Haithar et al.

BMC Pediatrics (2018) 18:289

https://doi.org/10.1186/s12887-018-1261-1

RESEARCH ARTICLE Open Access

Maternal depression and child severe acute

malnutrition: a case-control study from Kenya
S. Haithar1, M. W. Kuria2, A. Sheikh3, M. Kumar1,2* and A. Vander Stoep4

Abstract
Background: Depression is the leading cause of disease-related disability in women and adversely affects the
health and well-being of mothers and their children. Studies have shown maternal depression as a risk factor for
poor infant growth. Little is known about the situation in Sub-Saharan Africa. The aim of our study was to examine
the association between maternal depression and severe acute malnutrition in Kenyan children aged 6–60 months.
Methods: A matched case-control study was conducted in general paediatric wards at the Kenyatta National Hospital.
The cases were children admitted with severe acute malnutrition as determined by WHO criteria. The controls were
age and sex-matched children with normal weight admitted in the same wards with acute ailments. Mothers of the
cases and controls were assessed for depression using the PHQ-9 questionnaire. Child anthropometric and maternal
demographic data were captured. Logistic regression analyses were used to compare the odds of maternal depression
in cases and controls, taking into account other factors associated with child malnutrition status.
Results: The prevalence of moderate to severe depression among mothers of malnourished children was high (64.1%)
compared to mothers of normal weight children (5.1%). In multivariate analyses, the odds of maternal depression was
markedly higher in cases than in controls (adjusted OR = 53.5, 95% CI = 8.5–338.3), as was the odds of having very low
income (adjusted OR = 77.6 95% CI = 5.8–1033.2).
Conclusions: Kenyan mothers whose children are hospitalized with malnutrition were shown in this study to carry
a significant mental health burden. We strongly recommend formation of self-help groups that offer social support,
counseling, strategies to address food insecurity, and economic empowerment skills for mothers of children
hospitalized for malnourishment.
Keywords: Maternal depression, Child malnutrition, Kenya, Case control study, Poverty

Background attention, especially in low and middle income countries

The health and well-being of children is inextricably tied (LMIC) where poor growth due to under-nutrition is a
to their early social and emotional experiences. Since major problem.
feeding and caring for the young is primarily the Globally, nearly 50.6 million children under the age of
mother’s responsibility, poor maternal physical or mental five are malnourished; 90% of these children reside in
health can adversely affect nutrition, health, and psycho- LMIC [5]. Physical growth is a key indicator of child
logical well-being of children [1]. The impact of mater- health and nutritional status [6]. Rapid physical growth
nal mental health on children’s long-term emotional, and development occur in the first two years of life when
cognitive and behavioral problems has been well studied children are the most dependent on caregivers for meeting
in high income countries [2–4]. However, the impact on their nutritional needs [7]. Studies have shown that
child physical health and development has received less healthy maternal behavior and attitude have an essential
role in maintaining healthy nutrition in children [5–7].
* Correspondence: [email protected] Depression is the leading cause of disease-related dis-
1
Department of Psychiatry, College of Health Sciences, University of Nairobi, ability among reproductive aged women, globally [8].
P.O. Box 103140, Nairobi 00101, Kenya The first year after a woman gives birth to an infant is a
2
Research Department of Clinical, Health and Educational Psychology,
University College London, London WC1E 7HB, UK particularly high risk time for the occurrence of
Full list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Haithar et al. BMC Pediatrics (2018) 18:289 Page 2 of 9

