care after discharge of preterm infants -...

Faculteit Geneeskunde en Gezondheidswetenschappen

Academiejaar 2014-2015

CARE AFTER DISCHARGE OF PRETERM INFANTS

IN SOUTHERN RWANDA

Masterproef voorgelegd tot het behalen van de graad van

Master in Management en Beleid in de Gezondheidszorg

Door Louise Cauwelier

Promotor: Prof. Dr. Jan Philippé

Co-promotor: Prof. Dr. Luk Cannoodt

Verklaring publicatie en vermogensrecht

Abstract English

Context: Neonatal care for preterm infants was upgraded in KABUTARE, the district

hospital of HUYE, with a medium care unit (August 2010) and a Kangaroo Mother

Care room (January 2011). No medical information exists after they were discharged

alive from that hospital.

Aims: 1) Investigate to what extent and how long KMC has been practiced in the

hospital and in the community; 2) Identify barriers to KMC in the community; and 3)

Describe contacts with health professionals after these infants arrived at home.

Method: The data were collected through a survey among 124 caregivers and their

CHWs. SPSS was used to analyze the received information.

Results: Almost all respondents practiced KMC, but there is a large variation in the time

KMC is practiced, before and after hospital discharge. KMC is practiced more during

daytime than at night. Working in the fields and lack of support to the caregivers are

reported most frequently as barriers to practice KMC in the community. Some CHWs

reported that they never visited the children assigned to them at home (11%), 26% did

less than the guidelines of five home-visits in the first four weeks after discharge, 50 %

did more than five.

Conclusions: KMC-practice is not yet optimal in Rwanda. Special attention should be

directed towards KMC-practice at night and the reported difficulties when practicing

KMC in the community. The majority of CHWs are highly motivated to visit infants at

home. Preterm infants would benefit from more involvement by the assigned health

centers after discharge.

Number of words: 13.943 (exclusive references and appendix)

Abstract Nederlands

Achtergrond: De neonatale zorg voor vroeggeborenen in KABUTARE, het district

ziekenhuis van HUYE, werd opgewaardeerd met een medium-care afdeling (augustus

2010) en aangevuld met een Kangoeroe-kamer (januari 2011). Eenmaal deze neonati

levend ontslagen worden, is er echter geen medische informatie meer beschikbaar.

Doel: 1) Onderzoeken hoe intensief en hoelang KMC werd toegepast in het ziekenhuis

en na ontslag; 2) Identificeren van hindernissen om deze zorg ambulant toe te passen;

en 3) Beschrijven van contacten met zorgverleners na ontslag uit het ziekenhuis.

Methode: Data werd verzameld via interviews met 124 moeders en hun CHWs. SPSS

werd gebruikt om de verzamelde gegevens te analyseren.

Resultaten: KMC wordt ook ambulant toegepast door bijna alle ondervraagden, maar

vertoont een grote variatie in de toepassing ervan, zowel voor als na ontslag uit het

ziekenhuis. KMC wordt meer overdag toegepast dan ’s nachts. Het werken op het veld

en het gebrek aan ondersteuning van de moeders werden gerapporteerd als de meest

voorkomende hindernissen. Sommige CHWs gaven aan nooit het kind thuis te hebben

bezocht (11%), 26% bezochten het kind minder dan de richtlijn van vijf bezoeken in de

eerste vier weken, 50% deed meer dan vijf thuisbezoeken.

Conclusie: KMC wordt nog niet optimaal toegepast in Rwanda. Bijzondere aandacht is

gewenst betreffende de toepassing van KMC ‘s nachts en de hindernissen om deze zorg

ambulant toe te passen. Het merendeel van de CHWs is zeer gemotiveerd om de

kinderen thuis te bezoeken. Prematuren zouden baat hebben bij een grotere

betrokkenheid van de aangewezen gezondheidszorgcentra na het ziekenhuisverblijf.

Aantal woorden masterproef: 13.943 (exclusief bijlagen en bibliografie)

Table of content

Verklaring publicatie en vermogensrecht .................................................................... 4

Abstract English ..................................................................................................................... 5 Abstract Nederlands ............................................................................................................. 6

Table of content ..................................................................................................................... 7 List of abbreviations ............................................................................................................. 9

Foreword ............................................................................................................................... 10

Introduction ......................................................................................................................... 11 Chapter 1: Review of the literature .............................................................................. 14 1.1. Search strategy ................................................................................................................. 14 1.2. Prematurity ....................................................................................................................... 14 1.2.1. What is prematurity? ............................................................................................................... 14 1.2.2. Causes ............................................................................................................................................ 15 1.2.2.1. Spontaneous preterm birth .......................................................................................................... 15 1.2.2.2. Provider-‐initiated preterm birth ............................................................................................... 16

1.2.3. Common problems ................................................................................................................... 16 1.2.4. Follow-‐up ...................................................................................................................................... 16

1.3. Epidemiology .................................................................................................................... 17 1.3.1. Worldwide ................................................................................................................................... 17 1.3.1.1. Prevalence ........................................................................................................................................... 17 1.3.1.2. Mortality ............................................................................................................................................... 18

1.3.2. Rwanda .......................................................................................................................................... 18 1.3.2.1. Incidence .............................................................................................................................................. 18 1.3.2.2. Mortality ............................................................................................................................................... 18

1.3.3. Belgium .......................................................................................................................................... 19 1.3.3.1. Incidence .............................................................................................................................................. 19 1.3.3.2. Mortality ............................................................................................................................................... 19

1.4. The Kangaroo Mother Care Method .......................................................................... 20 1.4.1. What is KMC? .............................................................................................................................. 20 1.4.2. Impact of KMC on the health status .................................................................................. 21 1.4.3. Effectiveness of KMC ............................................................................................................... 21 1.4.4. Requirements for KMC ........................................................................................................... 22 1.4.5. KMC barriers in Africa ............................................................................................................ 24 1.4.5.1. Overall intervention ........................................................................................................................ 24 1.4.5.2. Position ................................................................................................................................................. 24 1.4.5.3. Nutrition ............................................................................................................................................... 25 1.4.5.4. Early discharge and follow-‐up policies ................................................................................... 25 1.4.5.5. Duration ................................................................................................................................................ 26 1.4.5.6. Support .................................................................................................................................................. 26 1.4.5.7. Staff ......................................................................................................................................................... 26 1.4.5.8. Logistical obstacles .......................................................................................................................... 27

1.5. Health care in Rwanda ................................................................................................... 27 1.5.1. Brief outline of the country ................................................................................................... 27 1.5.2. Evolution of the health care system .................................................................................. 28 1.5.3. Organization of the public sector ....................................................................................... 29 1.5.3.1. Central level ........................................................................................................................................ 30 1.5.3.2. Intermediate level ............................................................................................................................ 30 1.5.3.3. Peripheral level ................................................................................................................................. 30

1.5.4. The community health workers (CHWs) ........................................................................ 31 1.5.5. Health insurance ....................................................................................................................... 32 1.5.6. Health indicators ....................................................................................................................... 32

Chapter 2: Problem Statement ....................................................................................... 34 Chapter 3: Methods ............................................................................................................ 35 3.1. Study design .......................................................................................................................... 35 3.2. Study population and recruiting .................................................................................... 35 3.3. Measuring instrument ....................................................................................................... 35 3.4. Data collection ..................................................................................................................... 36 3.4.1. Demographic data and socio-‐economic data .................................................................... 36 3.4.2. Kangaroo Mother Care ................................................................................................................ 36 3.4.3. Questions for the Community Health Worker .................................................................. 37 3.4.4. Other health care contacts ........................................................................................................ 37

3.5. Data analysis ......................................................................................................................... 37 3.6. Ethical considerations ....................................................................................................... 37

Chapter 4: Results .............................................................................................................. 39 4.1. Recruiting .............................................................................................................................. 39 4.2. Demographic and socio-‐economic data ...................................................................... 39 4.3. KMC .......................................................................................................................................... 42 4.3.1. KMC in the hospital ...................................................................................................................... 43 4.3.2. KMC in the community ................................................................................................................ 44

4.4. Home visits by Community Health Workers .............................................................. 47 4.5. KMC-‐practices observed by the CHWs ......................................................................... 50 4.5.1. Observations during the first week at home ..................................................................... 50 4.5.2. Observations during the first month at home .................................................................. 51

4.6. How to improve the practice of KMC in the community? ...................................... 53 4.7. Other health care contacts ............................................................................................... 55 4.8. Correlations .......................................................................................................................... 56

Chapter 5: Summary and Discussion ........................................................................... 59 5.1. Aim of this master’s thesis ............................................................................................... 59 5.2. Main findings ........................................................................................................................ 59 5.3. Findings confronted with the existing literature ..................................................... 62 5.4. Limitations ............................................................................................................................ 63

Chapter 6: Conclusions and recommendations ....................................................... 64

References ............................................................................................................................ 66 Appendix ................................................................................................................................... I

List of tables and figures ...................................................................................................... i

List of abbreviations

AIDS: Acquired Immune Deficiency Syndrome

ASM: Animatrice de Santé Maternelle

CHW: Community Health Worker

CHUB: University Teaching Hospital of BUTARE

FUAPI: Follow-Up Assessment of Premature Infants

GDP: Gross Domestic Product

GNI: Gross National Income

HC: Health Center

HIV: Human Immunodeficiency Virus

ICCM: Integrated Community Care Management

IMPORE: Improving Maternal and Pediatric Outcome: the Rwandan Experience

KMC: Kangaroo Mother Care

LBW: Low Birth Weight

MDG: Millennium Development Goal

MoH: Ministry of Health

NGO: Non-Governmental Organization

PNC: Postnatal Care

SPE: Study Center for Perinatal Epidemiology

SPSS: Statistical Package for the Social Science

TBA: Traditional Birth Attendants

UNDP: United Nations Development Programme

UNICEF: United Nations Children’s Fund

WHO: World Health Organization

Foreword

This master’s thesis was written as completion of my Master of Science in Health Care

Management and Policy. To celebrate this moment I would like to thank some people

who helped me throughout this accomplishment.

First, I would like to thank Mrs. Claudine MUKESHIMANA, for welcoming me into

the temporary shelter for motherless babies you started in Rwanda and for fulfilling this

African dream of mine.

Second, I would like to thank Prof. Dr. Jan PHILIPPE and Prof. Dr. Luk CANNOODT

for the guidance and suggestions you provided throughout the process of this

masterproof.

Third, I would like to thank Leonard NTWARI for the long days of translating

interviews written in Kinyarwanda and all the members of ‘Chez Marraine’ for your

hospitality.

Last but not least, I am grateful for my family, friends and everyone who helped me in

any way. Your encouragements meant a lot to me. Especially my parents deserve a

thank you since they gave me the opportunity to study.

Ghent, August 2015

Louise Cauwelier

11

Introduction

According to the World Health Organization (2013), 15 million babies are born preterm

every year, over 60 % occurs in Africa and South Asia (WHO, 2012). Preterm birth is

defined by the WHO as “babies born alive before 37 weeks of pregnancy or 259 days of

gestation” (WHO, 2013, p. 1). It is the leading cause of newborn deaths and the second-

leading cause of death in children under the age of five, after pneumonia (WHO, 2012).

Preterm birth often results in low birth weight (LBW). The WHO defines LBW as

“weight at birth of less than 2.500 grams” (UNICEF & WHO, 2004, p 1). It is also more

common in poor countries. Low birth weight is associated with increased mortality and

morbidity, inhibited growth, cognitive development and chronic diseases later in life. In

order to achieve the Millennium Development Goals (MDG) in 2015, the under-five

mortality rate should be reduced by two thirds compared to 1990 (United Nations,

2014).

The morbidity associated with preterm birth often extends to later life, resulting in

enormous physical, psychological and economic costs (Beck et al., 2010). Thus,

interventions that reduce infant morbidity and mortality and the resulting economic

costs related to it would be an important advance in health care.

A feasible, cost-effective solution can be found in Kangaroo Mother Care (KMC). It

provides an alternative to incubator care, without separation from the mother (Akhtar,

Haque & Khatoon, 2013). It is proven to reduce mortality, severe infections,

hypothermia and the length of hospital stay; moreover it increases weight gain and

breastfeeding (Conde-Agudelo, Belizan & Diaz-Rossello, 2011). The systematic

implementation of this method might be an important factor to reach the fourth

millennium development goal. However, the implementation of KMC is often varied

(Bergh et al., 2014). This kind of care is especially recommended in low- and lower

middle-income countries; these are respectively characterized by an economy with a

GNI per capita of $1,045 or less and economies with a GNI per capita of $1,046 to

$4,125 (The World Bank, 2015).

12

This study takes place in Rwanda, a low-income country located in Sub-Saharan Africa

with a GNI per capita of $630 (The World Bank, 2013). It is one of the first studies

conducted in Africa concerning the follow-up of preterm infants.

