Influence of childcare practices on nutritional status of Ghanaian ...

14 downloads 7039 Views 2MB Size Report
The Ghana Demographic and Health Survey (DHS) data collected in 2008 were used for the analysis. These data are in the public domain and available from ...
BMJ Open

Childcare practices are associated with child nutritional status regardless of socio-demographic background: Findings from the Ghana Demographic and Health Survey 2008

rp Fo Journal:

Manuscript ID: Article Type:

Date Submitted by the Author:

bmjopen-2014-005340 Research 25-Mar-2014

ee

Complete List of Authors:

BMJ Open

Primary Subject Heading:

Nutrition and metabolism, Public health Community child health < PAEDIATRICS, Nutrition < TROPICAL MEDICINE, Epidemiology < TROPICAL MEDICINE

w

Keywords:

ie

Secondary Subject Heading:

Epidemiology

ev

rr

Amugsi, Dickson; University of Bergen, Health Promotion and Development Mittelmark, Maurice; University of Bergen, Department of Health Promotion and Development Lartey, Anna; Food and Agricultural Organization, Nutrion Division, Economic and Social Department Matanda, Dennis; University of Bergen, Department of Health Promotion and Development Urke, Helga; University of Bergen, Department of Health Promotion and Development

ly

on For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 1 of 28

Childcare practices are associated with child nutritional status regardless of socio-demographic background: Findings from the Ghana Demographic and Health Survey 2008 Dickson A. Amugsi1* Email: [email protected] Maurice B. Mittelmark1

rp Fo

Email: [email protected] Anna Lartey2

Email: [email protected] Dennis J. Matanda1

[email protected]

[email protected]

ie

ev

Helga B. Urke1

rr

ee

1. Department of Health Promotion and Development, University of Bergen,

w

Norway

on

2. Nutrition Division, Economic and Social Department, Food and Agriculture Organization, Rome, Italy

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

*Corresponding author: Department of Health Promotion and Development,

University of Bergen, Christiesgt. 13, 5015, Bergen, Norway. Phone: +4746270992, Fax:+4755589887

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

ABSTRACT

Objectives: To examine: (a) the association of childcare practices (CCP) to infant and young children’s growth (height-for–age Z scores, or HAZ), and (b) to establish whether care practices are more important to some socio-demographic subgroups of children compared to others.

rp Fo

Design: Cross sectional survey Setting: The 2008 Ghana Demographic and Health Survey, focused on maternal and child health.

ee

Participants: HAZ scores were calculated for children aged 6-36 months (393 urban and 794). Their mothers (n = 1,187) reported CCP provided to the child,

rr

their own education and occupation status, and their empowerment experience.

ev

Field staff collected data on maternal height/weight, anemia status, household assets using the wealth index, and on household water quality/availability and

ie

sanitation facilities and practices.

w

Results: Principal Components Analysis was used to create a CCP score, combining feeding and preventive health variables: dietary diversity,

on

solid/semisolid feeding frequency, breastfeeding status, vaccination status, iron

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

supplementation and treatment for intestinal parasites. Regression analysis examined the relationship of CCP to child nutritional status, adjusting for child, maternal and household level factors. The results showed that CCP was a significant predictor of HAZ, after controlling for covariates/confounders at child, maternal and household levels. Children with higher CCP scores had higher HAZ scores. In addition to CCP, child’s and mother’s age, number of children under five, place of resident, maternal weight, and wealth index were also 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 3 of 28

significantly associated with HAZ. No statistically significant interactions between CCP and the other independent variables were observed. Conclusions: This study found a significant, positive association between CCP and child growth, regardless of the socio-demographic characteristics of the mother and of the household. This calls for research into the effects on growth of various CCP components, with longitudinal cohort study designs that can

rp Fo

disentangle causal relationships. Keywords: Care practices, nutritional status, children, Ghana Demographic and Health Survey

w

ie

ev

rr

ee ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

Strengths and limitations of this study •

Use of high quality nationally representative sample to investigate the relationship between childcare practices and nutritional outcomes