depression. Postpartum depression (PPD) prevalence es- and child clinic reported a strong association between ma-
timates vary from 15 to 57% [9–13]. An estimated 10– ternal depression and both non-exclusive breastfeeding
15% of mothers who reside in high income countries are (OR = 7.1; 95% CI = 2.9–17.6) and infant underweight sta-
affected [13], with nearly double the prevalence reported tus (OR = 4.4; 95%CI = 1.8–11.0) [17]. To date no Kenyan
in South Asia (Pakistan 28%, India 23%) [9, 11, 14–16]. studies have examined the association between severe
PPD prevalence estimates from sub-Saharan Africa acute malnutrition (SAM) in infants and depression in
range from 6 to 30% [17]. mothers. By testing the hypothesis that mothers of chil-
Most research on maternal depression and child nutri- dren hospitalized with SAM would have a significantly
tional outcomes in LMIC has been conducted in S. Asia, higher likelihood of suffering from depression than chil-
where the majority of the world’s underweight children dren hospitalized with other medical conditions, our re-
reside [9, 10, 14, 18]. The South Asian studies suggest that search was designed to fill this gap.
poor maternal mental health, particularly maternal depres-
sion, is a risk factor for inadequate growth in young chil- Methods
dren. A case-control study conducted in South India Study design
reported a significant association between current major We carried out a matched case-control study to examine
depression in mothers and malnutrition in children (OR = differences in the prevalence of depression in mothers
3.2, 95% CI 1.1–9.5) [16, 19]. A cohort study conducted by with young children hospitalized for severe acute malnu-
Rahman et al. (2007) [20] in rural Pakistan found that trition (SAM) and mothers with young children hospi-
perinatal depression in mothers predicted poorer growth talized for other health problems.
and higher risk of diarrhea in infants [21].
In Brazil de Miranda et al. (1996) conducted a Sample
case-control study of women of San Paulo and found high Sample size was determined using open Epi formula for
levels of psychiatric morbidity among women with protein matched case control studies. The minimum number
energy malnourished children, with 63% of cases having was determined as 74 (37 cases and 37 controls).
high levels of mental distress compared to 38% in controls Estimates from the Husain et al. study (2000) were used
(OR = 2.8;95% CI =1.2–6.9) [22]. In Rio de Janeiro Hassan to ascertain expected prevalence of maternal depression
et al. (2016) found that maternal mental health was associ- in children with SAM, and the meta-analytic study by
ated with the nutritional status of infants at six months [23]. O’Hara et al. (1996) [13] for the expected prevalence of
Infants of depressed mothers were reportedly two standard maternal depression in normal children [9, 13].
deviations below average height on WHO norms [12, 24].
A number of studies of child nutritional status in Recruitment and consenting procedures
under-fives conducted in sub-Saharan Africa have exam- The study was approved by the Kenyatta National
ined related demographic, socioeconomic and cultural Hospital/University of Nairobi ethical review committee
factors [25–29], while a small literature focuses on ma- (approval no. KNH/ERC/A/180). Consent was adminis-
ternal mental health (Table 1). Adewuya et al. (2008) tered in English or Kiswahili, depending on the mother’s
conducted a case-control study in Nigeria and found language preference, and written informed consent was
that at both three and six months, infants of depressed obtained from the participants. Cases were malnourished
mothers had statistically significantly poorer growth than children ages 6–60 months admitted with severe acute
infants of non-depressed mothers, with odds ratios of malnutrition at Kenyatta National Hospital pediatric
3.28 and 3.34 for length and 3.21 and 4.21 for height wards between May and June 2014. The controls were
[30]. Depressed mothers reported that they discontinued gender and age-matched children who were normal
breastfeeding earlier, and their infants had more episodes weight and admitted to the same hospital for acute ail-
of diarrhea and other infectious illnesses [30]. Other ments. For each case found at the pediatric ward, the
sub-Saharan African studies have shown that maternal first author matched the control on age (up to +/−three
depression is associated with compromised parenting months) and sex (except for two pairs). Sampling for the
behavior, non-responsive care- giving practices and de- cases and controls was done sequentially in all the
crease in breast feeding, all of which contribute to child- pediatric wards at the ratio of one to one. Malnourished
hood malnutrition [17, 21, 22]. Ashaba et al. (2015), who children who were admitted for duration longer than
conducted a case-control study in Uganda, reported that 7 days or not within the age range were excluded.
42% of mothers of malnourished children were de- Mothers who could not communicate in either English
pressed, compared to 12% of mothers of normal weight or Kiswahili or were unable to give informed consent
children admitted to hospital for chronic illness (OR = 2.4; were also excluded. Consecutive convenience sampling
95% CI =1.18–4.79) [31]. A recent cross-sectional study was applied to obtain cases and controls until the de-
conducted in Kenya with mothers attending a maternal sired sample size was achieved.
Table 1 Summary of Key SAM and Maternal Depression Studies in LMIC Contexts
Study Design Sample size & methods Population & setting Tools & mode of administration Outcomes (ORs with 95% CI)
Ashaba et al. (2015) [31] Maternal Matched case N = 166 children (83 cases and 83 Rural population from low MINI (Mini International Prevalence of depression 42% among
depression and malnutrition in control study controls); Controls were age and socioeconomic background. Neuropsychiatric Interview) cases versus 12% among controls
SW Uganda Not blinded gender-matched chronically ill Hospital-based study. Clinician administered. OR 2.4 (95% CI = 1.18–4.79; p = 0.015)
children. Children aged 6–60 months.
Ross & Hanlon et al. (2010) [35] Cohort study N = 954 mother child pairs. Rural population of low SRQ 20 (Self- Reporting Prevalence of High CMD (SRQ20
Perinatal mental distress & infant socio-economic status. Questionnaire) score > 6) was 9.8% in pregnancy,
Haithar et al. BMC Pediatrics (2018) 18:289

morbidity in Ethiopia Population-based study. Self-administered. 2.1% post- natally: persistent high
Followed up from 3rd trimester CMD was 2.5%
through first 2 months Persistent perinatal CMD was
postpartum. associated with RR 2.15 (95%
CI = 1.39–3.24) increased risk of
infant diarrhea.
Ejaz et al. (2012) [37] Maternal Matched case N = 100 (50 cases, 50 controls with Urban population in Karachi HADS (Hamilton Anxiety and Cases were more likely than controls
psychiatric morbidity & childhood control study significant co-morbidities were of low socio-economic Depression Scale) to have depressed mothers OR 0.85
malnutrition in Pakistan Not blinded excluded. status. Clinician administered at time of (95% CI = 0.38–1.86; p = 0.68)
Controls were children with normal Hospital based study. hospital admission.
weight. Admitted with common
childhood illnesses, like acute
respiratory infections, diarrhea.
Rahman et al. (2004) [38] Maternal Case control study N = 172(82 cases, 90 controls) Urban and peri-urban. SRQ 20 (Self- Reporting Strong association between maternal
mental health & childhood Interviewer blinded Controls were children from same Mainly of low SES. Questionnaire), depression and poor weight gain.
growth in Rawalpindi, Pakistan to case-control locality whose weight for age was Immunization clinic based. Self- administered Adjusted OR 2.8 (95% CI 1.2–6.8,
status of infant. above the 10th percentile. Administered to mother when p < 0.05)
she came to clinic for child’s
9-mo. immunization
Patel et al. (2003) [14] Maternal Cohort study 171 infants age > 9 months Rural population in Goa, EPDS (Edinburgh Perinatal Babies under the 5th percentile for
depression & infant growth in Hospitalized 22% with depressed mothers. India of low SES. Depression Scale) weight were more likely to have
Goa, India controls. Hospital based. Clinician administered at depressed mothers Risk ratio 2.3
6–8 week immunization visit. (95% CI = 1.1–4.7, p = 0.01)
Anoop et al. (2004) [19] Maternal Case control study 72 cases and 72 controls, matched. Rural and peri-urban SCID (Structured Clinical Mothers with malnourished babies
depression as risk factor for Interviewer blind Cases were children 50–80% of of low SES Interview for DSM-IV) were more likely to have post- natal
malnutrition in children 6–12 months to child nutritional expected weight. Community based. Clinician administered. depression OR 7.4 (95% CI = 1.6–3.85;
in Kaniyambadi Block, Nadu status. Controls matched for age, sex, and p = 0.01)
locality were > 80% of expected
weight.
Page 3 of 9
Haithar et al. BMC Pediatrics (2018) 18:289 Page 4 of 9