Since August 2010 a neonatal medium care unit was introduced in the district hospital

of HUYE, named KABUTARE. Three incubators were given to optimize the care.

Later, in January 2011 a KMC-ward was introduced. However, little is known about the

care of the preterm infants after discharge from the hospital. If indicated, a structured

follow-up program could be introduced to solve this.

Therefore, the following research questions were drafted: (1) “To what extent and how

long has the KMC-method been applied in the hospital and at home?” (2) “Which

barriers can be identified concerning this method?” and (3) “What contacts did the

prematurely born child had with the health care system after having been discharged for

the first time from the hospital to home?”

The data were collected through a quantitative survey with some open-ended questions.

The interviews were divided in two parts. The first part (A) is addressed towards the

caregiver (usually the mother) and the second part (B) towards the Community Health

Worker. SPSS was used to analyze the received information.

This report is divided in six chapters. The first chapter consists of a review of the

literature concerning prematurity, epidemiology, the Kangaroo Mother Care method and

the Rwandan health care system. The second chapter describes the research questions.

The research methodology is explained in the third chapter, followed by the results.

Chapter five discusses these findings. The last chapter presents the conclusions and

some recommendations for the future.

This study is part of the FUAPI-project: Follow-Up Assessment of Preterm Infants. It

has been initiated by Prof. Dr. Luk Cannoodt (Principal Investigator) and approved by

the Rwandan National Ethics Committee in May 2014. This part of the FUAPI-project

is financed by the Rwandan NGO UMUBANO-IMPORE (sponsored by a grant from

13

Dr. Louis Ide – project August 2010) and the Catholic University of Rwanda. IMPORE

stands for ‘Improving Maternal and Pediatric Outcome: the Rwandan Experience’. As

the name suggests, it was created to help reduce maternal and child mortality in

Rwanda, in order to help meet the fourth and fifth Millennium Development Goals. One

of their objectives is to formulate policy recommendations, such as the introduction of a

structured follow-up program for preterm infants, if indicated.

14

Chapter 1: Review of the literature

The aim of this review is to take a closer look at prematurity, its epidemiology, its

relation to the Kangaroo Mother Care Method and the organization of the Rwandan

health care system. First, the search strategy will be elucidated, followed by a

substantive discussion of the literature based on peer reviewed and grey literature.

1.1. Search strategy

A broad electronic search of the literature in Limo, PubMed and Science Direct was

performed. Keywords used for the search were: ‘prematurity’, ‘preterm birth’,

‘premature birth’, ‘epidemiology’, ‘Kangaroo Mother Care’, ‘Kangaroo Care’, ‘skin-to-

skin contact’, ‘barriers’, ‘obstacles’, ‘implementation’, ‘discharge’, ‘follow-up’,

‘development’, ‘developing countries’, ‘low-income countries’, ‘Rwanda’, ‘Africa’,

‘health care’, ‘CHW’, ‘insurance’ and ‘indicators’. Different combinations were used.

The results were filtered by title and abstract, and the snowball effect was used to

determine additional literature. No time limit was inserted.

1.2. Prematurity

1.2.1. What is prematurity?

Preterm birth is defined by the World Health Organization as “babies born alive before

37 weeks of pregnancy or 259 days of gestation” (WHO, 2013, p. 1). It is the leading

cause of newborn deaths and the second-leading cause of death in children under the

age of five, after pneumonia. Among preterm birth, three levels of risk can be

distinguished (see figure 1): extremely preterm infants (<28 weeks), very preterm

infants (28 to <32 weeks) and moderate to late preterm infants (32 to <37 weeks)

(WHO, 2012). Preterm birth is closely associated with low birth weight (LBW), but also

restricted fetal (intrauterine) growth can lead to LBW. The WHO defines LBW as

“weight at birth of less than 2500 grams” (UNICEF & WHO, 2004, p 1).

15

Figure 1: Overview of definitions applied for pregnancy outcomes related to preterm birth and

stillbirths (Blencowe, et al., 2012)

1.2.2. Causes

Preterm birth can happen due to a variety of reasons. Two subtypes can be

distinguished: spontaneous preterm birth and provider-initiated preterm birth (WHO,

2012).

1.2.2.1. Spontaneous preterm birth

Risk factors for spontaneous preterm birth can be found in factors related to the mother

such as a young or advanced maternal age, a short inter-pregnancy interval, a low

maternal body mass index, maternal psychological health and violence towards the

mother. Other risk factors are related to a multiple pregnancy. The risk of preterm birth

can also be increased by infections such as urinary tract infections, malaria, HIV,

syphilis and bacterial vaginosis. Some lifestyle factors can also have an impact such as

16

smoking, excessive alcohol consumption, drug use, stress, excessive physical work and

long times spent standing. Last, the genetics family history and cervical incompetence

can be an important contributor to preterm birth (WHO, 2012).

1.2.2.2. Provider-initiated preterm birth

Maternal as well as fetal conditions can form a medical indication for preterm birth. The

birth can be induced medically or a C-section can be performed. Clinical conditions

related to the mother are renal disease, hypertension, obesity and diabetes. Clinical

conditions related to the fetus are severe pre-eclampsia, placental abruption, uterine

rupture, cholestasis, fetal distress and fetal growth restriction with abnormal tests. In

some cases a provider-initiated preterm birth occurs without a medical indication

(WHO, 2012).

1.2.3. Common problems

Common issues related to prematurity are respiratory problems, brain damage due to

immaturity of the central nervous system, thermoregulation problems, patent ductus

arteriosus, hypoglycaemia, infections, neonatal hyperbilirubinemia, hypocalcemia and

problems concerning fluids, electrolytes and nutrition. The severity of the problems

increase as the duration of the pregnancy is shorter (Van den Brande, Heymans &

Monnens, 1998).

1.2.4. Follow-up

Babies born prematurely are at high risk for developmental problems. A structured

follow-up system provides services for the family of those children. Such a program

helps to reassure the parents, gives them developmental expertise, timely diagnosis and

referral to needed services. However not all countries are able to provide such a follow-

up system for their inhabitants. In most developing countries follow-up policies do not

exist. Yet, studies show that infants who are not able to attend follow-up programs are

likely to have higher rates of disabilities, lower cognitive skills and less access to

17

required intervention services (Ballantyne, Stevens, Guttmann, Willan & Rosenbaum.,

2013).

1.3. Epidemiology

1.3.1. Worldwide

1.3.1.1. Prevalence

About 13 million preterm babies are born worldwide each year. In 2005, 9.6% of all

births were preterm. Africa and Asia bear 85% of this burden or approximately 11

million preterm births. Europe and North America achieve about the same number,

around 0.5 million preterm births. Last, 0.9 million preterm births occurred in Latin

America and the Caribbean. However these data are only an estimate due to the lack of

systematic data.

The highest rates occurred in Africa and North America, followed by Asia and Latin

America (see table 1). The lowest rates of preterm birth occurred in Europe and Oceania

(Beck et al., 2010).

Table 1: Estimated preterm birth prevalence rates for each region in 2005 (Beck et al., 2010)

Region Preterm births

(x1000)

Preterm birth

rate (%)

95% confidence

intervals

World total 12,870 9.6 9.1 – 10.1

Africa 4,047 11.9 11.1 – 12.6

Asia 6,907 9.1 8.3 – 9.8

Europe 466 6.2 5.8 – 6.7

Latin America and the

Caribbean

933 8.1 7.5 – 8.8

North America 480 10.6 10.5 – 10.6

Oceania 20 6.4 6.3 – 6.6

18

1.3.1.2. Mortality

The largest numbers of child deaths were in the African region and in the Southeast

Asian region (see table 2). A lower proportion of neonatal deaths (aged 0-27 days)

occurred in the African region than in the Southeast Asian region. However, Africa has

the highest proportion of deaths in children aged 1-59 months due to malaria and AIDS.

In America, Europe and Asia, a high proportion of child deaths occurred during the

neonatal period ranging from 48% (0.137 million / 0.284 million) in America to 54%

(1.295 million / 2.390 million) in Southeast Asia (Black et al., 2010).

Table 2: Child deaths occurred in 2008 worldwide (Black et al., 2010)

Region Number of deaths of children U5 (x1,000,000)

Neonatal deaths in %

Neonatal deaths due to preterm birth complications in %

Africa 4.199 29 8

America 0.284 48 18

Eastern

Mediterranean

1.239 45 14

Europe 0.148 53 18

Southeast Asia 2.390 54 14

Western Pacific 0.534 52 15

1.3.2. Rwanda

1.3.2.1. Incidence

In 2010, the incidence of preterm births in Rwanda per 100 births was 9.5 (WHO,

2012).

1.3.2.2. Mortality

The major leading causes of death in neonatology are ‘Prematurity’ (43%), ‘Asphyxia’

(32%) and ‘Neonatal infections’ (10%) (see table 3). Together they are accountable for

85% of all causes of neonatal deaths in Rwanda in 2012 (Ministry of Health, 2012).

19

Table 3: Top causes of death in neonatology in all health facilities in 2012 (N = 2,722)

(Ministry of Health, 2012)

Rank Causes of neonatal death 2012 % of total

1 Prematurity (22 to 37 weeks) 1,157 43%

2 Asphyxia 861 32%

3 Neonatal infections 259 10%

4 Congenital malformation 159 6%

5 Respiratory infections 58 2%

6 Others 228 8%

1.3.3. Belgium

1.3.3.1. Incidence

In 2010, the incidence of preterm births in Belgium per 100 births was 7.9 (WHO,

2012).

Each year the Study Center for Perinatal Epidemiology (SPE) publishes an annual

report concerning the most important trends in birth in the Flemish region (see table 4).

The average frequency of preterm birth is 7.4%. In singleton pregnancies 6.5% of the

babies is born too soon, against 57.8% in multiple pregnancies (Cammu, Martens,

Martens, Van Mol & Jacquemyn, 2013).

Table 4: Gestational age (2013) (Cammu et al., 2013)

Singleton Pregnancy

(N=64,978)

Multiple Pregnancy

(N=1,219)

% Count % Count

<28 weeks 0.4 260 3.7 45

28-316/7 weeks 0.6 384 8.6 105

32-366/7 weeks 5.5 3,548 45.5 555

≥37 weeks 93.5 60,786 42.2 514

20

1.3.3.2. Mortality

The incidence of prenatal mortality has barely changed in the past five years. From the

infants born before 28 weeks, only half survives (see table 5). When the birth occurs

between 28 and 32 weeks, 90% survives. Once the baby is carried at term the mortality

rate lowers to 1/625 (Cammu et al., 2013).

Table 5: Perinatal mortality by gestational age (2013) in ‰ (Cammu et al., 2013)

Gestational Age Fetal Early-Neonatal Perinatal

<28 weeks 387 243 536

28-31 weeks 90 16 105

32-36 weeks 14 4 17

≥37 weeks 1.0 0.6 1.6

1.4. The Kangaroo Mother Care Method

1.4.1. What is KMC?

Professor Edgar Rey, a neonatologist from Colombia, first described the Kangaroo

Mother Care method in 1978. He was looking for a solution to meet the incubator

shortages, the high infection rates and the abandonment among preterm births (Charpak

& Ruiz-Pelaez, 2006).

The method has been reviewed and adjusted, resulting in a new model consisting of

three main components: position, feeding and discharge/follow-up policies (Charpak &

Ruiz-Pelaez, 2006). The position demands a continuous skin-to-skin contact between

infant and adult, while only wearing a diaper, socks and a cap. Furthermore, the

newborn must be hold in a vertical position until he can regulate his own temperature.

When possible, the child must exclusively be breastfed. Supplementations are allowed if

there is an inadequate weight gain (less than 15g/kg per day). Once medically stable,

used to the KMC-method and in control of the breathing-sucking-swallowing

coordination the infant can be discharged if an appropriate follow-up system is in place

and if the KMC-provider is willing to stick to the program.

21

Kangaroo Mother Care can be practiced continuously or intermittently. Continuous

KMC requires day and night skin-to-skin contact for at least 20 hours or more a day.

Intermittent KMC on the other hand, uses multiple sessions of skin-to-skin contacts a

day of at least 70 minutes per session (Bergh et al., 2014).

1.4.2. Impact of KMC on the health status

KMC has shown to be effective in different areas. The effects of Kangaroo Mother Care

can be divided in five main categories. We can identify physiological effects such as a

reduction in nosocomial infections and the stabilization of the heart rate, the respiratory

rate, oxygenation and thermoregulation. Second, we have behavioral effects such as

better sleep patterns and less crying of the child. Third, breastfeeding starts sooner, the

milk production is better and there is a longer duration of lactation. Fourth, KMC leads

to a reduction in maternal anxiety, improves maternal satisfaction and facilitates the

attachment between mother and child. Neurobehavioral effects can be seen in the

improved general, mental and motorical development of the infants receiving KMC

(Bergh, Charpak, Ezeonodo, Udani & van Rooyen, 2012).