Creation of a composite childcare measure including home care and medical care dimensions



Cross-sectional survey design, while a longitudinal cohort design would

rp Fo be preferred



Lack of variables to measure cultural, socio-political and locality influences on child health

w

ie

ev

rr

ee ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 5 of 28

Introduction The foundational UNICEF framework for child health emphasizes that childcare practices (CCP) are vitally important in promoting child nutrition and health(1). Socio-demographic factors (e.g., parental education and income) are also emphasized in the UNICEF framework, and are consistently found to have a graded relationship with health(2). However, little is known about the degree to

rp Fo

which CCP are consistently related to child health in the face of the widely differing socio-demographic backgrounds that characterize societies.

Childcare is a complex concept including a range of behaviors and practices of

ee

care givers that provide the food, health care, stimulation, and emotional support

rr

necessary for children’s healthy survival, growth, and development (3). As part of CCP, feeding and health care underlie dietary sufficiency and protection from

ev

disease, which in turn impacts child health, for which physical growth is a critical

ie

marker (4).

w

A robust finding in public health research is that of a graded relationship

on

between socio-demographic status (SDS) and health(5). Low SDS translates predictably into lessened food security and reduced access to health care.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

However, even in households with food insecurity due to poverty and poor access to health care, families can optimize the use of the existing resources to promote health (3, 6). This calls for further research to illuminate the relationship between childcare and child health, in economically vulnerable as well as secure households and communities. An ecological approach to such research calls for specification not only of proximal influences on child health 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

such as feeding practices, but also consideration of more distal factors such as caregivers’ health literacy, availability of resources such as clean water and sanitary living conditions and accessible health care (7) .

In the Global South1, interventions to reduce child under-nutrition focus often on household food security (adequacy of food availability), without taking into

rp Fo

consideration the complementary role of quality childcare. This can be seen in most of the nutrition intervention programs in Ghana(8). Yet, food security alone is not enough to improve children’s nutritional status, and the significance of care practices to improving child nutritional status has been documented

ee

repeatedly (6, 9-16). Despite the fact that quality of childcare has a

rr

demonstrated role in alleviating child under-nutrition in resource-constrained settings like Ghana, there have been only two Ghanaian studies (of which we are

ev

aware) that have examined the role of childcare in relation to child nutritional

ie

status. The pioneering study of Ruel and her colleagues(6) in urban Accra used a

w

composite care practices variable (care practice index) to examine the importance of care for healthy child nutrition. The other study, by Nti and Lartey

on

(16), was conducted in one rural area; both studies found a significant association between care practices and child nutritional status. However, the

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

setting-specificity of these two studies limits the generalizability of their

findings. Addressing this limitation, this paper presents an analysis of the relationship between care practices and child nutritional status in Ghana, using a By ‘Global South’ we refer to parts of the world that are also termed ‘the third world’ and ‘developing countries’ (which may carry pejorative connotations). The Global South is a geopolitical concept including parts of the world located notably in the Equatorial Zone that have colonial pasts, challenging geopolitical conditions, and that are rising in economic, social and political resilience (6). Regions having these conditions are of course found outside the Equatorial Zone.

1

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 7 of 28

national representative sample. The primary objective of this analysis was to examine the influence of CCP on children’s HAZ, controlling for covariates and potentially confounding factors at child, maternal, household, and community levels. The secondary objective was to establish whether care practices were more important to growth in some socio-demographic subgroups of children compared to others.

rp Fo

METHODS

Data sources

Ghana Demographic and Health Survey (DHS) data collected in 2008 were used

ee

for the analysis. These data are in the public domain and available from

rr

MEASURE DHS website (17). The Ghana Statistical Service and the Ghana Health Service collected the data, using the 2000 national population census as a

ev

sampling frame. Ethical clearance was obtained from the Ghana Health Service

ie

Ethical Review Committee.

w

The participants were 1,187 children aged 6-36 months (393 urban and 794

on

rural) from whom anthropometry data were obtained. This excluded 224 children in the survey from whom complete and in-range anthropometry data

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

could not be obtained.