Study procedures available in Kiswahili language. Monahan et al. 2009 [34]

Once consent was given by the mothers, the anthropo- validated the tool in a Kenyan sample.
metric measures of the children were taken using the The PHQ-9 was administered orally by the lead author
normogram to ascertain weight and height. Usually the or one of her medical school assistants when the mother
cases had files which categorized these children as SAM. was unable to complete it (mainly due to poor literacy
Research data were collected on site in the hospital level) or by herself with the choice of filling the Kiswahili
ward. Two medical students who were trained by first or English version. A high percentage of participants pre-
author assisted in the collection of anthropometric data. ferred it to be administered orally due to their low literacy
The mother’s socio-demographic data was captured in level. Quality assurance criteria were instated in training
the study questionnaire, and the mother was invited for the study assistants in collecting maternal depression in-
depression assessment. According to study protocol, all formation including instructing them to stay close to the
mothers whose PHQ-9 scores indicated that they were tonal/semantic reference of the questionnaire.
experiencing severe depression or suicidality were re-
ferred to the KNH mental health unit for treatment. Data analytic approach
The data were entered, cleaned, and analyzed with SPSS
Data collection instruments version 17. Continuous and categorical variables were ana-
Socio-demographic information lyzed using descriptive statistics. Logistic regression ana-
We collected information about family income and size, lyses estimated the ratio of the odds of moderate to severe
maternal age, education, occupation including spouses’, depression (PHQ-9 score ≥ 10) among the cases compared
maternal empowerment and control over finances, HIV to controls. Adjusted odds ratios were calculated to take
status, exposure to chronic illnesses, and family and so- into account other risk factors of malnutrition. In analysis
cial support. of variables with missing data, participants with missing
values were excluded.
Measurement of child’s weight and height to confirm case
and control status Results
Once recruited, the children were undressed, and their We recruited seventy-eight mother-child dyads (39 cases
weights were measured using a digital compression scale and 39 controls) during the study period. The mean ages
and recorded to one decimal point (in kilograms). The of the cases and controls were similar (20.4 months
height of each child was measured (in centimetres) from (SD = 12.2) vs 20.3 months (SD = 20.3)), while the
crown to heel with the child in prone position using a tape mean height (72.8 (SD = 10.4) vs 78.8 (SD = 10.5) and
measure. Weight for height scores was generated from weights (7.2 (SD = 2.2 and 10.1 (SD = 2.8)) of the cases
WHO normograms, and these were used to confirm the were significantly lower than that of the controls with t
case definition. Children were classified according to (76) = − 2.5, p < 0.005 (height) and t (76) = − 5.2,
WHO criteria (severe wasting (< 70% weight for length or p < 0.001(weight). We recruited 19 girls and 19 boys as
< − 3 Z score) and/or oedema [12]. The cases stood out cases and 22 boys and 17 girls as controls.
clearly; therefore there were no changes in classification The participating mothers ranged from 16 to 46 years
after anthropometric measurements were taken, and none of age, with a mean age of 27.7 (SD = 6.4). With regard
of the mothers were excluded from the study. to their marital status, 81.2% of the mothers were mar-
ried; half had had some secondary schooling; and 45.1%
Measurement of maternal depression were employed. Of the children, 35.1% were firstborn.
Maternal depression was assessed using the PHQ-9 For those with siblings, the number ranged from 1 to 5.
which was developed in the U.S. [32]. The PHQ-9 is a As shown in Table 2, mothers of children hospitalized
self-administered depression scale with nine items that with malnutrition had significantly lower levels of family
asks about past two weeks with response options ran- income (X2 = 14.1, df = 2, p = .001) than mothers of chil-
ging from “not at all” to “nearly every day.” The items dren hospitalized with other conditions. Among those
reflect the nine criteria on which the diagnosis of who were married, mothers with children with SAM
DSM-V major depressive disorder is based [32]. In this were more likely to have spouses who were unemployed.
study we used the PHQ-9 to grade depressive symptom There were no statistically significant differences in
severity as none (score of 0–4), mild (5–9), moderate mothers of cases and controls with respect to age, mari-
(10–14), or moderately severe/severe (11–27), as recom- tal status, educational attainment, employment status,
mended by the scale developers [32]. The validity and chronic health conditions, or number of children under
usefulness of the PHQ-9 in East Africa has been dis- age 5 years. Similar proportions of mothers of malnour-
cussed by Gelaye et al. 2014 [33]. The PHQ-9 has been ished children (43.6%) and control children (38.5%) had
widely used in Kenya, and there is a translated version breastfed their infants for 12 months or more. The
Haithar et al. BMC Pediatrics (2018) 18:289 Page 5 of 9