1.4.3. Effectiveness of KMC

Measured by mortality, KMC is at least equivalent to conventional care in terms of

safety and thermal protection. The method facilitates breastfeeding and contributes to

better bonding between mother and child (WHO, 2003). Other studies reported that the

time taken for weight gain was significantly shorter in infants receiving KMC compared

to those with Conventional Care. The humane and psychosocial effects that accompany

KMC may explain this (Vahidi et al., 2014). Last, compelling evidence is found that

KMC is associated with a reduction in mortality, severe infections and sepsis, lower

respiratory tract disease, hypothermia and length of hospital stay. The method

contributes to an increase in weight gain, maternal satisfaction and mother-infant

attachment (Conde-Agudelo et al., 2011).

22

1.4.4. Requirements for KMC

The Kangaroo Mother Care method can be implemented in different settings, but mostly

in maternity facilities and referral hospitals. In order to achieve a good implementation,

health authorities at all levels need to be included and have to show support towards the

process. A national policy ensures education, training and a coherent and effective

integration of the practice within pre-existing structures of the health system. National

standards need to be developed with clear criteria for monitoring and evaluation. This

will help with the implementation of the local protocols. Not every health facility needs

to implement KMC in the same way. Policies and guidelines need to be adapted to the

local situation and culture and each protocol should cover follow-up. Involvement of the

staff can smoothen this process (WHO, 2003).

KMC does not involve more staff than conventional care, but training is necessary to

learn a new way of working. The program contains topics such as when and how to

initiate the KMC-method; how to position the baby between and during feeds; feedings

LBW and preterm infants; breastfeeding; alternative feeding methods until

breastfeeding becomes possible; involving the mother in all aspects of her baby’s care;

taking timely and appropriate action when a problem is detected or the mother is

concerned; deciding on the discharge; and ability to encourage and support the mother

and the family.

The mother needs to be informed about the advantages of the Kangaroo Mother Care

method as well as the advantages of the conventional method. Time must be given to

her so she can discuss the implications of KMC with her family. If she encounters any

obstacle, the mother must be able to talk about it with the staff towards finding

solutions. Health care workers also need to support the mother regardless which choices

she makes (WHO, 2003).

In order to make this all possible special KMC-rooms with comfortable beds and chairs

need to be arranged with attention towards privacy.

23

The mother can wear whatever she wants, as long as it is comfortable. There is only one

item needed: the support binder (see figure 2). The binder is a soft piece of fabric that

helps to hold the baby safely to the chest (WHO, 2003).

Figure 2: Carrying pouches for KMC babies (WHO, 2003)

After discharge, regular follow-up must be ensured. The quality of this follow-up

determines the timing of the discharge from the facility. The better the follow-up

system, the sooner the discharge can take place. The frequency of the home visits can

vary from daily in the beginning to weekly or monthly at the end. The WHO

recommends at least one visit for every preterm week. Mothers also need access to

health professionals for counseling and support related to the care of the baby. The

public health nurse should assess the home conditions, the home support and the ability

to travel at least one time. The set up of support groups leaded by mothers with previous

KMC experience can also be considered as help (WHO, 2003).

24

1.4.5. KMC barriers in Africa

The different barriers were divided in eight subcategories: Overall intervention,

position, nutrition, early discharge and follow-up policies, duration, support, staff and

logistical obstacles.

1.4.5.1. Overall intervention

The intervention is sometimes perceived as ‘a poor man’s alternative’. However once

the evidence is shown, people change their opinion (Bergh et al., 2012; Charpak &

Ruiz-Pelaez, 2006). Cultural and religious beliefs can also be an obstacle (Bergh et al.,

2014). Other concerns are fear because the baby is too small, fear because it is the first

time, the infant doesn’t look comfortable, fear of the KMC method, fear of not being

able to manage, difficulty of the intervention and fear of hurting the infant (Chisenga,

Chalanda & Ngwale, 2014). However, according to Nguah et al. (2011) 95.5% of the

mothers said KMC was beneficial to them, 96% thought KMC was beneficial to their

babies, 98% were willing to recommend KMC to other mothers and 61.9% thought it

was easy to practice. Age and literacy status of the mother and the birth weight of the

child were not significantly related to the use of KMC.

1.4.5.2. Position

The position of the infant during Kangaroo Mother Care provides most obstacles. In the

first place, direct skin-to-skin contact with the naked infant is considered as unusual or

improper. This problem exists only in cultures where physical expression of feelings is

not accepted. Secondly, mothers have inadequate privacy when they breastfeed or place

the infant in the adequate position. Thirdly, infants need to wear a cap and socks even in

warm climates. However, this is frequently not done due to the lack of knowledge of

neonatal physiology. Mothers perceive the position as capable of overheating the child

and inducing excessive sweating and therefore think KMC is dangerous. In the fourth

place, the use of diapers is problematic due to cultural, religious or economic reasons.

Traditional alternatives are often inappropriate when carrying the child on the chest.

25

Fifthly, continuous Kangaroo Mother Care is not practiced often enough in immature

infants resulting in poor growth, hypothermia and other complications. However an

intermittent position can be used in term or near-term infants. A sixth reason is that

some mothers find it overwhelming to provide continuous Kangaroo Mother Care. They

are also scared of losing their activity and independence during that time. Mothers have

trouble to imagine how to sleep, rest, work and take care of their own hygiene (Charpak

& Ruiz-Pelaez, 2006). Next, the KMC-practice can also interfere in the mother’s daily

work schedule (Bazzano et al., 2012; Charpak & Ruiz-Pelaez, 2006). Bazzano et al.

(2012) identified following obstacles: physical pain following labor during KMC, fear

of harming the baby through the cord stump, fear of causing internal pain to mother or

baby, the tradition of carrying baby on back in Ghana, the need for back support while

sitting or sleeping, fear of letting the baby fall and fear of rolling on the baby during

rest. Bergh et al. (2014) also mentioned the barrier of carrying the baby on the back in

African countries.

1.4.5.3. Nutrition

Charpak & Ruiz-Pelaez, (2006) reported three barriers concerning the feeding. As

previously mentioned, staff thinks training and supporting mothers in breastfeeding is

demanding in terms of skill, time and effort. Secondly, feasible alternatives to

supplement breastfeeding when indicated are not available. At last, families who are

able to afford more, think artificial formula feeding is better, because it is more

expensive. Luckily the worldwide efforts to promote breast milk are successful. In the

study of Nguah et al. (2011) 94.1% of the neonates were exclusively breastfed. Bazzano

et al. (2012) reported leaking of the breasts while applying KMC, which was

uncomfortable and caused reluctance towards trying different breastfeeding positions,

resulting in an interruption of Kangaroo Mother Care breastfeeding.

1.4.5.4. Early discharge and follow-up policies

In many developing countries follow-up policies do not exist, therefore it is found that

staff is reluctant to discharge infants even though the requirements for safe discharge

26

have been met (Bergh et al., 2014; Charpak & Ruiz-Pelaez, 2006). Record keeping and

the follow- up of the babies are one of the main challenges (Bergh et al., 2014).

1.4.5.5. Duration

Nguah et al. (2011) reported that 95.5% of the mothers had decided to continue KMC at

home, but only 71.8% outside their home. 93.1% were willing to practice KMC at

night.

1.4.5.6. Support

Bergh et al. (2012) state that the lack of support can be an obstacle. Sometimes

permission of husbands, partners and family to perform KMC is required. Also the

participation of the father can be difficult due to cultural barriers, the direct care of the

infant is actually still considered as the role of the mother (Charpak & Ruiz-Pelaez,

2006). However, Nguah et al. (2011) found that 87.1% of the mothers were confident

that their spouses would allow them to practice KMC at night but only 76.7% thought

their spouses or helpers at home would actually help them practicing KMC. Only 2.0%

thought members of the community would be supportive.

1.4.5.7. Staff

Frequently, health care staff has misconceptions concerning KMC, they think it will

represent extra work. However, once implemented it turns out to be a different way of

work and not additional work (Charpak & Ruiz-Pelaez, 2006). Moreover, most of the

workload is taken care of by the mother if she is well trained. According to Bergh et al.

(2012) staff turnover and rotations affect the continuity of KMC practice. Other barriers

are the reluctance to change existing arrangements related to space and additional costs.

Bergh et al. (2014) states that there is lack of clarity, which transpires during the

training, the trainers’ lack of knowledge, skills and experience. Furthermore, the

implementation in the workplace can be better. There is also a lack of supervision

27

because of staff workload, lack of transport, distances and decentralization. A last

problem is that the staff does not follow up recommendations of review meetings.

1.4.5.8. Logistical obstacles

Bergh et al. (2012) reported several possible obstacles such as roads, transport, space,

equipment, lack of communication mechanisms and technical problems such as power

failures and lack of printing facilities.

1.5. Health care in Rwanda

1.5.1. Brief outline of the country

Rwanda is a republic state located in Central to East Africa. It is bordered by Burundi,

the Democratic Republic of Congo, Tanzania and Uganda. This country is also known

as the land of a thousand hills, due to a geography characterized by several mountains,

volcanoes, lakes and rivers. Rwanda has a land area of 26,338 square kilometers and a

population of approximately 12 million inhabitants. The capital, Kigali, is also the

largest city of the country. Rwanda is a low-income country with a GNI per capita of

630$. The local language is Kinyarwanda, spoken by almost all Rwandans. Some also

speak French and/or English.

Apart from the Kigali area, Rwanda is composed of four provinces: Southern Province,

Western Province, Northern Province and Eastern Province. Each province is divided

into districts (see figure 3). The current study took place in the district of HUYE. The

district hospital named KABUTARE is located in the city of HUYE, previously called

BUTARE (Briggs & Booth, 2009).

28

Figure 3: Map of Rwanda (www.theiguides.org)

1.5.2. Evolution of the health care system

Since the genocide of 1994, Rwanda has made great progress. Improvements are

noticeable in living standards, infrastructure, education and health care. Today the

country has an economic growth rate of approximately 7%. This has helped in the

reduction of poverty, but income is unequally distributed with a distinct urban to rural

divide. Despite the efforts of the government and the progress made so far, many

Rwandans still live below the poverty line. Most of them live in rural areas (Pierce,

Heaton & Hoffmann, 2014).

In 2002, the government of Rwanda developed an ambitious plan called Vision 2020.

Three main goals can be identified: to quadruple the Gross Domestic Product (GDP),

achieve economic recovery and ending external aid by 2020. In terms of health care the

government wanted to improve the health-service delivery and increase equity of

access. This is aimed to be realized throughout three interventions: linkage of donor and

29

government programs, a nationwide community based health insurance and a

performance-based pay initiative to improve quality of health care (Pierce et al., 2014).

Chambers and Booth (2012) identified three common African obstacles regarding

Rwanda’s health care system.

In the first place, women and their families are often slow to seek medical help due to

suspicion or ignorance of modern health services, financial costs and weak incentives.

Especially traditional birth attendants (TBA) had a negative effect. TBA’s are untrained

women who are close to the women in the villages. As a solution, policy makers

incorporate those women into primary care system of Community Health workers

supervised by the nearest Health Center. So doing, the professional medical care

became more accessible to rural mothers.

Secondly, necessary and immediate emergency obstetric interventions are often delayed

due to the lack of ambulances and the lengthy time it takes to refer and transfer women

to another health facility. Now, the Rwandan government subsidizes 90% of the

ambulance cost through their community health insurance plan. As a result, more

women are able to arrive in a health facility on time to deliver in safer circumstances.

Last, the lack of equipment, drugs and blood supplies and the poorly trained staff affects

the quality of the care. Today, government policies have implemented training and

monitoring to ensure that health facilities respect their opening hours and deliver

qualitative care. Citizens are also encouraged to subscribe to the nationwide health

insurance program (Pierce et al., 2014).

1.5.3. Organization of the public sector

Three levels can be distinguished in the public sector. Each of them has a minimum

package of activities and communicates with the other levels to prevent overlap and

improve the use of the different services.

30

1.5.3.1. Central level

The Ministry of Health is based in the capital city and has five main responsibilities:

developing health policies and norms, establishing strategies and guidelines, conduct

monitoring and evaluation of the health situation, coordinate resources and managing

the national referral hospitals. These hospitals provide specialized inpatient and

outpatient services such ophthalmology and stomatology (Ministry of Health, 2011).

There are four national referral hospitals in Rwanda; one of them (CHUB) is located in

HUYE (Ministry of Health, 2012).

1.5.3.2. Intermediate level

The intermediate level helps the health centers with the implementation of the health

policies and norms developed by the central level. It also trains and supervises the staff

of the health centers, collects data and analyses and sends feedback. The district

hospitals need to manage all health problems for a well-defined population (Ministry of

Health, 2011). There are 40 district hospitals for inpatient and outpatient services in

Rwanda; one of them (KABUTARE) is located in the District of HUYE (Ministry of

Health, 2012).