Outcome variable The indicator of child nutritional status was Height-for-age Z-scores (HAZ).

7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

Childcare practices (CCP) measurement The variables used in creating the CCP score were feeding practices variables and use of preventive health service. The feeding practices variables included dietary diversity score, which was created using 16 food groups: 1) tinned/powder or fresh milk; 2) baby formula; 3) baby cereal; 4) bread, rice, noodles, other made from grains; 5) potatoes, cassava, or other tubers; 6) eggs;

rp Fo

7) meat (beef, pork, lamb, goat, chicken etc.); 8) dark green leafy vegetables; 9) mangoes, papayas, other vitamin A fruits; 10) other fruits; 11) pumpkin, carrots, squash (yellow or orange inside); 12) liver, kidney, heart, other organs; 13) fish or shellfish(fresh or dried); 14) food made from beans, peas, lentils, nuts; 15)

ee

oils, fats, butter, products made from them; 16) cheese, yogurt, other milk

rr

products. Details about the DDS are presented elsewhere (in print). Other feeding variables were frequency of feeding solid or semi-solid food and

ev

breastfeeding status. The preventive health service variables included; BCG

ie

vaccination, DPT, Hepatitis B, influenza, polio and measles vaccinations, iron

w

supplementation, and use of drugs for intestinal parasites.

on

The CCP score was created using the results of Principal Component Analysis (18-20). We employed the regression method, with the components loadings

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

adjusted to account for the initial correlations between variables. Since we did not anticipate the relationship between components to be orthogonal, we used the oblique factor rotation procedure. The results were used to create the composite care practices score, treated in subsequent analyses as a continuous variable.

8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 9 of 28

Other variables used in the analysis: (a) maternal age, height, weight, number of antenatal visits (ANC) education, occupation, anemia level, and parity;

(b), method of disposal of youngest child stools;

rp Fo

(b) empowerment variables including women’s role in household decision making, opinion regarding wife beating, and attitudes regarding sexual relations with husband;

ee

(c) household level variables including number of children under 5 years in

rr

household, Wealth Index, urban/rural place of residence, source of drinking water, religion and type of toilet facilities; and

ie

ev

(e) the child level variables sex and age (child’s age was transformed into age

w

squared and included in regression analyses to account for non-linearity of the age variable (21).

on

Some of the variables were recoded. Source of drinking water and toilet facilities

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

were recoded according to WHO and UNICEF (22) recommended classifications: ‘improved’ and ‘unimproved’ water and ‘improved’ and ‘unimproved’ sanitation facilities. The disposal of the youngest child stool was recoded into ‘appropriate’ and ‘inappropriate’ disposal methods. Maternal occupation was recoded into ‘white collar’ and ‘agriculture/labor’, and religion into ‘Christians’ and ‘other religions’. For the empowerment variables, three indices were created based on 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

the DHS (23) recommended procedure (number of household decision making, opinion regarding wife beating, and justified to refuse sexual intercourse with husband). High scores were coded ‘more empowered’ and low scores ‘less empowered’. The wife beating attitude variable was reversed coded so that a high score corresponded to be more empowered.

rp Fo

Analytical framework and methods This analysis is framed using the UNICEF conceptual framework in which food, health, and care are posited as the three key pillars influencing child survival, growth and development (1). The model identifies three levels of causes of child

ee

under-nutrition: immediate (operating at the individual level), underlying

rr

(influencing household and communities) and basic causes (structure and processes of societies). The model suggests that these causal factors affect child

ev

nutritional status in a chain-like manner—the basic factors affect the underlying

ie

factors, which in turn affect the immediate factors, in turn affecting child

w

nutrition status. The model was extended by Engle and colleagues (24) and the above levels reclassified broadly as context, resources and care giving. This

on

analysis used this framework to structure the hierarchical multiple regression analyses.

