Independent and combined effects of improved water, sanitation, and hygiene, and improved complementary feeding, on stunting and anaemia among HIV-exposed children in rural Zimbabwe: a cluster-randomised controlled trial

Background: Children exposed to HIV have a high prevalence of stunting and anaemia. We aimed to test the effect of improved infant and young child feeding (IYCF) and improved water, sanitation, and hygiene (WASH) on child linear growth and haemoglobin concentrations. Methods: We did a cluster randomised 2 × 2 factorial trial in two districts in rural Zimbabwe. Women were eligible for inclusion if they permanently lived in the trial clusters (ie, the catchment area of between one and four village health workers employed by the Zimbabwean Ministry of Health and Child Care) and were confirmed pregnant. Clusters were randomly allocated to standard of care (52 clusters); IYCF (20 g small-quantity lipid-based nutrient supplement daily for infants from 6 months to 18 months, complementary feeding counselling with context-specific messages, longitudinal delivery, and reinforcement; 53 clusters); WASH (ventilated, improved pit latrine, two hand-washing stations, liquid soap, chlorine, play space, and hygiene counselling; 53 clusters); or IYCF plus WASH (53 clusters). Participants and fieldworkers were not masked. Our co-primary outcomes were length for age Z score and haemoglobin in infants at 18 months of age. Here, we report these outcomes in the HIV-exposed children, analysed by intention to treat. We estimated the effects of the interventions by comparing the two IYCF groups with the two non-IYCF groups and the two WASH groups with the two non-WASH groups, except for outcomes with an important statistical interaction between the interventions. The trial is registered at ClinicalTrials.gov (NCT01824940) and is now complete. Findings: Between Nov 22, 2012, and March 27, 2015, 726 HIV-positive pregnant women were included in the trial. 668 children were evaluated at 18 months (147 from 46 standard of care clusters; 147 from 48 IYCF clusters; 184 from 44 WASH clusters; 190 from 47 IYCF plus WASH clusters). Of the 668 children, 22 (3%) were HIV-positive, 594 (89%) HIV-exposed uninfected, and 52 (8%) HIV-unknown. The IYCF intervention increased mean length for age Z score by 0·26 (95% CI 0·09–0·43; p=0·003) and haemoglobin concentration by 2·9 g/L (95% CI 0·90–4·90; p=0·005). 165 (50%) of 329 children in the non-IYCF groups were stunted, compared with 136 (40%) of 336 in the IYCF groups (absolute difference 10%, 95% CI 2–17); and the prevalence of anaemia was also lower in the IYCF groups (45 [14%] of 319) than in the non-IYCF groups (24 [7%] of 329; absolute difference 7%, 95% CI 2–12). The WASH intervention had no effect on length or haemoglobin concentration. There were no trial-related adverse or serious adverse events. Interpretation: Since HIV-exposed children are particularly vulnerable to undernutrition and responded well to improved complementary feeding, IYCF interventions could have considerable benefits in areas of high antenatal HIV prevalence. However, elementary WASH interventions did not lead to improvements in growth. Funding: Bill & Melinda Gates Foundation, UK Aid, Wellcome Trust, Swiss Development Cooperation, US National Institutes of Health, and UNICEF. The SHINE trial design has been reported previously;10, 11 the protocol and statistical analysis plan are available online. Briefly, SHINE was a cluster-randomised community-based 2 × 2 factorial trial done in two contiguous rural districts in Zimbabwe with 15% antenatal HIV prevalence. Clusters were defined as the catchment area of 1–4 village health workers employed by the Ministry of Health and Child Care. Village health workers did prospective pregnancy surveillance and established date of last menstrual period among pregnant women, and referred pregnant women to SHINE research nurses for trial enrolment. Women were eligible if they permanently resided in a study cluster and were confirmed pregnant. Over the recruitment period, the cutoff of gestational age for recruitment eligibility was gradually liberalised to maximise recruitment (appendix). The Medical Research Council of Zimbabwe and the Institutional Review Board of the Johns Hopkins Bloomberg School of Public Health approved the study protocol. All participants provided written informed consent. Clusters were allocated (1:1:1:1) to one of four treatment groups: standard of care, IYCF, WASH, or IYCF plus WASH at a public event. A highly constrained randomisation technique achieved balance across groups for 14 variables related to geography, demography, water access, and sanitation coverage (appendix).12 Masking of participants and fieldworkers was not possible, but investigators analysing the data were masked to group allocation. Interventions were informed by extensive formative research and piloting.10 All women were scheduled to receive 15 modules delivered by group-specific village health workers (VHWs), with behaviour-change messages and interactive tools between enrolment and 12 months postnatal (approximately one visit per month); other