1.5.3.3. Peripheral level

The peripheral level is an operational unit that works with representatives from the

community; it plans and provides primary health care (Ministry of Health, 2011). The

442 health centers take care of a defined population in a specific health area, seventeen

of them are located in the district of HUYE. They provide primary health care such as

prevention, but also inpatient services such as a maternity. Some health centers also

supervise health posts. There are 125 health posts that provide outreach activities such

as immunizations, antenatal care and family planning. These three activities are free of

charge (Ministry of Health, 2012).

31

1.5.4. The community health workers (CHWs)

Rwanda introduced the CHWs in 1995 after the genocide to improve access to health

services. They started with approximately 12,000 community health workers in 1995.

Today, the number has grown to 45,000. In each village, a male and a female CHW

called ‘binomes’ are in charge of the integrated community care management (ICCM).

One Animatrice de Santé Maternelle (ASM) is in charge of the maternal and newborn

health. The Community Health Desk of the Ministry of Health coordinates them all.

Qualifications to become elected by the village as a CHW include (1) the ability to read

and write, (2) aged between 20 and 50 years, (3) willing to volunteer, (4) living in the

local village and (5) being honest, reliable and trusted by the community (Ministry of

Health, 2013).

The binomes are in charge of the ICCM, malnutrition screening, community-based

provision of contraceptives, directly observed treatment for tuberculosis, prevention for

non-communicable diseases, prevention and behavior change activities and household

visits. They conduct household visits for children aged between six months and five

years. Children aged between 28 days and six months, can be visited by the binomes as

well as by the ASM.

The ASM is responsible for the follow-up of pregnant women and their newborns, the

malnutrition screening, community-based provision of contraceptives, prevention for

non-communicable diseases, preventive and behavior change activities and household

visits (Ministry of Health, 2013).Each pregnant woman should have three visits during

the pregnancy and three in the postnatal period. If the infant has a low birth weight, the

national guidelines schedule at least two additional home visits. The complete schedule

can be found in figure 4 (Rwanda PNC review, 2012). The ASM conducts household

visits from the date of conception until the baby is 28 days old. Children aged between

28 days and six months, can be visited by the binomes as well as by the ASM.

32

Figure 4: Home visit schedule of ASM (Rwanda PNC review, 2012)

1.5.5. Health insurance

The health facilities are mainly financed by the state, but individuals also need to

contribute through health insurance and out-of-pocket payments. The largest health

insurance agency is called “Mutuelles de Santé”, where an annual premium is asked per

family member, depending on the income level. This agency insures 91% of the

Rwandan population. For each visit to the health center a fee (10% of total costs) needs

to be paid out-of-pocket (Government of the Republic of Rwanda, 2014).

1.5.6. Health indicators

Since the genocide, Rwanda has come a long way. The country has achieved significant

improvements in different health indicators such as the under-five mortality rate, the

contraceptive prevalence rate and the childbirth conditions. The main reason is the rise

of the national budget for health care services. This allowed to increase the number of

staff, and to improve the national health supply and infrastructure. Between 2000 and

2007 a drop in the under-five mortality rate of nearly 50% has been registered.

Furthermore, the prevalence of the level of contraceptives has almost tripled between

33

2000 and 2008. Last, the proportion of women giving birth in a health facility has

increased drastically (Dhillon, Bonds, Fraden, Ndahiro & Ruxin, 2012). This all

resulted in a reduction of the maternal mortality ratio: from one of the world’s highest in

2005 at 750 deaths per 100,000 live births down to 325 in 2015 (UNDP, Rwanda,

2015). Despite those positive numbers, health issues associated with poverty still persist

(WHO, 2013).

34

Chapter 2: Problem Statement

In order to help reduce infant mortality and morbidity in the district of HUYE, neonatal

care in KABUTARE has been scaled up in the second half of 2010. In August 2010 a

neonatal medium care unit was introduced. Three incubators were donated to optimize

neonatal care for preterm and LBW newborns. Medical and nursing staff received extra

training in neonatal care and in the use of the KMC-method. In January 2011 a separate

KMC-ward was introduced.

There is no doubt that this has improved the survival chances of some of the smaller and

weaker newborns. However, what happens when these infants are discharged home

alive? CHWs are supposed to add two home visits in the first week after these LBW

infants have been discharged. Do they sometimes return to the hospital and if so, for

what reasons? Do they seek care in a health center regularly? The mothers are supposed

to continue the KMC-practice in the community. Do they do this? How frequently?

What are possible barriers to the KMC-practice at home and elsewhere? There is hardly

anything known about what happens in reality in relation to these issues. The aim of this

master’s thesis is to give some answers to these questions.

The answers to those questions could help policy makers with the implementation of a

structured follow-up program, if this could optimize the health status of this group of

children. It has been reported before that the absence of such a program is one of the

bigger health care challenges in developing countries (Bergh et al., 2014; Pratomo et al.,

2012; Charpak & Ruiz-Pelaez, 2006). The FUAPI-project (see further) aims to

contribute to taking on this challenge. This master proof attempts to increase knowledge

about certain aspects of the research questions introduced above through a survey of

caregivers (usually the mothers) of the study-population and the CHWs who know these

families best.

35

Chapter 3: Methods

3.1. Study design

This study is a quantitative study, namely a cross-sectional study, where interviews

were conducted to collect information about the characteristics of the preterm and LBW

infants who received neonatal care in KABUTARE and their caregivers. The interviews

have been designed to find some answers about the research questions formulated

above. The interviews were conducted in the home of the main respondent, usually the

mother of the each selected child. Four bachelors in Public Health conducted the

interviews in Kinyarwanda.

3.2. Study population and recruiting

The study includes 185 low birth weight infants, born between January 1st 2011 and

December 31th 2012, who received neonatal care in the district hospital KABUTARE

and who were known to be still alive at the time the interviews took place.

Inclusion criteria were: (1) low birth-weight infant (< 2500 gram), (2) born alive, (3)

between 01/01/2011 and 31/12/2012, (4) received neonatal care in KABUTARE, (5)

discharged from the hospital alive (6) discharged to the home of the child.

Exclusion criteria were: (1) infants with a birth weight of 2500 gram or above, (2)

infants born before 2011 or after 2012, (3) infants who were transferred to the

University Teaching Hospital of BUTARE (CHUB) to receive neonatal intensive care,

(5) infants who died in KABUTARE before discharge.

3.3. Measuring instrument A survey questionnaire was developed in English through collaboration between Prof.

Dr. Luk Cannoodt and the Department of Public Health, Catholic University of

Rwanda. The questionnaire was then translated in the local language, Kinyarwanda. In

total, 124 families could be reached. The survey was conducted by four interviewers, all

36

bachelors in Public Health, who recently graduated at the Catholic University of

Rwanda. The questionnaires are divided into four sections (Appendix 1). Before

conducting the interviews, an informed consent was signed by the respondents (section

0 see Appendix 2). Next, in section 1, background information on child and household

were collected. In section 2, background information about the interview was collected.

Finally, section 3 collected information from the CHWs who know the families best and

were present during the interview in the home of the main respondent. All answers were

coded so statistical data analysis with SPSS Statistics 22 could be made possible.

3.4. Data collection

3.4.1. Demographic data and socio-economic data

Demographic and socio-economic data were collected in relation to both the main

respondent and the child. The interviewer asked who the main respondent was, in which

district the mother lives, the age of the main respondent, the marital status of the main

respondent, the level of education of the main respondent, the profession of the main

respondent and whether the main respondent had a phone (the lowest income-classes

usually have no phone). If the mother was not present during the interview, the

interviewer asked if she was still alive and, if not, when she passed away. Questions

related to the child were the date of birth, the gender and if the child was part of a twin.

The date of birth has been asked to verify whether the in- and exclusion criteria were

met. Last, the insurance status was also added in the questionnaire.

3.4.2. Kangaroo Mother Care

Data concerning the KMC-method was collected through the main respondent as well

as through the community health worker. The answers of the main respondent were

further divided into the use of KMC in the hospital and at home, also a subdivision was

made concerning the use of KMC during the day and during the night. Mothers were

asked if they received any information regarding KMC while in the hospital and, if so,

when this happened and by whom, whether they started KMC in the hospital and when,

37

whether they stayed in the KMC-room and, if so, how many days. Questions related to

the KMC-practice in the community were regarding the continuation of KMC after

discharge from the hospital and the duration of it, the average number of hours practiced

during the day, during the night and the number of nights, the difficulty to continue

KMC in the community and the experienced obstacles. Last, the main respondent could

formulate suggestions that would facilitate the KMC-practice in the community.

3.4.3. Questions for the Community Health Worker

First, the community health worker was asked how many follow-up visits she

conducted. Next, questions were asked about how often KMC practiced by the mother

was observed. Finally, they were encouraged to make suggestions about facilitating the

KMC-practice in the community.

3.4.4. Other health care contacts

The main respondents were asked what health centers they usually visited when the

child was sick or needed vaccinations and whether the child had to return to a hospital

after the first discharge and why.

3.5. Data analysis The analysis of the data was performed through SPSS Statistics 22. Each variable was

thoroughly checked by running the frequencies. All indistinct data was reported as

‘missing’ or through the code ‘9999’. All qualitative data was translated into

quantitative data to make more statistical tests possible.

3.6. Ethical considerations

To perform this research the approval from the Rwanda National Ethics Committee was

required. The approval came in May 2014 (see Appendix 3). All the interviews were

conducted after signing an informed consent (see Appendix 2). The gathered data are

treated in an anonymously way. Authorization was also requested and granted from

38

KABUTARE, the health centers, the head of the community health workers of each

sector and local authorities. As the author of this work did not directly perform

interviews or communicate with the mothers or children, no approval of the Medical

Committee of the University Hospital of Ghent was asked nor needed.

39

Chapter 4: Results

4.1. Recruiting The database included 185 LBW children. Of these, 30 children died after they were

already discharged home from KABUTARE. Thirty one families could not be reached,

either because they moved outside the area or because the address in the hospital files

was incomplete or incorrect (they were unknown in the recorded village). There were

124 children that could be reached for the interviews. Although, the target population is

the child, it is the mother or another close member to the child who was interviewed

(the one who took most care of the child). All caregivers contacted (= main

respondents) were willing to participate and signed the informed consent form.

4.2. Demographic and socio-economic data Table 6, 7 and 8 describe some demographic and socio-economic data of the sample.

For each variable, the exact values as well as the percentages are being presented. Table

6 presents the data related to the main respondents.

Table 6: Demographic data concerning the main respondents (N=124)

Variables N (%) Cumulative %

What is the relationship of each main respondent towards his/her selected child?

• Mother 114 (91.9)

• Grandmother 7 (5.6)

• Other 3 (2.4)

District where the households live:

• Huye

• Nyaruguru

• Gisagara

101 (81.5)

17 (13.7)

6 (4.8)

Age of the main respondents:

• 25 and younger

• 25 until 29 years


20 (16.1)

38 (30.6)

27 (21.8)

16.1

46.7

68.5

40



• 45 and older

17 (13.7)

11 (8.9)

11 (8.9)

82.2

91.1

100.0

Marital status of the main respondents:

• Married

• Unmarried

• Separated

• Widow/widower

89 (71.8)

16 (12.9)

14 (11.3)

5 (4.0)

Level of education of the main respondents:

• None to two years of primary school

• Three to six years of primary school

• At least one year of secondary school

31 (25.0)

84 (67.7)

9 (7.3)

Profession main respondents:

• Farmer/Cultivator/Planter

• Casual jobs

• None

• Business dealer

111 (89.5)

7 (5.6)

4 (3.2)

2 (1.6)

Table 6 shows that in all but ten cases, the main respondent was the mother of the child

(92%). Most of the respondents (81.5%) live in the district of HUYE, where

KABUTARE is located, but it also shows that the district hospital attracts a significant

number of preterm newborns coming from neighboring districts (18.5%). Almost 47%

of the main respondents was 29 years or younger at the time of the interview; the large

majority was married (71.8%) at that time. Almost 90% of the main respondents is

either farmer, cultivator or planter. 67.7% finished the first three years of primary

school and one, two or three additional years. One fourth of the respondents had either

no education, or one or two years of education. Less than half of the main respondents

(42%) in this rural area own a phone. This gives an indication of the socio-economic

class of the family.

Three of the 124 children selected and reached have lost their mother. One mother died

during birth, the second mother past away two months after discharge from the hospital

41

and in the third case the mother died one year after her child (of the study population)

left the hospital.

Table 7 presents some characteristics about the children in the sample.