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

The General Linear Model (GLM) in the SPSS 21 Complex Samples command was used to perform the multivariate analysis. The GLM was used to allow adjustment for survey design effects (sample weight, strata, and cluster). The analysis involved four steps. The first step (model A) contained only the basic characteristics of the mother (age) and child (age and sex), to examine the direct 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 11 of 28

effects of these factors on HAZ. The second step (model B) introduced context variables (place of residence and religion) in the model in the presence of the basic factors to establish how the context variables were directly related to HAZ. The third step (model C) introduces resource variables (education, occupation, anemia level, parity, disposal of youngest child stool, household decision making, opinion regarding wife beating, justified to refuse sexual intercourse with

rp Fo

husband, number of children under 5, wealth index, source of drinking water, type of toilet facilities), controlling for basic and contextual factors. In the final step (model D), CCP score was introduced, controlling for basic, contextual and resource factors. Tests of interactions between the CCP score and other predictor

ee

variable were undertaken, because previous research has documented that

rr

children from poorer households and/or those of mothers with less education may be more likely to benefit more from better care practices, compared to

ev

children of wealthier households or those of mothers with better education(6).

w

ie

RESULTS Characteristics of the sample

on

Tables 1 and 2 present the descriptive statistics of the sample. The average age of children used in the analysis was about 20 months. The mean Height-for-Age

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Z-score for the sample was little above -1 (-1.09), while that of weight-for-age and weight-for-height Z-scores respectively were lower than -1 (-0.81 and -0.33). The average prevalence of stunting, underweight and wasting were 29.1%, 16.0%, and 11.5% respectively. The average age of the mothers was 28 years. The number of Antenatal visits was relatively low (1.74 visits). Breastfeeding was generally above average in this population (67%). The average frequency of 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

feeding the child with solid or semi-solid food within 24hours was 2.59. Immunization rates were high among this population. BCG, which is given at birth, was as high as 94%. Additionally, 87.7% of children older than three months had received all their DPT vaccination and 85.6% received polio 3 vaccinations. For children older than 9 months, 86.7% received measles vaccination. Fewer children in the sample received iron supplement (29.0%).

rp Fo

The use of drugs for intestinal parasites was low (37.2%), probably because the children in the sample were relatively young. With regards to water and sanitation, 22.2% of this population did not have access to improved source of water and 47% used unimproved sanitation facilities. Also, a high proportion of

ee

mothers (63%) used inappropriate ways to dispose the youngest child stool.

w

ie

ev

rr ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 13 of 28

Table 1: Characteristics of the sample (N = 1,187), continuous variables

Variables SD

Child age

19,8

8,55

Child Height-for-age Z-scores (HAZ)

-1,09

1,72

Child weight-for-age Z-scores (WAZ)

-0,81

1,3

Child weight-for-height Z-scores (WHZ)

-0,33

1,49

Maternal age (units)

28,18

6,77

Maternal height (units)

1,59

0,07

Maternal weight (units)

58,17

10,83

No. of children U5 in household

1,88

1.00

No. of ANC visits

1,74

0,53

Number of times child ate solid, semisolid or soft food yesterday

2,59

1,32

Dietary diversity score for child

5,92

3,06

N

(%)

600

50,5

587

49,5

ie

ev

rr

ee

Mean

rp Fo

Table 2: Characteristics of the sample (N = 1,187) categorical variables

w

Characteristics Sex of child

on

Male Female

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Still breastfeeding (yes)

797

67,1

Received BCG (yes)

1120

94,5

Received DPT/Hep B/Influenza 3 (yes)

1037

87,7

Received POLIO 3 (yes)

1013

85,6

Received Measles>=9 months (yes)

927

86,7

Taking iron pills, sprinkles or syrup (last 7 days) (yes)

343

29,0

Drugs for intestinal parasites (yes)

438

37,2

Use of preventive health service

13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

Anemia level (some anemia)