Table 2 Baseline Social and Demographic Characteristics of the Mother-child Dyads

Variable Mothers of Hospitalized Severely Mothers of Hospitalized Normal Difference Significance
Malnourished Children (Cases) Weight Children (Controls) Statistic
Mean age (SD) 28.4 (7.6) 27.0 (5.0) t (75) = 1.032 0.32
Mean number of children under age 5 yrs (SD) 1.2 (0.38) 1.3 (0.57) t (75) = .741 0.46
Marital status N (%)
Single/divorced/ widowed 9 (23.7) 5 (12.8) X2 (1) = 2.110 0.28
Married 29 (76.3) 34 (87.2)
Unknown 1 0
Mother’s education level
None 0 1 (2.6) X2 (3) = 3.24 0.35
Primary 22 (56.4) 15 (38.5)
Secondary 13 (33.3) 18 (46.2)
Post-secondary 4 (10.3) 5 (12.8)
Spouse’s education level*
Primary 10 (34.5) 11 (32.4) X 2 (3) = 1.97 0.37
Secondary 16 (55.2) 15 (44.1)
Post-secondary 3 (10.3) 8 (23.5)
Chronic illnesses (i.e. hypertension, diabetes)
Yes 5 (12.8) 5 (12.8) X2 (1) = 0.00 1.00
No 34 (87.2) 34 (87.2)
Self-reported HIV status
Positive 7 (17.9) 2 (5.1) X2 (2) = 3.35 0.20
Negative 31 (79.5) 35 (89.7)
Unknown 1 (2.6) 2 (5.1)
Duration of breastfeeding
≤12 months 22 (56.4) 24 (61.5) X 2 (1) = 0.21 0.64
> 12 months 17 (43.6) 15 (38.5)
Mother’s occupation
Unemployed 28 (73.7) 24 (61.5) X2 (1) = 1.29 0.25
Employed 10 (26.3) 15 (38.5)
Spouse’s occupation*
Unemployed 15 (51.7) 8 (23.5) X2 (1) = 5.37 0.02
Employed 14 (48.3) 26 (76.5)
Family income per annum
< 36,000 14 (36.8) 1 (2.8) X2(2) 14.15 0.001
36,000–150,000 16 (42.1) 18 (52.8)
> 150,000 8 (21.1) 16 (44.4)
Social support from others
Yes 28 (71.8) 33 (84.6) X2(1) = 1.88 0.17
No 11 (28.2) 6 (15.6)
Mother’s level of control over family finances
Total control 16 (53.3) 14 (45.2) X2(1) = 0.41 0.52
Partial control 14 (46.7) 17 (54.8)
None 9 (23.1) 8 (20.5)
• No data for mothers who are not married (N = 14) or marital status is unknown (N = 1)
Haithar et al. BMC Pediatrics (2018) 18:289 Page 6 of 9

self-reported prevalence of HIV was 17.9% among Table 3 Logistic Regression Model: Depression Status of Mothers
mothers of the cases and 5.1% among mothers of the with Children Hospitalized with Severe Acute Malnutrition or
controls. The majority of the mothers of cases (71.8%) Other Health Conditions
and controls (84.6%) reported that they were receiving Model Variable OR (95% CI) p value
social support from family members or friends. Over 1 Depression Status
three quarters of mothers in both groups reported hav- Depressed 33.0 (6.9–158.2) < 0.001
ing some control over family finances. Not depressed 1.0
The prevalence of mild, moderate, or moderately se-
2 Depression Status
vere depression was 64.1% (N = 25) among mothers of
severely malnourished children. This statistically signifi- Depressed 53.5 (8.5–338.3) < 0.001
cantly higher than the 5.1% (N = 2) prevalence of depres- Not depressed 1.0
sion identified in mothers of normal weight children, Family income per annum
OR = 33.0; 95% CI 6.9–158.2, p < 0.001 (Fig. 1). Among < 36,000 77.6 (5.8–1033.2) 0.001
the 25 case mothers who were depressed, 13 had mild 36,000-150,000 3.3 (0.6–18.0) 0.162
depression, 9 had moderate depression and 3 had mod-
> 150,000 1.0
erately severe depression. In the control group one
mother had mild depression, and the other had moder-
ately severe depression (Fig. 1). discrepancy between the prevalence of depression in
Results of multivariate logistic models showed child these mothers compared to mothers of children hospi-
nutrition to be significantly associated with maternal de- talized with other illnesses.
pression (AOR = 53.5; 95% CI: 8.5–338.3) and low family The hospital-based case control study conducted in
income (AOR = 77.6; 95% CI: 5.8–1033.2). Besides family Pakistan by Ejaz et al. reported high psychiatric morbidity
income, none of the covariates were statistically signifi- of 50% in the cases, but with nearly as high a prevalence
cantly associated with child malnutrition in multivariate of depression (46%) in controls who were mothers of hos-
analyses (see Tables 2 and 3). pitalized normal weight children [37]. This high mental
health morbidity in both cases and controls reflected the
Discussion generally high prevalence of mental health problems
Our study demonstrated that infant malnutrition is sig- amongst women in Pakistan [38]. Although the prevalence
nificantly associated with both maternal depression and of depression in cases in the current study (64.1%) is
family income. Several studies in low income countries higher than estimates of 15–63% reported among mothers
such as India, Pakistan, Ethiopia and Uganda have in other LMICs, what is more striking in our study is the
shown similar findings [19, 21, 31, 35]. In a low prevalence of depression (5.1%) in the controls [14,
meta-analysis of seventeen studies from eleven different 16, 21, 30, 31, 39]. A prior Kenyan study conducted by
countries, Sukran et al. reported an OR of 2.2 in the as- Madeghe et al. (2016) with women with infants attending
sociation between maternal depression and underweight well-child visits reported a PPD prevalence (EPDS score of
and an OR of 2.0 in the association between maternal 10 or higher) of 13% [17].
depression and stunting [36]. Our study findings stand Several features of our study sample and methods may
out in both the high prevalence of depression in mothers have contributed to differences between our findings
of hospitalized malnourished children and in the and those of previous studies of hospitalized children.