Table 7: Demographic data concerning the child (N=124)

Variable N (%)

Gender of the child:

• Male

• Female

64 (51.6)

60 (48.4)

Twin child:

• Yes

• No

42 (33.9)

82 (66.1)

Year of birth

• In 2011

• In 2012

56 (45.2)

68 (54.8)

Table 7 shows an almost even distribution between the two genders. One third of the

children in the sample was born with a twin sister or brother. Of those 42 children, there

are 28 children where both twins were still alive at the time of the interview, and there

were 14 children where the other child of the twin died before the interview was

conducted (either before or after having been discharged from the hospital). Of the 124

children in the sample, 56 were born in 2011 en 68 in 2012.

Table 8 gives some information about the health insurance coverage. In Rwanda, health

insurance is mandatory. In rural areas people are usually self-employed. Each year they

have to pay a premium to be in the system called “Mutuelle de Santé” set up by the

Rwandan government. Health insurance coverage is free for the poorest as the

government pays their annual premium. Table 8 shows that 85 % of the family declared

having health insurance in the period that their child was less than one year old. For

about 10 % of the sample, the government paid the annual premium. Note that maternal

and neonatal care is free of charge for those who have health insurance (no out-of-

pocket payments) if the pregnant mother had four prenatal consultations in the health

42

center. Despite these benefits, 15 % declared that their household did not have health

insurance during the period that the child was less than one year old (see table 8).

Table 8: Health insurance coverage

Variable N (%)

Household health insurance when the child was < 1 year: (N=124)

• Mutuelle de santé

• None

105 (84.7)

19 (15.3)

If insured, who paid the annual premium? (N=105)

• The family paid the annual premium

• The premium was paid by the government

93 (88.6)

12 (11.4)

Current health insurance of the household: (N=124)

• Mutuelle de santé

• None

• RAMA

72 (58.1)

51 (41.1)

1 (0.8)

If insured, who paid the annual premium? (N=73)

• The family paid the annual premium

• The premium was paid by the government

• Their employer paid the annual premium

62 (84.9)

10 (13.7)

1 (1.4)

Table 8 also reports the answers concerning the health insurance coverage at the

moment of the interview. At that time, the children of the sample were between 2.5 and

4 years old. Only 50% of the sample was then in the system of “Mutuelle de Santé”, ten

families received free health insurance coverage by the government. For one family, the

employer paid the annual premium.

4.3. KMC

In KABUTARE, mothers with preterm newborns (born there or transferred to this

hospital after birth) usually receive information about the KMC-method. In order to get

to know the method and feel comfortable using it, they stay in the KMC-room for one

or more days. Once the infant weights at least 2 kg, mother and child can be discharged.

However, if the mother has a lot of work or has other children at home to take care of,

discharge can be arranged at 1.8 kg.

43

4.3.1. KMC in the hospital

Most main respondents (102 cases or 82.3%) stated that they received information about

the KMC-method in the hospital. Of those, 100 respondents also explained when they

received that information (see table 9).

Table 9: Moment KMC-information was received (N=100)

Variable N (%) Cumulative %

• Within one hour after delivery

• Between 1 and 12 hours after delivery

• Between 12 hours and one day after delivery

• Between 1 and 3 days after delivery



• More than 14 days after delivery

10 (10.0)

18 (18.0)

21 (21.0)

14 (14.0)

21 (21.0)

7 (7.0)

9 (9.0)

10.0

28.0

49.0

63.0

84.0

91.0

100.0

Table 9 shows that there is a large variation in the timing of receiving this information,

going from less than one hour after delivery (10 %) to more than two weeks after

delivery (9 %). Of those who remembered who gave them the information, 72.1%

responded that it was a medical doctor. About 91% of the interviewed (113 cases)

confirmed that the KMC-practice started in the hospital. In 6 cases of those 113 it was

not the mother who practiced KMC in the hospital (see table 10).

Table 10: KMC-practice in the hospital (N=113)

Variable N (%)

Person who has been practicing KMC:

• Mother

• Father

• Grandmother

• Aunt

107 (94.7)

3 (2.6)

2 (1.8)

1 (0.9)

There is also a large variation in the time that the persons concerned started using KMC

in the hospital. Some already started within one hour after delivery (18.6 %), while a

small number (7.1 %) started only one month after delivery (see table 11).

44

Table 11: Moment that KMC started in the hospital (N=113)


• Within one hour after delivery

• Between 1 and 12 hours after delivery

• Between 12 hours and one day after delivery




• After more than one month after delivery

21 (18.6)

19 (16.8)

15 (13.3)

17 (15.0)

20 (17.7)

13 (11.5)

8 (7.1)

18.6

35.4

48.7

63.7

81.4

92.9

100.0

Not all who practiced KMC in the hospital did so in the KMC room. In fact, only 96 of

the main respondents (77.4 %) stayed at least one day in the KMC-room. This is 85% of

those who started practicing KMC in the hospital. Of those who stayed in the KMC-

room, 32.3% left the room within one week, while 27.1% stayed there one month or

more (see table 12).

Table 12: Number of days in the KMC-room (N=96)


• Between 1 and 7 days


• Between 14 days and 21 days


• Between 1 and 2 months

• More than 2 months

31 (32.3)

13 (13.5)

10 (10.4)

16 (16.7)

23 (24.0)

3 (3.1)

32.3

45.8

56.2

72.9

96.9

100.0

4.3.2. KMC in the community

Only 6 of the main respondents, who reported that they started KMC in the hospital, did

not continue KMC after discharge from the hospital. In total, 98 interviewed persons

reported to have continued KMC at least one day in the community (= after discharge).

Of them, 82 (= 83.7 %) did so during at least one night. So, there is some difference

between the KMC-practice during daylight and at night. Table 13 shows how many

days KMC has been practiced during daylight, while table 14 presents the number of

hours per day KMC has been practiced after discharge during daylight.

45

Table 13: Duration of KMC-practice in the community during daylight (N=98)



• Between 7 and 14days





8 (8.2)

10 (10.2)

3 (3.1)

35 (35.7)

25 (25.5)

17 (17.3)

8.2

18.4

21.5

57.2

82.7

100.0

Table 14: Number of hours per day of KMC-practice during daylight (N=96)


• Between 1 and 3 hours



• More than 9 hours

41 (42.7)

29 (30.2)

13 (13.5)

13 (13.5)

42.7

72.9

86.4

100.0

Of those who practiced KMC in the community during daylight, only 8.2 % stopped

after one week, while 42.8 % continued during one month or more (see table 13).

Of those who practiced KMC in the community during daylight, 42.7 % did so only 1 to

3 hours per day, while 27% did so for more than 6 hours per day (see table 14).

Table 15 and 16 present the results of KMC-practice at night.

Table 15: Duration of KMC-practice at home during the night (N=80)


• Between 1 and 7 nights






19 (23.8)

6 (7.5)

5 (6.3)

27 (33.7)

14 (17.5)

9 (11.2)

23.8

31.3

37.6

71.3

88.8

100.0

46

Of those who practiced KMC at home during the night (82 of the interviewed), all but 2

gave the duration that they continued doing so. Among those, 23.8 % stopped after one

week, while 28.7 % continued during one month or more (see table 15).

Of those who continued KMC at night, almost half did so only 1 to 3 hours per night,

while more than 1/4 did so the whole night (more than 9 hours a night) (see table 16).

Table 16: Number of hours of KMC-practice at home during the night (N=81)





• More than 9 hours

40 (49.4)

14 (17.3)

6 (7.4)

21 (25.9)

49.4

66.7

74.1

100.0

Table 17 gives more clarification of why KMC is not practiced more frequently in the

community. In fact, 72 % of the interviewed indicated that they did experience

obstacles.

Table17: Obstacles in practicing KMC reported by the main respondents

Variable N (%)

Was it difficult to continue KMC at home? (N=100)

• Yes

• No

72 (72.0)

28 (28.0)

Note: Several respondents mentioned more than one obstacle. N gives the number of

respondents who mentioned the specified types of obstacles. The next column calculates what

% this is of the number of respondents who said that it was difficult to continue KMC at home

(= 72 respondents). The last column calculates the percentage of N in relation to the 124

respondents of the survey.

N % of ‘yes’ % of total sample

• Difficult to work (in the fields) with KMC 51 70.8 41.1

• Difficult getting food 38 52.8 30.6

• Lack of equipment such as warm clothes, a mattress,.. 31 43.1 25.0

• Health problems such as HIV, back pain… 18 25.0 14.5

• Having twins 14 19.4 11.3

47

• Difficulties to carry the baby on the chest 10 13.9 8.1

• Difficulty to sleep 9 12.5 7.3

• Disagreements with the partner 8 11.1 6.5

• Having a disability 2 2.8 1.6

Each respondent could give one or more obstacles/difficulties. Of those who did

mention at least one barrier/obstacle (72 cases), more than 70 % mentioned that it is

difficult to work when practicing KMC (most of them specified ‘working in the fields’),

almost 53 % mentioned that it is difficult to get food, 43 % mentioned lack of

equipment, and one fourth mentioned health problems (see table 17 for a complete

overview of barriers). The frequency can also be expressed in percentage of all

caregivers interviewed. ‘Difficult to work’ was mentioned by 41 % of all 124 main

respondents, difficult getting food was mentioned by 31 % and lack of equipment 25 %

of all main respondents. In 14 cases, the main respondent reported “having twins” as a

barrier. Of those, 12 had both twins alive at the moment of the interview. Two other

mothers with both twins alive did not report this as a barrier. Fourteen mothers have lost

one of both twins before the interview took place. Only one of them reported that

having twins was a barrier for practicing KMC. In one case the barrier ‘twins’ was

mentioned by a mother with a singleton child in the sample.

4.4. Home visits by Community Health Workers

Currently, CHWs play a key role in the follow-up of infants after discharge in Rwanda.

For the preterm infants, they are supposed to visit the families five times during the first

4 weeks at home, of which three times during the first week. This is why it was decided

to also ask some questions to those CHWs. In all cases, the respondent was the CHW in

charge of maternal and child health who knows the family best. She is living in the

same village (Umudugudu) as the family involved and is normally the one who does the

home visits during pregnancy and after delivery. For each child that could be reached,

the CHW involved was interviewed. Some of them may have more than one child of the

study population under her supervision.

48

Table 18 presents the frequency of home visits during the first week after being

discharged from the hospital, as reported by the CHWs of 122 children.

Table 18: Frequency of the home-visits conducted by the CHW during the first week after

discharge (N=122)

Number of CHWs Absolute Percentage Cumulative

percentage

No visits 14 11.5 11.5

1 visit 13 10.7 22.2

2 visits 44 36.1 58.3

3 visits 35 28.7 87.0

4 visits 4 3.3 90.3

5 visits 4 3.3 93.6

6 visits 2 1.6 95.2

7 visits 6 4.9 100.0

Surprisingly, 14 CHWs (or more than 1 out of ten interviewed) admitted that they

visited the family not even once during the first week that the LBW child arrived at

home. More than 1 out of 3 did so twice the first week (this is the recommended

frequency for normal births).

Almost 60 % did not reach the recommended three visits in the first week. On average,

the number of home visits during the first week was 2.4 times.

On the other side, 16 CHWs (13 %) visited the preterm infant more often than what is

recommended by the Ministry of Health (MoH). Some visited them daily, not only in

the first week, but in the following weeks as well. As could be expected the frequency

of visits reduces as the number of weeks that the LBW infant is at home increases (see

table 19).

Table 19: Frequency of home-visits by the CHW, week by week. (N=122)

Week 1 Week 2 Week 3 Week 4 After week 4

No visits 14 (11.5) 19 (15.6) 33 (27.1) 37 (30.3) 36 (29.5)

1 visit 13 (10.7) 46 (37.7) 51 (41.8) 57 (46.7) 53 (43.4)

2 visits 44 (36.1) 37 (30.3) 27 (22.1) 15 (12.3) 15 (12.3)

49

3 visits 35 (28.7) 9 (7.4) 2 (1.6) 2 (1.6) 6 (4.9)

4 visits 4 (3.3) 4 (3.3) 2 (1.6) 4 (3.3) 5 (4.1)

5 visits 4 (3.3) 2 (1.6) 2 (1.6) 0 (0.0) 4 (3.3)

6 visits 2 (1.6) 0 (0.0) 0 (0.0) 2 (1.6) 0 (0.0)

7 visits 6 (4.9) 5 (4.1) 5 (4.1) 5 (4.1) 3 (2.5)

Table 20 presents the results of the cumulative number of home visits for each CHW

over a period of four weeks. For LBW infants the Rwandan MoH recommends that this

be at least five times over the period of four weeks.