718

61,5

343

28,9

267

23,1

173

15,1

262

22,2

557

47,3

740

62,6

802

67,7

383

32,3

394

33,1

793

66,9

Empowerment Participation in decision-making Low participation

Husband justified in beating wife Less sense of empowerment Wife justified in refusing sex

rp Fo

Less Empowered

Water and sanitation

Source of drinking water Unimproved

Type of toilet facilities Unimproved

rr

ee

Disposal of youngest child stool inappropriate disposal practice Religion Christian

ie

ev

Other religions Place of residence

w

Urban Rural

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 15 of 28

Bivariate analysis of the association between CCP and HAZ Bivariate analysis was carried out to examine the associations between CCP and child nutritional status. The results show a strong positive association between care practices and child HAZ (P 1,

25

and four principal components were extracted that explained 70% of the 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 37 of 58

1

variance. No item had a loading less than .4 (20). Therefore, all the items were

2

used to create the composite care practices score, treated in subsequent analyses

3

as a continuous variable.

4 5

Other variables used in the analysis:

6

(a) maternal age, height, weight, number of antenatal visits (ANC) education,

7 8 9

occupation, anemia level, and parity;

(b), method of disposal of youngest child stools;

ee

10

rp Fo

11

(b) empowerment variables including women’s role in household decision

12

making, opinion regarding wife beating, and attitudes regarding sexual

13

relations with husband;

ie

15

ev

14

rr

(c) household level variables including number of children under 5 years in

w

16

household, Wealth Index, urban/rural place of residence, source of drinking

17

water, religion and type of toilet facilities; and

on

18 19

(e) the child level variables sex and age (child’s age was transformed into age

20

squared and included in regression analyses to account for non-linearity of the

21

age variable (23).

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

22 23

Some of the variables were recoded. Source of drinking water and toilet facilities

24

were recoded according to WHO and UNICEF (24) recommended classifications:

25

‘improved’ and ‘unimproved’ water and ‘improved’ and ‘unimproved’ sanitation 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

1

facilities. The disposal of the youngest child stool was recoded into ‘appropriate’

2

and ‘inappropriate’ disposal methods. Maternal occupation was recoded into

3

‘white collar’ and ‘agriculture/labor’, and religion into ‘Christians’ and ‘other

4

religions’. For the empowerment variables, three indices were created based on

5

the DHS (18) recommended procedure (number of household decision making,

6

opinion regarding wife beating, and justified to refuse sexual intercourse with

7

husband). High scores were coded ‘more empowered’ and low scores ‘less

8

empowered’. The wife beating attitude variable was reversed coded so that a

9

high score corresponded to be more empowered.

ee

10

rp Fo

11

Analytical framework and methods

12

This analysis is framed using the UNICEF conceptual framework in which food,

13

health, and care are posited as the three key pillars influencing child survival,

14

growth and development (1). The model identifies three levels of causes of child

15

under-nutrition: immediate (operating at the individual level), underlying

16

(influencing household and communities) and basic causes (structure and

17

processes of societies). The model suggests that these causal factors affect child

18

nutritional status in a chain-like manner—the basic factors affect the underlying

19

factors, which in turn affect the immediate factors, in turn affecting child

20

nutrition status. The model was extended by Engle and colleagues (25) and the

21

above levels reclassified broadly as context, resources and care giving. This

22

analysis used this framework to structure the hierarchical multiple regression

23

analyses.

w

ie

ev

rr

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 58

24

9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 39 of 58

1

The General Linear Model (GLM) in the SPSS 21 Complex Samples command was

2

used to perform the multivariate analysis. The GLM was used to allow

3

adjustment for survey design effects (sample weight, strata, and cluster). The

4

analysis involved four steps. The first step (model A) contained only the basic

5

characteristics of the mother (age) and child (age and sex), to examine the direct

6

effects of these factors on HAZ. The second step (model B) introduced context

7

variables (place of residence and religion) in the model in the presence of the

8

basic factors to establish how the context variables were directly related to HAZ.