Fig. 1 Severity of depression among mothers of cases and controls

Haithar et al. BMC Pediatrics (2018) 18:289 Page 7 of 9

Our study sample was restricted to mothers whose chil- undernutrition in children by compromising parenting be-
dren had been hospitalized for seven days or less, whereas havior. Depression can adversely affect the mother’s ability
the children in Ashaba et al. (2015) were not restricted to to perform caregiving activities such as breast feeding,
those with brief length of hospitalization and, subse- stimulation, hygiene and overall care [27]. This interferes
quently, their control mothers may have been suffering with formation of a secure early attachment and bonding
psychological effects of their children’s long hospital stays behaviors with the baby [19, 21] which, in turn affect a
(as high as 2–3 months) [31]. We only include those chil- child’s physical and emotional well-being. Conversely, hav-
dren admitted fewer than seven days previously in order ing a child who is severely malnourished is highly distres-
to mitigate this potential contributor to maternal distress. sing. In the current study the malnourished children had
The higher prevalence of depression in mothers of our been ill intermittently with general deterioration of health
cases may be due to differences in study populations, with that could trigger sustained psychological distress in the
the current sample being predominantly urban slum mothers. Additionally, the fact that the infant was physic-
dwellers of low socio economic status, while the Ashaba ally extremely fragile, and this was visually apparent to the
et al. sample was mainly rural. Table 1 highlights that a mother as she waited for the infant to recover, could
variety of tools, including the MINI, EPDS, HADS etc., heighten feelings of hopelessness and helplessness in the
were used in different studies. We administered the mother. Children in the hospital wards where the study
PHQ-9 because it has been validated in Kenya. was conducted have high mortality rates with conse-
Our study had several limitations. Although the study quences for the mothers’ level of stress and low mood. In
was adequately powered to evaluate the primary study contrast, the controls may have been ill for a shorter win-
question, the small sample size contributed to the very dow of time, so the mothers may not have been subjected
wide confidence intervals around estimated odds ratios to sustained distress.
for maternal depression and family income. We were not Our study illustrates the juxtaposition of two health
able to draw conclusions about the contribution to child conditions that have serious adverse effects on large seg-
malnutrition status of factors such as mother’s HIV status ments of populations in low income countries. In this
and father’s unemployment status that may have distin- case-control study, we draw attention to tremendous chal-
guished cases from controls in a larger study. Researchers lenges parents face in caring for malnourished children
were aware of the case-control status of the mother at the and the burgeoning challenges children face when their
time they administered the depression questionnaire. A caregivers are debilitated with depression. While our study
high proportion of participating mothers requested that is inconclusive regarding the temporal sequence in the
the questionnaire be administered orally. While the med- causal association between mother’s depression and child’s
ical students were carefully trained to administer the nutritional status, the empirical evidence regarding the eti-
PHQ-9 in a systematic way to both case and control ology of depression would support the argument that
mothers, there may have been errors in understanding the there is considerable reciprocity, with maternal depression
intent of the questions or in the data collectors’ sensitivity, affecting feeding and other child-rearing practices, and the
based on the health status of the child or if the mother stress of caring for a malnourished child affecting the
was perceived as highly distressed. Additionally, because mental health status of the caregiver [31, 40, 41].
the case-control study was organized around the outcome Hospitalization is an added burden on caregiving re-
of the child’s hospitalization, it is difficult to establish tem- sources. By matching cases and controls on the condition
poral sequence between maternal depression and the of hospitalization and by taking family income into ac-
child’s nutritional status. We were not able to determine count in multivariate analyses, our study was able to con-
which mothers in this study suffered from depression trol for these sources of parental stress.
before or during pregnancy. In addition we did not gather Our study findings suggest that mothers of malnour-
information about which children were born preterm or ished children are a very vulnerable group for whom
were underweight at the time of their birth. Knowledge of emotional health support and economic empowerment
the date of onset and temporal ordering of depression in programs are warranted. Mothers of malnourished infants
the mother and malnourishment in the infant would help are discharged from hospital to carry out feeding proto-
to determine the optimal timing for targeting inter- cols that require time, effort, new skills, and financial re-
vention strategies. sources. It may be difficult for mothers with depression
Having a child who is severely malnourished and who is and low income to comply with recommendations. Our
undergoing hospitalization requires high reserves of par- findings strongly suggest that the need for clinicians who
ental energy. From what we know about how depression care for families with malnourished infants should learn
affects functioning, a mother with moderate depression to recognize and treat maternal mental health conditions
will have difficulty in carrying out ordinary work and so- that can impede attainment of desired nutritional goals. In
cial activities. Maternal depression may contribute to addition, the association between malnourishment and
Haithar et al. BMC Pediatrics (2018) 18:289 Page 8 of 9

