Neurosurgery and Psychiatry, 80(6), 667–670.

Wilson, R. I., & Mainen, Z. F. (2006). Early events in olfactory processing. Annu. Rev. Neurosci., 29, 163–201.

Wilson, R., Barnes, L., Schneider, J., Bach, J., Bennett, D., Beckett, L., & Evans, D. (2002). Individual differences in rates of change in cognitive abilities of older persons. Psychology and Aging, 17(2), 179–193.

Wilson, R. S., Arnold, S. E., Schneider, J. A., Tang, Y., & Bennett, D. A. (2007). The relationship between cerebral Alzheimer’s disease pathology and odour identification in old age. Journal of Neurology Neurosurgery and Psychiatry, 78(1), 30–35.

Wilson, R. S., Arnold, S. E., Tang, Y., & Bennett, D. A. (2006). Odor identification and decline in different cognitive domains in old age. Neuroepidemiology, 26(2), 61–67.

Wilson, R. S., Yu, L., & Bennett, D. A. (2011). Odor Identification and Mortality in Old Age. Chemical Senses, 36(1), 63–67.

Winblad, B., Amouyel, P., Andrieu, S., Ballard, C., Brayne, C., Brodaty, H., … Zetterberg, H. (2016). Defeating Alzheimer’s disease and other dementias: a priority for European science and society. Lancet Neurology, 15(5), 455–532.

Winston, J. S., Gottfried, J. A., Kilner, J. M., & Dolan, R. J. (2005). Integrated neu-ral representations of odor intensity and affective valence in human amygdala. Journal of Neuroscience, 25(39), 8903–8907.

Wolk, D. A., Dickerson, B. C., & Alzheimer’s Disease Neuroimaging Initiative. (2010). Apolipoprotein E (APOE) genotype has dissociable effects on memory and attentional-executive network function in Alzheimer’s disease. Proceedings of the National Academy of Sciences of the United States of America, 107(22), 10256–10261.

Yaffe, K., Freimer, D., Chen, H., Asao, K., Rosso, A., Rubin, S., … Simonsick, E. (2017). Olfaction and risk of dementia in a biracial cohort of older adults. Neu-rology, 88(5), 456–462.

Zelano, C., Bensafi, M., Porter, J., Mainland, J., Johnson, B., Bremner, E., … Sobel, N. (2005). Attentional modulation in human primary olfactory cortex. Nature Neuroscience, 8(1), 114.

Zou, D.-J., Chesler, A., & Firestein, S. (2009). How the olfactory bulb got its glo-meruli: a just so story? Nature Reviews Neuroscience, 10(8), 611–618.

su.diva-portal.orgsu.diva-portal.org/smash/get/diva2:1198320/fulltext01.pdf · receptor cells do...

Documents