Table 20: Number of home visits per CHW from week 1 to week 4 (N=122)

Week 1 to 4 Absolute percentage Cumulative percentage

No visits 14 11.5 11.5

1 to 4 visits 18 14.8 26.3

5 visits 25 20.5 46.8

6 visits 25 20.5 67.3

7 to 10 visits 25 20.5 87.8

11 to 19 visits 10 8.2 95.9

28 visits 5 4.1 100.0

Table 20 shows that the same 14 CHWs who did not visit the family during the first

week, did not visit these families in the following weeks either. Another 15 % of the

CHWs visited the families of LBW infants in their village less than the recommended

five times over a period of four weeks. About 20 % of the CHWs visited them just the

five times as recommended. More than 50 % of the CHWs visited their assigned

families with LBW infants more than the recommended five times in the first four

weeks (average = 6.7 times in four weeks). Five CHWs reported that they visited the

families concerned each day in the first month. One CHW was herself the mother of one

of the LBW infants.

50

4.5. KMC-practices observed by the CHWs The CHWs interviewed were also asked if they observed that KMC was being practiced

after discharge from the hospital (in the first week after discharge, in the 2nd week, in

the 3rd week, in the 4th week, in the following weeks). Possible answers were: yes, no or

cannot remember.

4.5.1. Observations during the first week at home

Of those interviewed, all but one answered ‘yes’ or ‘no’ in the first week after

discharge. More than 2/3 of the CHWs interviewed (N=85), confirmed that they

observed the caregiver (usually the mother) practicing KMC at least once during the

first week that the child arrived at home. Of those 85, there were 84 CHWs who

answered for another question that they visited the family at least once during the first

week after discharge of the child (see table 21). Of the 38 who answered ‘no’, there

were 13 who responded that they never visited the family in the first week. If they

would have visited the family at home, they would probably have seen the caregiver

practicing KMC because 98 caregivers declared that they practiced KMC after the child

was discharged from the hospital. Table 21 also shows that 23 CHWs who visited the

family at home at least once, did not observe the caregiver practicing KMC in the

community. This corresponds with the number of caregivers who declared that they did

not continue KMC in the community.

Table 21: Did the CHW observe the use of KMC in week 1after discharge (N=124)?

Variable Week 1

Yes, I observed the use of KMC 85 (68.6)

- At least one visit 84 (98.8)

- Never visited 1 (1.2)

No, I did not observe the use of KMC 38 (30.7)

- At least one visit 23 (60.5)

- Never visited 13 (34.2)

- Not answered 2 (5.3)

Do not remember 1 (0.8)

51

The CHWs were also asked how many times they observed the caregiver practicing

KMC. Linking this information with the number of times they reported having visited

the family at home might also give an idea about how frequent KMC has been practiced

during daytime. Two CHWs did not answer this question and 12 others could not

remember the number of times they saw the caregiver practicing KMC.

For those who did remember the number of times they saw the caregiver practicing

KMC, a distinction was made between those who never saw the caregiver practicing

KMC during their home visit (=’never’), those who saw the caregiver practicing KMC

each time they visited the family at home (=’always’), and those who saw the caregiver

practicing KMC less times than the number of times they did home visits during that

week (=’sometimes’). Table 22 presents the results.

Table 22: Frequency of KMC-practice observation during home visits: first week after

discharge (N=110)

Frequency Week 1

Never 28 (25.5)

Sometimes 10 (9.1)

Always 72 (65.4)

Table 22 shows that in the first week 65.4% of the caregivers were observed to practice

KMC during each home visit. One fourth of the CHWs never observed the practice of

KMC during their home visits.

4.5.2. Observations during the first month at home

The same questions about the observations by the CHW were also asked for each week

after the first week that the child arrived at home. Of those interviewed, all but four,

answered ‘yes’ or ‘no’ to the question whether they observed the caregiver practicing

KMC after the first week that the child arrived home (see table 23).

The percentage of CHWs who observed the practice of KMC among the caregivers goes

down week after week. The fourth week it falls down under 50 %. The large majority

(more than 90%) of those who observed the KMC-practice at least once also visited the

family concerned at least once that week. A small number of CHWs also observed the

52

use of KMC although they did not visit the family that week (suggesting KMC is

sometimes also practiced outside the home).

Table 23: Did the CHW observe the use of KMC in the first month after discharge (N=124)?

The percentage of CHWs who did not observe the caregiver practicing KMC increases

week after week. As shown in table 23 this is strongly related to the growing number of

CHWs who did not visit the family in the weeks following the first week after

discharge.

The CHWs were also asked how many times they observed the caregiver practicing

KMC during the weeks after the first week that the baby arrived home (see table 24).

Table 24: Frequency of KMC-practice observation during home visits: first month after

discharge

Week 1

N=110

Week 2

N=104

Week 3

N=104

Week 4

N=99

Never 28 (25.5) 32 (30.8) 41 (39.4) 42 (42.4)

Sometimes 10 (9.1) 8 (7.7) 6 (5.8) 7 (7.1)

Always 72 (65.4) 64 (61.5) 57 (54.8) 50 (50.5)

While in the first week, about 65 % of CHWs always saw the caregiver practicing KMC

during each home visit, this percentage goes down every week reaching 50 % during the

4th week after discharge. The percentage of CHWs who were never observed the

Variable Week 1 Week 2 Week 3 Week 4

Yes, I observed KMC use 85 (68.6) 76 (61.3) 66 (53.2) 61 (49.2)

- At least 1 visit that week 84 (98.8) 74 (97.4) 60 (90.9) 58 (95.1)

- Never visited that week 1 (1.2) 2 (2.6) 6 (9.1) 3 (4.9)

No, I didn’t observe KMC 38 (30.7) 44 (35.5) 54 (43.6) 59 (47.6)

- At least 1 visit that week 23 (60.5) 25 (56.8) 25 (46.3) 23 (39.0)

- Never visited that week 13 (34.2) 17 (38.6) 27 (50.0) 34 (57.6)

- Not answered 2 (5.3) 2 (4.6) 2 (3.7) 2 (3.4)

Do not remember 1 (0.8) 4 (3.2) 4 (3.2) 4 (3.2)

53

practice of KMC during the home visits increases from 25.5 % in the first week to

42.4% in the fourth week.

4.6. How to improve the practice of KMC in the community?

Both the main respondents and the CHWs were asked to give some suggestions about

how the practice of KMC in the community can be improved. In both cases, each

respondent could give more than one suggestion.

Table 25 presents the different suggestions made by the caregivers.

Table 25: Suggestions that can improve future KMC-practice in the community, as reported by

the caregivers (N=124)

Variable N (%)

• Mothers need more support in the provision of food

• Mothers and family members need to be better informed

• Mothers need more support in the provision of insurance

• Mothers need more support in the provision of equipment to keep warm

• Mothers should spend more time taking care of the baby

• Mothers need to better follow the instructions given by the CHW

• Mothers need more support in the provision of modern carrying pouches

• Mothers need to better feed the baby

• Health care staff should visit the mothers at home

• Mothers need to be encouraged to give birth at a health facility

• The mothers need to have more patience

• Mothers should not do difficult activities

30 (24.2)

26 (21.0)

24 (19.4)

21 (16.9)

18 (14.5)

14 (11.3)

9 (7.3)

7 (5.6)

6 (4.8)

4 (3.2)

2 (1.6)

2 (1.6)

Table 25 indicates that most of the caregivers suggest giving more support to them. This

includes more support in the provision of food, health insurance, equipment to carry the

baby and to keep the baby warm, as well as to give birth in health facilities rather than

at home. A second set of suggestions refers to the responsibilities of the

mothers/caregivers. According to some caregivers, they should follow the instructions

54

of the CHWs better, have more patience, make more time for feeding and caring the

baby and avoid activities that make it difficult to practice KMC. A third set of

suggestions refers to responsibilities of health care personnel (visit the mothers at home,

inform the families more).

The suggestions made by the interviewed CHWs to improve future KMC-practice in the

community are shown in table 26. The types of suggestions were similar to those given

by the caregivers, but they formulated more suggestions. Support in accessing food,

provide specific equipment and health insurance and more attention for information and

training in KMC were cited most frequently. Several CHWs also referred to the

importance of involving others, including health centers, to support KMC and to care

for LBW infants after discharge from the hospital. Some CHWs stressed the importance

of hygiene for the mother and of practicing KMC during the night (see table 26).

Table 26: Suggestions that can improve future KMC-practice in the community, as reported by

the CHWs (N=124)

Variable N (%)

• Support vulnerable mothers in the access to food during KMC

• Keep explaining the benefits of KMC to mother and family

• Provide equipment for the mothers such as sheets, a mattress…

• Keep the baby in warm conditions using a hat and socks

• Ask and advice others to support the mother during KMC

• Provide health insurance for mothers using KMC

• Provide modern carrying pouches

• Provide more training for the CHWs

• Empower HCs to care for LBW infants at their level

• Mothers need to take care of their hygiene

• Bring the child to check-ups

• Get help when the child has some problems

• Practice it more at night

42 (33.9)

30 (24.2)

27 (21.8)

18 (14.5)

15 (12.1)

14 (11.3)

13 (10.5)

12 (9.7)

10 (8.1)

9 (7.3)

8 (6.4)

4 (3.2)

3 (2.4)

55

4.7. Other health care contacts In order to get a more complete picture of health care practices after LBW newborns

were discharged from KABUTARE, some additional questions were asked to the

caregivers. The first question deals with the care sought in health centers (HCs) when

the child is sick. The large majority goes to the HC of their own sector. A few go to a

HC outside their sector, and 12.9 % declared that they do not go to any HC (see table

27). The last group rather goes the CHW, the local pharmacist or the hospital when the

child is sick, while 4.8 % never seek professional health care (they stay home during the

sickness of the child).

Rwanda has a very high vaccination rate of almost 100 % (WHO, 2015). Usually, the

caregivers go to the HC of the sector where the family lives for getting the baby

vaccinated (see table 27). Some told the interviewer that they go to another HC (7.3 %)

or to the cell office (5.6%).

Table 27: Contacts with HCs (N=124)

Variable N (%)

Which health center do you visit when your child is sick?

• Go to HC in own sector

• Go to HC in other sector

• No HC

103 (83.1)

5 (4.0)

16 (12.9)

Where did you go for the vaccinations of the child?

• To the HC in own sector

• To the HC in another sector

• To the cell office

108 (87.1)

9 (7.3)

7 (5.6)

Another question explored the use of hospital care by the LBW children of the study

sample after they were first discharged from KABUTARE. Almost 30% of the

caregivers responded that they sought care in a hospital at least once, after having been

discharge before. If they do, they usually went back to KABUTARE (92 %). The most

common reason given by them was ‘when the child is sick’. Ten caregivers explained

that they went back to the hospital for ‘regular check-ups’ (see table 28).

56

Table 28: Hospital care

Variable N (%)

Did you go back to the hospital after the birth of your child? (N=124)

• Yes

• No

38 (30.6)

86 (69.4)

If yes, which hospital? (N=38)

• KABUTARE

• CHUB

• Other

35 (92.1)

2 (5.3)

1 (2.6)

What was the reason? (N=38)

• Sickness

• Regular check-up

• The child was not improving

25 (65.8)

10 (26.3)

3 (7.9)

4.8. Correlations

In order to calculate the correlations, four dependent and six independent variables were

selected. Different combinations were made through a chi-squared test. Therefore two

assumptions needed to be fulfilled each time: 80 % of the cells must be bigger than five

and all cells must be bigger than one. Our predetermined alpha level of significance is

10 % (p < 0.10). Each dependent variable represents the extent to which KMC was

practiced: (1) duration (number of days) of KMC during daylight, (2) duration of KMC

at night, (3) number of hours of KMC during daylight and (4) number of hours of KMC

at night. The independent variables on the other hand, show different characteristics of

the main caregiver or the child: (1) level of education, (2) marital status, (3) year of

birth, (4) part of twin. Finally, the correlation with health insurance coverage was

analyzed (when the infant was < 1 year and at the moment that the main respondent was

interviewed).

First, the correlation with the dependent variable ‘duration of KMC during daylight’

was analyzed. Four independent variables were not significantly correlated with the

number of days that KMC was practiced in the community during daylight: the level of

education (p = 0.678), the year of birth (p = 0.500), part of twin (p = 0.490) and health

insurance at the time of birth (p = 0.191) and at time of the interview (p = 0.267). Only

57

one dependent variable, the marital status did show a significant relationship with the

duration of KMC during the day (see table 29).

Table 29: Correlation between number of days KMC during daylight and marital status

Duration KMC during daylight

Marital Status Married Others Total

# % # % # %

None 18 81.8% 4 18.2% 22 100%

1-14 days 15 83.3% 3 16.7% 18 100%

15 days – 1 month 31 81.6% 7 18.4% 38 100%

> 1 month 24 57.1% 18 42.9% 42 100%

TOTAL 88 73.3% 32 26.7% 120 100%

Chi2 = 8.681, p = 0.034

Table 29 shows that 81.8 % of those who never practiced KMC during the day were

married and almost 43 % of those who practiced KMC during daylight for more than

one month were not married (divorced, separated, widowed).