very low income calls for measures to ensure that families Authors’ contributions
have adequate economic resources so that mothers and The work was carried out by SH as part of the Masters degree in Psychiatry
at the department of Psychiatry University of Nairobi. SH with support from
infants do not suffer the health consequences of having AS collected data and wrote the findings, MK and MK were her University
insufficient food. mentors and helped in conceptualization, writing up and conducting statistical
Promising findings have emerged from a trial of the analysis. AVS was her third mentor who assisted during planning of the
research concept, reviewed the writing and data analysis and helped shape the
WHO endorsed Thinking Healthy Program conducted in manuscript for submission. All authors read and approved the final manuscript.
rural Pakistan, where mothers with depression who re-
ceived cognitive behavioral therapy experienced significant Ethics approval and consent to participate
reduction in depression. Additionally their infants had Ethical approval was obtained from The Kenyatta National Hospital / University of
Nairobi Ethical and Research Committee (KNH/UoN-ERC) Ref. no. KNH/ERC/A/180.
fewer episodes of diarrhea in the first year, compared to The study purpose was explained to the participants. A written informed
women whose depression was not treated [21, 42]. The consent was signed by the respondent, based on willingness to participate
Thinking Healthy Program targets maternal depression in the study. Informed consent was given from participants in this research
for future uses of data, such as publication, preservation and long-term use
and infant health promotion and has been endorsed by of research data. Confidentiality was assured. The information collected was
WHO [43] for use in LMIC. The intervention can be of- kept confident, serial numbers were used instead of names.
fered by lay health workers and could be implemented in
Kenya to provide greater support to vulnerable women. Consent for publication
Not applicable.

Conclusions Competing interests

The authors declare that they have no competing interests.
The prevalence of depression in Kenyan mothers of chil-
dren under five years of age who were hospitalized for
malnutrition was found to be significant. We found ma- Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published
ternal depression in these women significantly and maps and institutional affiliations.
markedly higher than in mothers of children hospital-
ized for other conditions. We strongly recommend for- Author details
1
Department of Psychiatry, College of Health Sciences, University of Nairobi,
mation of hospital-based support and self-help groups P.O. Box 103140, Nairobi 00101, Kenya. 2Research Department of Clinical,
for mothers of children hospitalized with severe acute Health and Educational Psychology, University College London, London
malnutrition. The implementation of WHO endorsed WC1E 7HB, UK. 3Department of Clinical Medicine and Therapeutics, College
of Health Sciences, University of Nairobi, P.O. Box 19676, Nairobi 00202,
Thinking Healthy Program at community and health Kenya. 4Psychiatry & Behavioral Sciences and Epidemiology, 6200 NE 74th
care levels to strengthen mothers’ ability to shoulder Street, Suite 210, Seattle, WA 88115-1538, USA.
and share the heavy burden of rearing children who are
Received: 29 August 2017 Accepted: 20 August 2018
at risk of life-threatening malnutrition may be consid-
ered. In this model, lay health workers including health
facility staff can be trained in basic psychosocial sup- References
port to bolster maternal mental health. 1. Cummings EM, Kouros CD. Maternal depression and its relation to
Children’s development and adjustment. In: Tremblay RE, Boivin M, RDeV P,
editors. Encycl early child dev; 2007. p. 1–10.
Abbreviations 2. Haas JD, Murdoch S, Rivera J, Martorell R. Early nutrition and later physical
DSM: Diagnostic and statistical manual of mental disorders; EPDS: Edinburgh work capacity. Nutr Rev. 2009;54:S41–8. https://doi.org/10.1111/j.1753-4887.
postnatal depression inventory; HADS: Hamilton anxiety and depression 1996.tb03869.x.
scale; LMIC: Lower and middle-income countries; MINI: Mini international 3. Victora CG, Adair L, Fall C, Hallal PC, Martorell R, Richter L, et al. Maternal and child
neuropsychiatric interview; PHQ-9: Patient health questionnaire 9; PPD: Post undernutrition: consequences for adult health and human capital. Lancet (London,
partum depression; SAM: Severe acute malnutrition; WHO: World Health England). 2008;371:340–57. https://doi.org/10.1016/S0140-6736(07)61692-4.
Organization 4. Pollitt E, Gorman KS, Engle PL, Rivera JA, Martorell R. Nutrition in early life
and the fulfillment of intellectual potential. J Nutr. 1995;125(4 Suppl):1111S–
8S. http://www.ncbi.nlm.nih.gov/pubmed/7536831. Accessed 15 July 2017.
Acknowledgements
5. Faruque ASG, Ahmed AMS, Ahmed T, Islam MM, Hossain MI, Roy SK, et al.
Kenneth Mutai assisted with statistical analyses and Winnie Sharon Kiche for
Nutrition: basis for healthy children and mothers in Bangladesh. J Health
help with referencing. The authors wish to acknowledge our participants,
Popul Nutr. 2008;26:325–39. http://www.ncbi.nlm.nih.gov/pubmed/
both children and their mothers.
18831228. Accessed 15 July 2017.
6. Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. Maternal and
Funding child undernutrition: global and regional exposures and health consequences.
The design, field work, and writing for the study were supported through a Lancet. 2008;371:243–60. https://doi.org/10.1016/S0140-6736(07)61690-0.
Medical Education Partnership Initiative (MEPI) Award from the U.S. National 7. Semba RD, Victora CG. Low birth weight and neonatal mortality. In:
Institute of Mental Health/Fogarty International Center, R25 MH099132. Nutrition and health in developing countries. Totowa, NJ: Humana Press;
2008. p. 63–86. https://doi.org/10.1007/978-1-59745-464-3_3.
8. Kessler RC. Epidemiology of women and depression. J Affect Disord. 2003;74:
Availability of data and materials 5–13. http://www.ncbi.nlm.nih.gov/pubmed/12646294. Accessed 15 July 2017.
All personal identifiers have been removed from the data. Original data in 9. Husain N, Creed F, Tomenson B. Depression and social stress in Pakistan.
SPSS data base format will be made available on request. The corresponding Psychol Med. 2000;30:395–402. http://www.ncbi.nlm.nih.gov/pubmed/
author could be contacted using the email provided to procure the data. 10824659. Accessed 15 July 2017.
Haithar et al. BMC Pediatrics (2018) 18:289 Page 9 of 9