family members were encouraged to participate. At each visit, previous information was reviewed before introducing new information to create a sequenced integrated longitudinal intervention. Between 13 and 17 months, VHWs visited monthly, providing routine care and delivering intervention supplies; during these visits VHWs encouraged participants to practise relevant behaviours, although structured modules were not implemented (see key messages and supplies in the appendix, and lesson plans and interactive tools online). Standard of care messages comprised promotion of exclusive breastfeeding to 6 months, uptake of antenatal and neonatal care, prevention of mother-to-child HIV transmission, immunisations, family planning, and standard IYCF information based on WHO recommend-ations. Groups with the IYCF component received all standard of care messages plus information about the importance of nutrition for infant health, growth, and development; feeding nutrient-dense food and a 20 g small-quantity lipid-based nutrient supplement (SQ-LNS; Nutriset, Malaumay, France) daily from age 6 months to 18 months; processing locally available foods to facilitate mastication and swallowing; feeding during illness; and dietary diversity. The IYCF modules therefore addressed specific contextual barriers through a sequential longitudinal intervention based on successive messages and reinforcement. VHWs also made monthly deliveries of 30 20 g sachets of small-quantity lipid-based nutrient supplement from infant age 6 months through to 18 months. The WASH component included all standard of care messages plus information about safe disposal of faeces; hand-washing with soap after faecal contact and before preparing food, eating food or feeding children; protection of infants from geophagia and animal faeces ingestion; chlorination of drinking water; and hygienic preparation of complementary food. Additionally, a ventilated improved pit latrine was provided within 6 weeks of enrolment; two hand-washing stations, plastic mat and play yard (North States, Minneapolis, MN, USA), and monthly delivery of soap and chlorine (WaterGuard, Nelspot, Zimbabwe) were provided. A latrine was constructed in the non-WASH groups following trial completion. Research nurses made home visits at baseline (approximately 2 weeks after consent), 32 weeks’ gestation, and at 1 month, 3 months, 6 months, 12 months, and 18 months post partum to assess maternal and household characteristics and trial outcomes. At baseline, mothers had height, weight, and mid-upper arm circumference measured, and were tested for haemoglobin concentrations (Hemocue, Ängelholm, Sweden), Schistosoma haematobium infection (by urinary microscopy), and HIV. HIV-positive women were encouraged to seek immediate antenatal care to prevent mother-to-child transmission. Other maternal and household characteristics were assessed, including dietary diversity, food insecurity, household wealth, and maternal capabilities.13 Infant birth date, weight, and delivery details were transcribed from health facility records. The trial provided Tanita BD-590 infant scales (Weigh & Measure, Olney, MD, USA) to all health institutions in the study area and trained facility staff. Gestational age at delivery was calculated from last menstrual period dates. At the 18-month postnatal visit (trial endpoint), mothers and infants were visited anywhere in the country for the intention-to-treat analyses of primary outcomes; however, given the household-based nature of the interventions, intermediate visits were done only when the mother was available in the household where she consented. At 18 months postnatal, infant point-of-care haemoglobin concentration was measured (HemoCue, Ängelholm, Sweden). Infant length was calculated as the median of three measurements; weight, head circumference, and mid-upper arm circumference were also measured (appendix). Infant diarrhoea (three or more loose or watery stools in 24 h), dysentery (stool with blood or mucus), and acute respiratory infection (fast or difficult breathing) were assessed by 7 day maternal recall at postnatal visits. Infants with acute malnutrition or illness were referred to clinics. Adverse events and serious adverse events were ascertained by research nurses during visits, and by village health workers during intervention delivery contacts, and reported to a senior research nurse who collected details. Events were reviewed by the study physician (AJP) to determine relatedness to trial interventions before reporting to the responsible institutional review boards. An independent data safety and monitoring board comprising two physicians from Zimbabwe and a statistician from the UK reviewed interim adverse event data. Mothers were tested for HIV status at baseline with a rapid test algorithm (Determine HIV 1/2 test [Alere International Limited, Ballybrit, Ireland], followed by INSTI HIV 1/2 test [bioLytical Laboratories Inc., Richmond, BC, Canada] if positive). HIV-positive women had CD4 counts measured (Pima Analyser [Alere International Limited, Ballybrit, Ireland]) and were referred to local clinics. Viral load was not measured in