9

The third step (model C) introduces resource variables (education, occupation,

rp Fo

10

anemia level, parity, disposal of youngest child stool, household decision making,

11

opinion regarding wife beating, justified to refuse sexual intercourse with

12

husband, number of children under 5, wealth index, source of drinking water,

13

type of toilet facilities), controlling for basic and contextual factors. In the final

14

step (model D), CCP score was introduced, controlling for basic, contextual and

15

resource factors. Tests of interactions between the CCP score and other predictor

16

variable were undertaken, because previous research has documented that

17

children from poorer households and/or those of mothers with less education

18

may be more likely to benefit more from better care practices, compared to

19

children of wealthier households or those of mothers with better education(6).

w

ie

ev

rr

ee

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

20 21

RESULTS

22

Characteristics of the sample

23

Tables 1 and 2 present the descriptive statistics of the sample. The average age

24

of children used in the analysis was about 20 months. The mean Height-for-Age

25

Z-score for the sample was -1.09 (S.D. = 1.7), while that of weight-for-age and 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

1

weight-for-height Z-scores respectively were -0.81 (S.D. = 1.3) and -0.33 (S.D. =

2

1.5). The average prevalence of stunting, underweight and wasting were 29.1%,

3

16.0%, and 11.5% respectively. The average age of the mothers was 28 years.

4

The number of Antenatal visits was relatively low (1.74 visits). Breastfeeding

5

was generally above average in this population (67%). The average frequency of

6

feeding the child with solid or semi-solid food within 24hours was 2.59.

7

Immunization rates were high among this population. BCG, which is given at

8

birth, was as high as 94%. Additionally, 87.7% of children older than three

9

months had received all their DPT vaccination and 85.6% received polio 3

rp Fo

10

vaccinations. For children older than 9 months, 86.7% received measles

11

vaccination. Fewer children in the sample received iron supplement (29.0%).

12

The use of drugs for intestinal parasites was low (37.2%), probably because the

13

children in the sample were relatively young. With regards to water and

14

sanitation, 22.2% of this population did not have access to improved source of

15

water and 47% used unimproved sanitation facilities. Also, a high proportion of

16

mothers (63%) used inappropriate ways to dispose the youngest child stool.

w

ie

ev

rr

ee

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 58

11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 41 of 58

Table 1: Characteristics of the sample (N = 1,187), continuous variables

Variables Mean

SD

Child age

19,8

8,55

Child Height-for-age Z-scores (HAZ)

-1,09

1,72

Child weight-for-age Z-scores (WAZ)

-0,81

1,3

Child weight-for-height Z-scores (WHZ)

-0,33

1,49

Maternal age ( in years)

28,18

6,77

Maternal height (in cm)

1,59

0,07

Maternal weight (in cm)

58,17

10,83

No. of children U5 in household

1,88

1.00

No. of ANC visits

1,74

0,53

Number of times child ate solid, semisolid or soft food yesterday

2,59

1,32

Dietary diversity score for child

5,92

3,06

N

(%)

600

50,5

587

49,5

797

67,1

Received BCG (yes)

1120

94,5

Received DPT/Hep B/Influenza 3 (yes)

1037

87,7

Received POLIO 3 (yes)

1013

85,6

Received Measles > = 9 months (yes)

927

86,7

ev

rr

ee

rp Fo

1 2

ie

3

w

Table 2: Characteristics of the sample (N = 1,187) categorical variables

Characteristics

on

Sex of child Male

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Female Still breastfeeding (yes) Use of preventive health service

12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

Taking iron pills, sprinkles or syrup (last 7 days) (yes)

343

29,0

Drugs for intestinal parasites (yes)

438

37,2

Anemia level (some anemia)

718

61,5

343

28,9

267

23,1

173

15,1

262

22,2

557

47,3

740

62,6

802

67,7

383

32,3

394

33,1

793

66,9

Empowerment Participation in decision-making Low participation

Husband justified in beating wife

rp Fo

Less sense of empowerment Wife justified in refusing sex Less Empowered

Water and sanitation

ee

Source of drinking water Unimproved Type of toilet facilities Unimproved

Disposal of youngest child stool inappropriate disposal practice

ie

ev

rr

Religion Christian

w

Other religions Place of residence

on

Urban Rural

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 58

1 2

Bivariate analysis of the association between CCP and HAZ

3

Bivariate analysis was carried out to examine the associations between CCP and

4

child nutritional status. The results show a strong positive association between

5

care practices and child HAZ (Beta = .12, t = 3.73, P 1,

25

and four principal components were extracted that explained 70% of the 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 37 of 58

1

variance. No item had a loading less than .4 (20). Therefore, all the items were

2

used to create the composite care practices score, treated in subsequent analyses

3

as a continuous variable.