10. Gelaye B, Rondon MB, Araya R, Williams MA. Epidemiology of maternal depression, 31. Ashaba S, Rukundo GZ, Beinempaka F, Ntaro M, LeBlanc JC. Maternal depression
risk factors, and child outcomes in low-income and middle-income countries. and malnutrition in children in Southwest Uganda: a case control study. BMC
Lancet Psychiatry. 2016;3:973–82. https://doi.org/10.1016/S2215-0366(16)30284-X. Public Health. 2015;15:1303. https://doi.org/10.1186/s12889-015-2644-y.
11. Mirza I, Jenkins R. Risk factors, prevalence, and treatment of anxiety and 32. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression
depressive disorders in Pakistan: systematic review. BMJ. 2004;328:794–7. severity measure. J Gen Intern Med. 2001;16:606–13. http://www.ncbi.nlm.
http://www.bmj.com/content/328/7443/794. Accessed 15 July 2017. nih.gov/pubmed/11556941. Accessed 15 July 2017
12. WHO child growth standards and the identification of severe acute 33. Gelaye B, Williams MA, Lemma S, Deyessa N, Bahretibeb Y, Shibre T, et al.
malnutrition in infants and children. http://apps.who.int/iris/bitstream/ Validity of the patient health questionnaire-9 for depression screening and
10665/44129/1/9789241598163_eng.pdf?ua=1. Accessed 15 July 2017. diagnosis in East Africa. Psychiatry Res. 2013;210:653–61. https://doi.org/10.
13. O’hara MW, Swain AM. Rates and risk of postpartum depression—a meta- 1016/j.psychres.2013.07.015.
analysis. Int Rev Psychiatry. 1996;8:37–54. https://doi.org/10.3109/ 34. Monahan PO, Shacham E, Reece M, Kroenke K, Ong’or WO, Omollo O, et al.
09540269609037816. Validity/reliability of PHQ-9 and PHQ-2 depression scales among adults
14. Patel V, DeSouza N, Rodrigues M. Postnatal depression and infant growth living with HIV/AIDS in western Kenya. J Gen Intern Med. 2009;24:189–97.
and development in low income countries; a cohort study from Goa, India. https://doi.org/10.1007/s11606-008-0846-z.
Arch Dis Child. 2003;88(1):34–7. 35. Ross J, Hanlon C, Medhin G, Alem A, Tesfaye F, Worku B, et al. Perinatal
15. Afzal S, Khalid R. Social Support and Postnatal Depression in Pakistani mental distress and infant morbidity in Ethiopia: a cohort study. Arch Dis
Context. Pakistan J Soc Clin Psychol. 2014;12(1):34–8. www.gcu.edu.pk/ Child - Fetal Neonatal Ed. 2010;96:F59–64. https://doi.org/10.1136/adc.2010.
Fulltextjour/PJSCS/2014/5.sidra Afzal.pdf 183327 .
16. Rahman A, Iqbal Z, Harrington R. Life events, social support and depression 36. Sukran P, Kennedy C, Hurley K, Black MWHO|. Maternal depression and early
in childbirth: perspectives from a rural community in the developing world. childhood growth in developing countries: systematic review and meta-
Psychol Med. 2003;33:1161–7. http://www.ncbi.nlm.nih.gov/pubmed/ analysis. Bull World Health Organ. 2011;89:545–620. https://doi.org/10.2471/
14580070. Accessed 15 July 2017. BLT.11.088187.
17. Madeghe BA, Kimani VN, Vander Stoep A, Nicodimos S, Kumar M. 37. Ejaz M, Sarwat A, Aisha T. Maternal psychiatric morbidity and childhood
Postpartum depression and infant feeding practices in a low income urban malnutrition. Pak J Med Sci. 2012;28:874–8. http://pjms.com.pk/index.php/
settlement in Nairobi-Kenya. BMC Res Notes. 2016;9:506. https://doi.org/10. pjms/article/viewFile/2747/942. Accessed 15 July 2017.
1186/s13104-016-2307-9. 38. Rahman A, Lovel H, Bunn J, Iqbal Z, Harrington R. Mothers’ mental health and
18. The State of the world’s children 1998. 1998;16–9. https://www.unicef.org/ infant growth: a case-control study from Rawalpindi, Pakistan. Child Care
sowc/archive/ENGLISH/ Health Dev. 2004;30:21–7. https://doi.org/10.1111/j.1365-2214.2004.00382.x.
The%20State%20of%20the%20World%27s%20Children%201998.pdf . 39. Makoka D. The Impact of Maternal Education on Child Nutrition : Evidence
19. Anoop S, Saravanan B, Joseph A, Cherian A, Jacob KS. Maternal depression and from Malawi,Tanzania and Zimbabwe. DHS Work Pap. 2013;84:1–32. www.