the trial. National guidelines for prevention of mother-to-child HIV transmission changed from WHO Option B (maternal antiretrovial therapy [ART] from 14 gestational weeks until the end of breastfeeding) to Option B+ (lifelong ART for all pregnant and breastfeeding women) in November, 2013. Women were encouraged to initiate co-trimoxazole and ART, to exclusively breastfeed, and to attend clinic at 6 weeks post partum for early infant diagnosis and initiation of infant co-trimoxazole. Women testing HIV-negative at baseline were retested at 32 gestational weeks and 18 months post partum to detect sero-conversion. HIV-positive mothers were invited to enrol in a substudy, in which infant blood was collected at 1 month, 3 months, 6 months, 12 months, and 18 months and tested for HIV; infants of mothers declining substudy enrolment were only tested at 18 months. Children were classified as HIV-positive or HIV-exposed uninfected based on results at 18 months, or their last available test. Children not tested at 18 months owing to caregiver refusal, defaulted visits, or loss to follow-up were classified as HIV-unknown. Inconclusive or discordant results were retested to confirm status; if no further samples were available or repeat testing was inconclusive, children were classified as HIV-unknown. Before 18 months of age, HIV was diagnosed by DNA PCR on dried blood-spot samples or RNA PCR on plasma; and after 18 months, by PCR or rapid test algorithm, depending on samples provided. All HIV-positive children were referred for ART initiation. The coprimary outcomes were length for age Z score and haemoglobin concentration in infants at age 18 months (allowable age range 76–130 weeks). Secondary outcomes were stunting (length for age Z score <–2), severe stunting (length for age Z score <–3), anaemia (haemoglobin <105 g/L), severe anaemia (<70 g/L); weight for age Z score, underweight (weight for age Z score <–2), weight for length Z score, wasting (weight for length Z score <–2), mid-upper arm circumference for age Z score and head circumference for age Z score at 18 months; 7-day maternal recall of diarrhoea, dysentery, and acute respiratory infection at 12 months and 18 months; and all-cause mortality up to 18 months. Intervention uptake was assessed at all visits and reported here for the 12-month visit. Sample size calculation was done for HIV-unexposed infants.11 No specific sample size calculation for outcomes among HIV-exposed infants was done. All analyses were done on an intention-to-treat basis at the child level. For primary analyses, we used generalised estimating equations that accounted for within-cluster correlation and contained two dummy variables representing the main effect of the IYCF intervention (the two IYCF-containing groups compared to the two groups without IYCF) and the WASH intervention (the two WASH-containing groups compared to the two groups without WASH), unadjusted for other covariates, with an exchangeable working correlation structure.11 Although the study was not powered to detect a statistical interaction between the IYCF and WASH interventions, we estimated these interactions for each outcome. When the interaction was significant (ie, p<0·05 according to the Wald test) or had a sizeable point estimate (ie, RR >2 or <0·5 when comparing ratio-of-ratios, or difference-of-differences >0·25 SDs when comparing continuous outcomes), results are based on a regression model with three dummy variables to represent IYCF, WASH and IYCF plus WASH compared to standard of care instead of the model of two terms. In adjusted analyses we controlled for prespecified baseline covariates, which were initially assessed in bivariate analyses to identify those with an important association with the outcome (ie, p<0·2 or RR >2·0 or <0·5 for dichotomous outcomes, and p<0·2 or difference >0·25 SDs for continuous outcomes). Selected covariates were entered in a multivariable regression model; a forward stepwise selection procedure was implemented with p<0·2 to enter. A log-binomial specification was used to facilitate estimation of relative risks (RR). Depending on the analysis, other methods for comparing groups while accounting for within-cluster correlation included multinomial and ordinal regression models with robust variance estimation, and Somers' D for medians. In a per-protocol analysis, we examined the effect of the interventions when behaviour-change modules were delivered at high fidelity (which was predefined for the IYCF plus WASH group as receiving all ten core modules and for the other study groups as receiving all modules scheduled at the same timepoints when IYCF plus WASH core modules were delivered). A prespecified subgroup analysis of primary outcomes by infant sex was planned if the interaction terms were significant (p<0·05). A sensitivity analysis excluded children testing HIV-positive or HIV-unknown at 18 months. We used Stata (version 14) for all analyses. The study is registered with ClinicalTrials.gov, number {"type":"clinical-trial","attrs":{"text":"NCT01824940","term_id":"NCT01824940"}}NCT01824940. The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.