4 5

Other variables used in the analysis:

6

(a) maternal age, height, weight, number of antenatal visits (ANC) education,

7 8 9

occupation, anemia level, and parity;

(b), method of disposal of youngest child stools;

ee

10

rp Fo

11

(b) empowerment variables including women’s role in household decision

12

making, opinion regarding wife beating, and attitudes regarding sexual

13

relations with husband;

ie

15

ev

14

rr

(c) household level variables including number of children under 5 years in

w

16

household, Wealth Index, urban/rural place of residence, source of drinking

17

water, religion and type of toilet facilities; and

on

18 19

(e) the child level variables sex and age (child’s age was transformed into age

20

squared and included in regression analyses to account for non-linearity of the

21

age variable (23).

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

22 23

Some of the variables were recoded. Source of drinking water and toilet facilities

24

were recoded according to WHO and UNICEF (24) recommended classifications:

25

‘improved’ and ‘unimproved’ water and ‘improved’ and ‘unimproved’ sanitation 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

1

facilities. The disposal of the youngest child stool was recoded into ‘appropriate’

2

and ‘inappropriate’ disposal methods. Maternal occupation was recoded into

3

‘white collar’ and ‘agriculture/labor’, and religion into ‘Christians’ and ‘other

4

religions’. For the empowerment variables, three indices were created based on

5

the DHS (18) recommended procedure (number of household decision making,

6

opinion regarding wife beating, and justified to refuse sexual intercourse with

7

husband). High scores were coded ‘more empowered’ and low scores ‘less

8

empowered’. The wife beating attitude variable was reversed coded so that a

9

high score corresponded to be more empowered.

ee

10

rp Fo

11

Analytical framework and methods

12

This analysis is framed using the UNICEF conceptual framework in which food,

13

health, and care are posited as the three key pillars influencing child survival,

14

growth and development (1). The model identifies three levels of causes of child

15

under-nutrition: immediate (operating at the individual level), underlying

16

(influencing household and communities) and basic causes (structure and

17

processes of societies). The model suggests that these causal factors affect child

18

nutritional status in a chain-like manner—the basic factors affect the underlying

19

factors, which in turn affect the immediate factors, in turn affecting child

20

nutrition status. The model was extended by Engle and colleagues (25) and the

21

above levels reclassified broadly as context, resources and care giving. This

22

analysis used this framework to structure the hierarchical multiple regression

23

analyses.

w

ie

ev

rr

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 58

24

9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 39 of 58

1

The General Linear Model (GLM) in the SPSS 21 Complex Samples command was

2

used to perform the multivariate analysis. The GLM was used to allow

3

adjustment for survey design effects (sample weight, strata, and cluster). The

4

analysis involved four steps. The first step (model A) contained only the basic

5

characteristics of the mother (age) and child (age and sex), to examine the direct

6

effects of these factors on HAZ. The second step (model B) introduced context

7

variables (place of residence and religion) in the model in the presence of the

8

basic factors to establish how the context variables were directly related to HAZ.