Secondly, the correlation with the dependent variable ‘duration of KMC during the

night’ was analyzed. All but one selected independent variables were not significantly

correlated with the number of nights KMC was practiced at home: marital status (p =

0.242), year of birth (p = 0.113), part of twin (p = 0.950), health insurance when child

was <1 year (p=0.179) and at time of interview (p = 0.466).

Only the level of education showed a significant relationship with the number of nights

KMC was practiced. Table 30 shows that only 9 of 20 mothers with less than 3 years of

primary education versus more than half of mothers with at least 3 years of primary

education continued KMC at home during at least 15 nights.

58

Table 30: Correlation between number of nights of KMC at home and level of education

Duration KMC

at night

Level of education None or Max 2 years primary

At least 3 years primary

Total

# % # % # %

None 9 23.1% 30 76.9% 39 100%

1-14 nights 11 44.0% 14 56.0% 25 100%

15 nights – 1 month 3 9.4% 29 90.6% 32 100%

> 1 month 6 26.1% 17 73.9% 23 100%

TOTAL 29 24.4% 90 75.6% 119 100%

Chi2 = 9.203, p = 0.027

Finally, the same selected independent variables were also correlated with the number

of hours that KMC was practiced in the community during daylight and at night. Aside

from a slight significance with the independent variable “health insurance at the time of

interview” (see table 31), none of the selected characteristics of the caregiver or the

child had any significance with the number of hours per day that KMC was practiced in

the community.

Table 31: Correlation between number of hours KMC during daytime and health insurance

coverage at time of the interview

Number of hours KMC

during daytime

Health insurance at time of interview Yes No Total

# % # % # %

None 15 68.2% 7 31.8% 22 100%

1-3 hours/day 18 43.9% 23 56.1% 41 100%

4-6 hours/day 21 72.4% 8 27.6% 29 100%

> 6 hours/day 13 50.0% 13 50.0% 26 100%

TOTAL 67 56.8% 51 43.2% 118 100%

Chi2 = 7.311, p = 0.063

59

Chapter 5: Summary and Discussion

5.1. Aim of this master’s thesis

The aim of this study was to collect information about some health care practices in

relation to LBW infants who received neonatal care in 2011 and 2012 in the district

hospital KABUTARE. Three research questions were formulated: (1) To what extent

and how long has the KMC-method been applied in the hospital and after discharge?,

(2) What barriers have been identified concerning the practice of KMC in the

community? and (3) What types of planned and/or unplanned health care contacts did

these LBW infants receive during their first three years of living?

5.2. Main findings

Of the 124 main respondents, 113 confirmed that the KMC-practice started in the

hospital. In 6 cases it was not the mother who practiced KMC in the hospital. Of those

who started practicing KMC in the hospital, 96 (= 85 %) stayed at least one day in the

KMC-room. Of those 96, 32.3% left the room within one week, while 27.1% stayed

there one month or more.

Of the main respondents who started KMC in the hospital, only 6 did not continue

KMC in the community during daytime, but 22 never practiced KMC at night. During

daytime, 98 respondents practiced KMC with an average of 47 days during four hours

per day on average. Of those 98, only 8 stopped already within one week, while 42

continued during one month or more. About 43% practiced KMC only 1 to 3 hours

during daytime, while 27% did so for more than 6 hours during daytime.

Of the 82 who continued KMC at least one night after being discharged home, they did

with an average of 27 nights during 3.5 hours per night on average. Of those 82, 19

stopped already within one week, while 22 continued during one month or more. About

50% practiced KMC only 1 to 3 hours during per night, while 27% did so for more than

6 hours per night.

The CHWs who were most familiar with the selected study population were also

interviewed. Their observations with the practice of KMC at home were in line with

60

what was reported by the main respondents: Almost 2/3 of 110 CHWs observed the

caregiver practicing KMC during each home visit. One fourth of the CHWs never

observed the practice of KMC during their home visits. The percentage of CHWs who

observed the practice of KMC among the caregivers goes down week after week. The

fourth week it falls down under 50 %.

Note that 14 CHWs (11 %) reported that they never visited the child assigned to them at

home. More in depth questioning is needed to know what may be the reasons for this.

Another 26 % did so less than what the guidelines of the MoH specify for LBW-infants

(5 visits in 4 weeks). However, it is encouraging to learn that more than 50% do more

home visits than the required 5 visits. It suggests the high motivation of many CHWs to

visit the families with preterm infants in their neighborhood. Some visited them even

daily, not only in the first week, but in the following weeks as well.

Yet, according to the same government guidelines, the CHW should visit the LBW-

infant families 3 times in the first week after discharge. Sixty percent of the CHWs

interviewed reported less than 3 visits in the first week. Could it be that the CHWs are

not always informed about whether the child was born prematurely? The overall

average number of home visits during the first week was 2.4 times.

Aside from the contacts with the CHWs, 83% of the caregivers also had contacts with a

health center when the child is sick, although not always with the HC of the sector

where they live. Surprisingly, 13 % of the caregivers declared that they stay home when

their child is sick. About all the caregivers get their child vaccinated, also not always in

the health center of their sector. Some caregivers (30%) went back at least once after

discharge to a hospital, most often KABUTARE. Reasons given were: sickness of the

child (25 cases) or regular check-ups (10 cases).

Finally, both the caregivers and CHWs were also asked about their opinions concerning

barriers / obstacles for practicing KMC in the community and how this practice can be

improved in the future. One hundred caregivers answered to the question whether it was

difficult or not to continue practicing KMC in the community.

Of those who did mention at least one obstacle (72 cases), the most frequent answers

were: difficult to work in the fields with KMC (70 % of the 72 caregivers), trouble

getting food (53 %) while practicing KMC, lack of equipment such as warm cloths, a

61

mattress, etc… (43 %) and health problems (25 %). Of the 14 caregivers who reported

“having twins” as a barrier, 12 had both twins alive at the moment of the interview.

Both the main respondents and the CHWs were asked to give some suggestions about

how the practice of KMC in the community can be improved. Most of the caregivers

suggest giving more support to them, such as support in the provision of food, health

insurance, equipment to carry the baby and to keep the baby warm. A second set of

suggestions refers to the responsibilities of the mothers/caregivers, such as following

the instructions of the CHWs better, have more patience, make more time for feeding

and caring the baby and avoid activities that make it difficult to practice KMC. A third

set of suggestions refers to responsibilities of health care personnel, such as visit the

mothers at home and inform the families more about the benefits of KMC. Similar

suggestions were made by the CHWs interviewed. In addition, several CHWs also

referred to the importance of involving others, including HCs, to support KMC and to

care for LBW infants after discharge from the hospital. Some CHWs stressed the

importance of hygiene for the mother and of practicing KMC during the night.

Given the large variation in KMC-practice duration and hours of KMC practice per day,

analyses were conducted to find out whether this variation can be, in part, explained by

characteristics of the study population.

No significant correlations were found between:

1) KMC-practice among infants born within one year that KMC was introduced in

KABUTARE (2011) versus KMC-practice among infants born after the first

year that KMC was introduced in that hospital (2012);

2) KMC-practice among infants born as singletons versus part of twins (although

12 out 14 mothers with both twins alive reported that having twins causes

difficulties in practicing KMC);

3) The families with health insurance while the child was less than 1 year versus

those without health insurance in the same period.

A highly significant correlation was found between marital status and KMC-practice

during daytime. Married mothers are less likely to continue KMC for more than one

month in the community than those who are separated, divorced or widows. Further

research is necessary to found whether the health status at birth may be better among

62

married mothers or whether the environment of married mothers may cause them to

stop KMC-practice sooner than the mothers with another marital status.

A significant correlation was also found between level of education and duration of

KMC-practice at night. Those who finished at least 3 years of primary education tend to

practice KMC longer during the night than those who had no education or less than 3

years of primary education.

Finally, there is a slight significant correlation between having health insurance at the

time of the interview and the number of hours per day that KMC is practiced. Note,

however, that only 57 % of the respondents said that they have health insurance at the

moment of the interview, while 85% reported to have health insurance when the child

was less than 1 year old. Maybe this could be due to the fact that the benefits of health

insurance are larger during the period of pregnancy, delivery and post-partum care.

However, it may also be the lower level of health insurance coverage is due to the fact

that most interviews were conducted at a moment that many people did not renew yet

their health insurance coverage for the year July 2014-June 2015.

5.3. Findings confronted with the existing literature

Nguah et al. (2011) reported that 61.9% of 202 mothers in Ghana thought KMC was

easy to practice. The current study indicates that only 28 of the 124 cases experienced

no difficulties in the practice of KMC. Nguah et al. (2011) also reported that 95.5 % of

the mothers decided to continue KMC at home and 93.1% were willing to practice it

during the night. In this study practicing KMC at home during the night is less likely to

occur, namely 73 % of those who started KMC in the hospital.

The interference of KMC in the mother’s daily work schedule is a frequently reported

barrier in the literature (Bazzano et al., 2012; Charpak & Ruiz-Pelaez, 2006; Hunter et

al., 2014; Parikh, Banker, Shah & Bala, 2013). Similar results are found in this study

where the most frequently reported barrier is the difficulty to work (in the fields) while

simultaneously practicing KMC (70.8 % of the 72 caregivers). Note that this study was

conducted in a rural area of Rwanda. This is also visible through the reported

profession, 89.5% said to be farmer, cultivator or planter. Rural women may choose to

63

stop or interrupt the KMC-practice as it is common practice in the Sub-Saharan region

to work in the fields with the baby on the back. Bazzano et al. (2012) and Bergh et al.

(2014) also reported this tradition in studies conducted in Ghana, Malawi, Mali and

Uganda.

According to Bergh et al. (2012) it is beneficial to have support from the partner during

KMC. In this study, 11 % of the 72 caregivers reported disagreements with their partner

concerning KMC as one of the barriers. The correlation analysis also indicates that

being married is not necessarily an advantage to practice KMC at home.

WHO (2003) recommends the use of a support binder for a better practice of KMC.

When asking for suggestions that can improve the practice of KMC this was brought up

by 7.3 % of the caregivers and by 10.5 % of the CHWs.

In this study, difficulties to sleep were reported by 12.5 % of the 72 caregivers. Charpak

& Ruiz-Pelaez (2006) published that this should be one of the main concerns before

starting KMC.

Charpak & Ruiz-Pelaez (2006) also mentioned that mothers experience difficulty to

take care of their own hygiene. This was reported by 7.3 % of the CHWs as an issue to

be dealt with in order to improve the practice of KMC.

5.4. Limitations

There were a few limitations to this study. A first limitation is memory bias due to the

time between the date of birth of the child and the time when the interviews were

conducted. All children were born between January 1st 2011 and December 31th 2012,

however the interviews only started July 2014. As a result, the received information

could be less accurate then if the interviews were conduct in an earlier period of time.

Not all main respondents were present in the hospital at the time the child received

neonatal care. This may explain ‘missing values’ among the answers of the

questionnaire.

Last, a large majority of respondents were living in the district of HUYE. The others

live in two neighboring districts. The results are, therefore, not necessarily

representative for Rwanda as a whole.

64

Chapter 6: Conclusions and recommendations

This master’s thesis sought some answers to three research questions: “To what extent

and how long has the KMC-method been applied in the hospital and after discharge?”,

“What barriers have been identified concerning the practice of KMC in the

community?” and “What types of planned and/or unplanned health care contacts did

these LBW infants receive during their first three years of living?”.

KMC started in January 2011 in the district hospital KABUTARE. Since then a large

majority of LBW newborns who received neonatal care in that hospital, also received

KMC during their hospital stay and after they were discharged. However, the responses

of 124 caregivers showed that there is a large variation in the timing that they received

information about KMC, the number of days that the caregivers stayed in the KMC

room, the number of days and nights that KMC was practiced in the community, as well

as the number of hours per day and per night. Some characteristics of the caregivers

(married mothers, mothers with little or no school years) are significantly correlated

with some measures of KMC-practice, but more research is necessary to explain the

causes of these significant relationships.

As the results of this study indicate that KMC-practice is not yet optimal, it can be

recommended that all health care providers would scale up their efforts to provide

information concerning the benefits of KMC and how to practice it continuously to all

mothers giving birth to LBW children, as well as their partners. Special attention should

be directed towards KMC-practice at night as KMC appears to be less commonly

practiced then.

Many respondents reported difficulties to practice KMC in the community. The answers

suggest that several difficulties can be overcome by providing more support to the

caregivers in getting food, health insurance, and material that help them to practice

KMC. As “working in the fields” is the most frequently mentioned obstacle to practice

KMC in rural areas, specific strategies have to be worked out to deal with this obstacle.

The responses of the CHWs interviewed indicate that the majority of CHWs tend to be

highly motivated to visit the infants at home, especially in the first month after birth.