low maternal intelligence as risk factors for malnutrition in children: a dhsprogram.com/pubs/pdf/WP84/WP84.pdf
community based case-control study from South India. Arch Dis Child. 2004;89: 40. Martins C, Gaffan EA. Effects of early maternal depression on patterns of
325–9. http://www.ncbi.nlm.nih.gov/pubmed/15033840. Accessed 15 Jul 2017 infant-mother attachment: a meta-analytic investigation. J Child Psychol
20. Rahman A, Creed F. Outcome of prenatal depression and risk factors Psychiatry. 2000;41:737–46. http://www.ncbi.nlm.nih.gov/pubmed/11039686.
associated with persistence in the first postnatal year: prospective study Accessed 15 July 2017.
from Rawalpindi, Pakistan. J Affect Disord. 2007;100:115–21. https://doi.org/ 41. Murray L, Cooper P. Effects of postnatal depression on infant development.
10.1016/j.jad.2006.10.004. Arch Dis Child. 1997;77:99–101. https://doi.org/10.1136/ADC.77.2.99.
21. Rahman A, Iqbal Z, Bunn J, Lovel H, Harrington R. Impact of maternal 42. Rahman A, Patel V, Maselko J, Kirkwood B. The neglected “m” in MCH
depression on infant nutritional status and illness. Arch Gen Psychiatry. programmes - why mental health of mothers is important for child
2004;61:946–52. https://doi.org/10.1001/archpsyc.61.9.946. nutrition. Trop Med Int Heal. 2008;13:579–83. https://doi.org/10.1111/j.1365-
22. de Miranda CT, Turecki G, de Jesus Mari J, Andreoli SB, Marcolim MA, 3156.2008.02036.x.
Goihman S, et al. Mental health of the mothers of malnourished children. 43. Thinking healthy: A manual for psychosocial management of perinatal
Int J Epidemiol. 1996;25:128–33. http://www.ncbi.nlm.nih.gov/pubmed/ depression WHO generic field-trial version 1.0, 2015 Series on Low-
8666480. Accessed 15 July 2017. Intensity Psychological Interventions – 1. http://apps.who.int/iris/
bitstream/10665/152936/1/WHO_MSD_MER_15.1_eng.pdf?ua=1&ua=1.
23. Hassan BK, Werneck GL, Hasselmann MH. Maternal mental health and
Accessed 15 July 2017.
nutritional status of six-month-old infants. Rev Saude Publica. 2016;50:7.
https://doi.org/10.1590/S1518-8787.2016050006237.
24. Patel V, Rahman A, Jacob KS, Hughes M. Effect of maternal mental health
on infant growth in low income countries: new evidence from South Asia.
BMJ. 2004;328:820–3. https://doi.org/10.1136/bmj.328.7443.820.
25. Reed BA, Habicht JP, Niameogo C. The effects of maternal education on
child nutritional status depend on socio-environmental conditions. Int J
Epidemiol. 1996;25:585–92. http://www.ncbi.nlm.nih.gov/pubmed/8671560.
Accessed 15 July 2017.
26. Madise NJ, Matthews Z, Margetts B. Heterogeneity of child nutritional status
between households: a comparison of six sub-saharan African countries.
Popul Stud (NY). 53:331–43. https://doi.org/10.2307/2584703.
27. Gwatkin DR, Rutstein S, Johnson K, Suliman E, Wagstaff A, Amouzou A.
Socio-economic differences in health, nutrition, and population within
developing countries: an overview. https://siteresources.worldbank.org/
INTPAH/Resources/IndicatorsOverview.pdf. Accessed 15 July 2017.
28. Owor M, Tumwine JK, Kikafunda JK. Socio-economic risk factors for severe
protein energy malnutrition among children in Mulago hospital, Kampala.
East Afr Med J. 2000;77:471–5. http://www.ncbi.nlm.nih.gov/pubmed/
12862136. Accessed 15 July 2017.
29. Pongou R, Ezzati M, Salomon JA. Household and Community
Socioeconomic and Environmental Determinants of Child Nutritional Status
in Cameroon. 2006. https://doi.org/10.1186/1471-2458-6-98.
30. Adewuya AO, Ola BO, Aloba OO, Mapayi BM, Okeniyi JAO. Impact of
postnatal depression on infants’ growth in Nigeria. J Affect Disord. 2008;108:
191–3. https://doi.org/10.1016/j.jad.2007.09.013.
 Copyright © 2018. This work is licensed under
http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding
the ProQuest Terms and Conditions, you may use this content in accordance
with the terms of the License.