9

The third step (model C) introduces resource variables (education, occupation,

rp Fo

10

anemia level, parity, disposal of youngest child stool, household decision making,

11

opinion regarding wife beating, justified to refuse sexual intercourse with

12

husband, number of children under 5, wealth index, source of drinking water,

13

type of toilet facilities), controlling for basic and contextual factors. In the final

14

step (model D), CCP score was introduced, controlling for basic, contextual and

15

resource factors. Tests of interactions between the CCP score and other predictor

16

variable were undertaken, because previous research has documented that

17

children from poorer households and/or those of mothers with less education

18

may be more likely to benefit more from better care practices, compared to

19

children of wealthier households or those of mothers with better education(6).

w

ie

ev

rr

ee

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

20 21

RESULTS

22

Characteristics of the sample

23

Tables 1 and 2 present the descriptive statistics of the sample. The average age

24

of children used in the analysis was about 20 months. The mean Height-for-Age

25

Z-score for the sample was -1.09 (S.D. = 1.7), while that of weight-for-age and 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

1

weight-for-height Z-scores respectively were -0.81 (S.D. = 1.3) and -0.33 (S.D. =

2

1.5). The average prevalence of stunting, underweight and wasting were 29.1%,

3

16.0%, and 11.5% respectively. The average age of the mothers was 28 years.

4

The number of Antenatal visits was relatively low (1.74 visits). Breastfeeding

5

was generally above average in this population (67%). The average frequency of

6

feeding the child with solid or semi-solid food within 24hours was 2.59.

7

Immunization rates were high among this population. BCG, which is given at

8

birth, was as high as 94%. Additionally, 87.7% of children older than three

9

months had received all their DPT vaccination and 85.6% received polio 3

rp Fo

10

vaccinations. For children older than 9 months, 86.7% received measles

11

vaccination. Fewer children in the sample received iron supplement (29.0%).

12

The use of drugs for intestinal parasites was low (37.2%), probably because the

13

children in the sample were relatively young. With regards to water and

14

sanitation, 22.2% of this population did not have access to improved source of

15

water and 47% used unimproved sanitation facilities. Also, a high proportion of

16

mothers (63%) used inappropriate ways to dispose the youngest child stool.

w

ie

ev

rr

ee

ly

on

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 58

11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

Page 41 of 58

Table 1: Characteristics of the sample (N = 1,187), continuous variables

Variables Mean

SD

Child age

19,8

8,55

Child Height-for-age Z-scores (HAZ)

-1,09

1,72

Child weight-for-age Z-scores (WAZ)

-0,81

1,3

Child weight-for-height Z-scores (WHZ)

-0,33

1,49

Maternal age ( in years)

28,18

6,77

Maternal height (in cm)

1,59

0,07

Maternal weight (in cm)

58,17

10,83

No. of children U5 in household

1,88

1.00

No. of ANC visits

1,74

0,53

Number of times child ate solid, semisolid or soft food yesterday

2,59

1,32

Dietary diversity score for child

5,92

3,06

N

(%)

600

50,5

587

49,5

797

67,1

Received BCG (yes)

1120

94,5

Received DPT/Hep B/Influenza 3 (yes)

1037

87,7

Received POLIO 3 (yes)

1013

85,6

Received Measles > = 9 months (yes)

927

86,7

ev

rr

ee

rp Fo

1 2

ie

3

w

Table 2: Characteristics of the sample (N = 1,187) categorical variables

Characteristics

on

Sex of child Male

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMJ Open

Female Still breastfeeding (yes) Use of preventive health service

12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open

Taking iron pills, sprinkles or syrup (last 7 days) (yes)

343

29,0

Drugs for intestinal parasites (yes)

438

37,2

Anemia level (some anemia)

718

61,5

343

28,9

267

23,1

173

15,1

262

22,2

557

47,3

740

62,6

802

67,7

383

32,3

394

33,1

793

66,9

Empowerment Participation in decision-making Low participation

Husband justified in beating wife

rp Fo

Less sense of empowerment Wife justified in refusing sex Less Empowered

Water and sanitation

ee

Source of drinking water Unimproved Type of toilet facilities Unimproved

Disposal of youngest child stool inappropriate disposal practice

ie

ev

rr

Religion Christian

w

Other religions Place of residence

on

Urban Rural

ly

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 58

1 2

Bivariate analysis of the association between CCP and HAZ

3

Bivariate analysis was carried out to examine the associations between CCP and

4

child nutritional status. The results show a strong positive association between

5

care practices and child HAZ (Beta = .12, t = 3.73, P