65

Yet, 11 % of the CHWs reported that they never visited the child assigned to them at

home. Another 26 % did so less than what the guidelines of the MoH specify for LBW-

infants (five visits in four weeks).

It is recommended that the CHW-coordinators of the health centers would supervise the

CHWs more closely and give the necessary training to be sure that they are all aware of

the importance of more frequent home visits for preterm infants, and of continuing these

visits for a longer period after birth. This also requires that the health centers are well

informed by the hospital immediately after each discharge of a preterm neonate.

This study is limited to the experience in a few districts in the South of Rwanda. As it is

the first in Rwanda and surrounding countries, it brings important new information

about the post-neonatal care to LBW-infants. Preferably, this study should be repeated

in other districts and other Sub-Saharan countries on a larger scale and at different times

after the children were discharged from the hospital (six months, one year, etc…) to

collect additional evidence to support policy-makers, health care providers and health

services researchers.

66

References

Akhtar, K., Haque, M. & Khatoon, S. (2013). Kangaroo Mother Care: A simple method

to care for low-birth-weight infants in developing countries. Journal of Shaheed

Suhrawardy Medical College, 5(1), 49-54.

Ballantyne, M., Stevens, B., Guttmann, A., Willan, A.R. & Rosenbaum, P. (2013).

Maternal and infant predictors of attendance at Neonatal Follow-Up programs. Child:

Care, Health and Development, 40(2), 250-258.

Bazzano, A., Hill, Z., Tawiah-Agyemang, C., Manu, A., ten Asbroek, G. & Kirkwood,

B. (2012). Introducing home based skin-to-skin care for low birth weight newborns: a

pilot approach to education and counseling in Ghana. Global health promotion, 19(3),

42-49.

Beck, S., Wojdyla, D., Say, L., Betran, A.P., Merialdi, M., Requejo, et al. (2010). The

worldwide incidence of preterm birth: a systematic review of maternal mortality and

morbidity. Bulletin of the World Health Organization, 88(1), 1-80.

Bergh, A.M., Charpak, N., Ezeonodo, A., Udani, R.H. & van Rooyen, E. (2012).

Education and training in the implementation of kangaroo mother care. South African

Journal of Child Health, 6(2), 38-45.

Bergh, A.M., Kerber, K., Abwao, S., de-Graft Johnson, J., Aliganyira, P., Davy, K., et

al. (2014). Implementing facility-based Kangaroo Mother Care services: lessons from a

multi-country study in Africa. BMC Health Services Research, 14, 293.

Bergh, A.M., Manu, R., Davy, K., van Rooyen, E., Asare, G.Q., Williams, J.K.A., et al.

(2012). Translating research findings into practice – the implementation of Kangaroo

Mother Care in Ghana. Implementation Science, 7(75), 2-9.

67

Black, R.E., Cousens, S., Johnson, H.L., Lawn, J.E., Rudan, I., Bassani, D.G., et al.

(2010). Global, regional, and national causes of child mortality in 2008: a systematic

analysis. The Lancet, 375, 1969-1987.

Blencowe, H., Cousens, S., Oestergaard, M.Z., Chou, D., Moller, A.B., Narwal, R. et al.

(2012). National, regional, and worldwide estimates of preterm birth rates in the year

2010 with time trends since 1990 for selected countries: a systematic analysis and

implications. The Lancet, 379, 2162-2172.

Briggs, P. & Booth, J. (2009). Rwanda. Guilford: The Globe Pequot Press Inc.

Cammu, H., Martens, E., Martens, G., Van Mol, C. & Jacquemyn, Y. (2013).

Perinatale activiteiten in Vlaanderen 2013. Brussel: Studiecentrum voor Perinatale

Epidemiologie.

Chambers, V. & Booth, D. (2012). Delivering maternal health. Why is Rwanda doing

better than Malawi, Niger and Uganda? Retrieved March 6, 2015, from

http://www.odi.org/sites/odi.org.uk/files/odi-assets/publications-opinion-files/7696.pdf

Charpak, N. & Ruiz-Pelaez, J.G. (2006). Resistance to implementing Kangaroo Mother

Care in developing countries, and proposed solutions. Acta Paediatrica, 95, 529-534.

Chisenga, J.Z., Chalanda, M. & Ngwale, M. (2014). Kangaroo Mother Care: A review

of mothers’ experiences at Bwaila hospital and Zomba Central hospital (Malawi).

Journal of Midwifery, 4(8), 1-10.

Conde-Agudelo, A., Belizan, J.M. & Diaz-Rossello, J. (2011). Kangaroo Mother Care

to reduce morbidity and mortality in low birth weight infants (Review). Cochrane

Database of Systematic Reviews, 3, 1-116.

68

Dhillon, R.S., Bonds, M.H., Fraden, M., Ndahiro, D. & Ruxin, J. (2012).The impact of

reducing financial barriers on utilization of a primary health care facility in Rwanda.

Global Public Health, 7(1), 71-86.

Government of the Republic of Rwanda (2014). Health system. Retrieved March 29,

2015 from http://www.gov.rw/services/health_system/

Hunter, E.C., Callaghan-Koru, J.A., Mahmud, A.A., Shah, R., Farzin, A., Cristofalo, et

al. (2014). Newborn care practices in rural Bangladesh: Implications for the adaptation

of kangaroo mother care for community- based interventions. Social Science &

Medicine, 122, 21-30.

IGuide Rwanda (2015). Map of Rwanda. Retrieved February 23, 2015 from

http://www.theiguides.org/public-docs/guides/Rwanda

Ministry of Health (2011). Comprehensive multi-year plan 2011-2015. Retrieved

February 20, 2015 from http://www.gavi.org/country/rwanda/documents/

Ministry of Health (2012). Rwanda Health Statistics Booklet 2011. Retrieved February

21, 2015fromhttp://www.moh.gov.rw/fileadmin/templates/HMIS_Docs/

MOH_Annual_booklet-2011.pdf

Ministry of Health (2012). Rwanda Annual Health Statistics Booklet 2012. Retrieved

February 21, 2015 from http://www.moh.gov.rw/fileadmin/templates/MOH-

Reports/MOH_Booklet_2012_final_September_2013.pdf

Ministry of Health (2013). National Community Health Strategic Plan. Retrieved

March 29, 2015 from

http://www.nationalplanningcycles.org/sites/default/files/country_docs/Rwanda/chd-

strategic_plan.pdf

69

Nguah, S.B., Wobil, P.N.L., Obeng, R., Yakubu, A., Kerber, K.J., Lawn, J.E., et al.

(2011). Perception and practice of Kangaroo Mother Care after discharge from hospital

in Kumasi, Ghana: A longitudinal study. BMC Pregnancy and childbirth, 11(99), 1-8.

Parikh, S., Banker, D., Shah, U. & Bala, D.V. (2013). Barriers in implementing

community based Kangaroo Mother Care in low-income community. Journal of

Medical Sciences, 2(1), 36-38.

Pierce, H., Heaton, T.B. & Hoffmann, J. (2014). Increasing maternal healthcare use in

Rwanda: Implications for child nutrition and survival. Social Science & Medicine, 107,

61-67.

Pratomo, H., Uhudiyah, U., Sidi, I.P.S., Rustina, Y., Suradi, R., Bergh, A.M., et al.

(2012). Supporting factors and barriers in implementing kangaroo mother care in

Indonesia. Paediatrica Indonesiana, 52(1), 43-50.

Rwanda PNC review (2012). Summary of Rwanda Country experience – implementing

PNC home visits. Retrieved February 22, 2015 from

http://www.healthynewbornnetwork.org/sites/default/files/resources/Country%20Report

%20PNC%20Home%20Visits%20-%20Rwanda.pdf

The World Bank (2015). Data: Country and lending groups. Retrieved October 16,

2014, from http://data.worldbank.org/about/country-and-lending-groups

The World Bank (2013). Rwanda. Retrieved April 3, 2015 from

http://data.worldbank.org/country/rwanda

UNDP (2015). Improve maternal health. Retrieved April 25, 2015 from

http://www.rw.undp.org/content/rwanda/en/home/mdgoverview/overview/mdg5.html

UNICEF & WHO (2004). Low Birthweight. Country, regional and global estimates.

Retrieved January 28, 2015, from

70

http://www.unicef.org/publications/files/low_birthweight_from_EY.pdf

United Nations (2014). We can end poverty: Millennium development goals and beyond

2015. Retrieved October 15, 2014, from

http://www.un.org/millenniumgoals/childhealth.shtml

Vahidi, R.G., Gholipour, K., Jannati, A., Hosseini, M.B., Ghoddoosi-Nejad, J. & Bayan,

H. (2014). Cost and Effectiveness Analysis of Kangaroo Mother Care and Conventional

Care Method in Low Birth Weight Neonates in Tabriz 2010 – 2011. Journal of Clinical

Neonatology, 3(3), 148-152.

Van den Brande, J.L., Heymans, H.S.A. & Monnens, L.A.H. (1998).

Kindergeneeskunde (3de dr.).Maarssen: ELSEVIER Gezondheidszorg.

World Health Organization (2003). Kangaroo Mother Care: A practical guide.

Retrieved February 12, 2015 from

http://whqlibdoc.who.int/publications/2003/9241590351.pdf?ua=1

World Health Organization (2012). Born too soon: The global action report on preterm

birth. Retrieved October 19, 2014, from

http://www.who.int/pmnch/media/news/2012/201204_borntoosoon-report.pdf

World Health Organization (2013). Media centre: Preterm birth. Retrieved October 15,

2014, from http://www.who.int/mediacentre/factsheets/fs363/en/

World Health Organization (2015). World Health Statistics. Retrieved June 23, 2015

from

http://apps.who.int/iris/bitstream/10665/170250/1/9789240694439_eng.pdf?ua=1&ua=

1

I

Appendix

1. Questionnaire

VI

2. Informed Consent Form

VII

3. Approval National Ethics Committee Rwanda

List of tables and figures

Tables

Table 1: Estimated preterm birth prevalence rates for each region in 2005……..…...………17

Table 2: Child deaths occurred in 2008 worldwide………………………………..…………18

Table 3: Top causes of death in neonatology in all health facilities in 2012………....………19

Table 4: Gestational age (2013)………………………………………………………....……19

Table 5: Perinatal mortality by gestational age (2013) in ‰………………………………....20

Table 6: Demographic data concerning the main respondent.……………….…………..…...39

Table 7: Demographic data concerning the child……………………………..……………...41

Table 8: Health insurance coverage………………………………………….……………….42

Table 9: Moment KMC-information was received………………...……………...……….…43

Table 10: KMC-practice in the hospital………………………………………………………43

Table 11: Moment that KMC was started in the hospital……………………………….........44

Table 12: Number of days in the KMC-room….…………………………………………..…44

Table 13: Duration of KMC-practice in the community during daylight………………….…45

Table 14: Number of hours per day of KMC-practice during daylight…………...……….…45

Table 15: Duration of KMC-practice at home during the night……………….…………......45

Table 16: Number of hours of KMC-practice at home during the night………………..……46

Table 17: Obstacles in practicing KMC reported by the main respondents……………….…46

Table 18: Frequency of the home-visits conducted by the CHW during the first week after

discharge………………………………………………………………………………...……48

Table 19: Frequency of home-visits by the CHW, week by week.…………...…………...…49

Table 20: Number of home visits per CHW from week 1 to week 4……………………...…49

Table 21: Did the CHW observe the use of KMC in week 1 after discharge?……….............50

Table 22: Frequency of KMC-practice observation during home visits: first week after

discharge……………………………………………………………………………………...51

Table 23: Did the CHW observe the use of KMC in the first month after discharge?……….52

Table 24: Frequency of KMC-practice observation during home visits: first month after

discharge…………………………………………………………………………….…..……52

Table 25: Suggestions that can improve future KMC-practice in the community, as reported

by the caregivers……………………………………………………………………...………53

Table 26: Suggestions that can improve future KMC-practice in the community, as reported

by the CHWs………………………………………………………….……………..………..54

Table 27: Contacts with HCs……………………………………………………………..…..55

Table 28: Hospital Care………………………………………………………………………56

Table 29: Correlation between number of days KMC during daylight and marital

status………………………………………………………………………………………….57

Table 30: Correlation between number of nights KMC at home and level of education…….58

Table 31: Correlation between number of hours KMC during daytime and health insurance

coverage at time of the interview…………………………………………………..…………58

Figures

Figure 1: Overview of definitions applied for pregnancy outcomes related to preterm birth and

stillbirths……………………...…………………………………….……...…………………15

Figure 2: Carrying pouches for KMC babies…………………………………………………23

Figure 3: Map of Rwanda…………………………………………………………………….28

Figure 4: Home visit schedule of ASM……………………………………..………………..32

care after discharge of preterm infants -...

Documents