age differences in treatment response to a collaborative care intervention for anxiety ......
TRANSCRIPT
Julie Wetherell, Ph.D., Andrew Petkus, M.A.,
Murray Stein, M.D., M.P.H., Michelle Craske, Ph.D., Denise Chavira, Ph.D., Catherine Sherbourne, Ph.D.,
Greer Sullivan, M.D., Lin Liu, Ph.D., &
Peter Roy-Byrne, M.D.
Age Differences in Treatment Response to a Collaborative Care Intervention for Anxiety Disorders in Primary Care
Disclosures Source Grant/Research Support
Forest Laboratories, Inc. Past
National Institute of Mental Health Current
National Center for Complementary
and Alternative Medicine
Current
US Veterans Health Administration Current
Overview Anxiety disorders are the most common mental health
problem in older adults
Lead to poorer health and cognitive decline
Present in primary care; typically unrecognized and untreated
If treated, usually benzodiazepines
Benzodiazepines cause falls and cognitive decline in older people
CBT is considered very effective for anxiety in younger and middle-aged adults
Meta-analyses suggest that CBT doesn’t work as well in older individuals
Most research on Generalized Anxiety Disorder
Effect sizes from meta-analysis of CBT for anxiety in older adults
Gould RL et al. J Am Geriatr Soc 2012;60:218-29
N = 215/348 N = 164 N = 202 N = 172
Large
Medium
Small
Meta-analyses comparing psychotherapy and medication for geriatric depression and anxiety
Pinquart M et al. Am J Psychiatry 2006;163:1493-1501 Pinquart M, Duberstein JF. Am J Geriatr Psychiatry 2007;15:639-51
0.0
0.5
1.0
1.5
2.0
Depression Anxiety
Un
con
tro
lled
eff
ect
:
clin
icia
n-r
ate
d m
eas
ure
s
Psychotherapy
Medication
P < 0.001
89 studies N = 5,328
32 studies N = 2,484
Not significant
Response to treatment for late-life GAD
Conçalves DC, Byrne GJ. J Anx Disord 2012;26:1-11
Favors treatment
Favors control
Benzo vs. Placebo
Antidep vs. Placebo
CBT vs. Wait List
CBT vs. Usual Care
CBT vs. Active
Control
CBT vs. Other
Therapy
Effect sizes for CBT for GAD in younger and older adults
Covin et al. J Anx Disord 2008;22:108-16
Younger Adults Older Adults
Controlling Anxiety with Learning and Management (CALM) Study Effectiveness trial in primary care
Collaborative care intervention vs. treatment as usual
Algorithm-driven treatment with medications and computer-assisted CBT
Administered by care manager
4 anxiety disorders: generalized anxiety disorder, panic disorder, social anxiety disorder, posttraumatic stress disorder
18 month duration
Age 18-75
First study to allow for age comparisons within a single trial
The present study Compared response to the CALM intervention for younger
(age 18-59) and older (age 60-75) adults
Examined anxiety symptoms (BSI Anxiety) and depressive symptoms (PHQ) in the full sample
Examined generalized anxiety disorder symptoms among the subsample with GAD
Used linear mixed models to examine treatment by time interaction within each age group
Examined intervention/treatment as usual differences at each time point
Controlled for appropriate medication use
Demographics
Age 18-59 (n = 870)
Age 60-75 (n = 134)
Female 71% 71%
Latino 20% 16%
> 12 years of education* 79% 70%
Employed* 76% 40%
Number of chronic medical conditions* 2.1 (1.9) 3.7 (2.2)
Number of primary care visits 4.3 (4.2) 5.0 (5.8)
Number of medical specialist visits* 1.0 (2.4) 1.7 (2.8)
*p < .05
Baseline clinical variables
Age 18-59 (n = 870)
Age 60-74 (n = 134)
Brief Symptom Inventory (BSI) Anxiety T-score
65.2 (8.6) 65.7 (8.4)
Patient Health Questionnaire-9 (PHQ) 12.7 (6.3) 13.4 (6.5)
SF-12 Physical* 50.0 (11.2) 43.8 (11.6)
SF-12 Mental 31.7 (10.1) 32.9 (9.8)
Generalized Anxiety Disorder Symptom Scale (GADS)
13.9 (3.9) n = 655 with GAD
13.9 (3.7) n = 100 with GAD
*p < .001
3
5
7
9
11
Baseline 6 Months 12 Months 18 Months
BS
I A
nx
iety
Young TX
Young TAU
3
5
7
9
11
Baseline 6 Months 12 Months 18 Months
BS
I A
nx
iety
Old TX
Old TAU
BSI Anxiety
Treatment by time interactions significant in both age groups
Among younger adults, the CALM intervention was superior to treatment as usual on overall anxiety symptoms at all time points, after controlling for appropriate medication use
Among older adults, the CALM intervention was superior to treatment as usual on overall anxiety symptoms at 12 months but did not differ at 6 or 18 months
p < .01
Treatment x time F = 3.36, p = .02
p < .01
p < .01
p = .05
Treatment x time F = 7.45, p < .0001
PHQ Depression
Treatment by time interactions significant in both age groups
Among younger adults, the CALM intervention was superior to treatment as usual on depressive symptoms at all time points, after controlling for appropriate medication use
Among older adults, the CALM intervention was superior to treatment as usual on depressive symptoms at 12 months but did not differ at 6 or 18 months
5
7
9
11
13
15
Baseline 6 Months 12 Months 18 Months
PH
Q 8
Old TX
Old TAU
5
7
9
11
13
15
Baseline 6 Months 12 Months 18 Months
PH
Q 8
Young TXYoung TAU
Treatment x time F = 9.55, p < .0001
p < .01 p < .01
p < .01
Treatment x time F = 3.32, p = .02
p = .04
7
9
11
13
15
Baseline 6 Months 12 Months 18 Months
GA
DS
Old TX
Old TAU
7
9
11
13
15
Baseline 6 Months 12 Months 18 Months
GA
DS
Young TX
Young TAU
GAD Symptoms
Treatment by time interactions significant in younger but not older
Among the younger adults with generalized anxiety disorder, the CALM intervention was superior to treatment as usual on GAD symptoms at all time points after controlling for appropriate medication use.
Among the older adults, the CALM intervention was not superior to treatment as usual on GAD symptoms at any time point.
p < .01 p < .01
p < .01
Treatment x time F = 2.49, p = .06
Treatment x time F = 7.28, p < .0001
Limitations and conclusions Insufficient power to examine three-way
interactions (age by treatment by time)
Relatively small number of older adults
Unable to analyze medications and CBT separately
No active control condition
CALM intervention appears to work less well for older adults, particularly for individuals with GAD
Older adults may benefit from booster sessions at 18 months