BMJGH 2020 003508

Original research
Effect of a home-visiting parenting
BMJ Glob Health: first published as 10.1136/bmjgh-2020-003508 on 29 January 2021. Downloaded from http://gh.bmj.com/ on August 11, 2021 by guest. Protected by copyright.
program to promote early childhood
development and prevent violence: a
cluster-randomized trial in Rwanda
Sarah KG Jensen,1 Matias Placencio-Castro,2 Shauna M Murray,1
Robert T Brennan,1,3 Simo Goshev,4 Jordan Farrar,1 Aisha Yousafzai,5
Laura B Rawlings,6 Briana Wilson,6 Emmanuel Habyarimana,7 Vincent Sezibera,8
Theresa S Betancourt1
To cite: Jensen SKG, ABSTRACT

Placencio-Castro M, Murray SM, Key questions
Introduction Families living in extreme poverty require
et al. Effect of a home-visiting
interventions to support early-childhood development
parenting program to promote What is already known?
(ECD) due to broad risks. This longitudinal cluster
early childhood development ►► Early child development interventions can improve
and prevent violence: a randomised trial examines the effectiveness of Sugira developmental outcomes of children in severe
cluster-randomized trial in Muryango (SM), a home-visiting intervention linked to poverty.
Rwanda. BMJ Global Health Rwanda’s social protection system to promote ECD and ►► Common risk exposures affecting child development
2021;6:e003508. doi:10.1136/ reduce violence compared with usual care (UC). in low-resource settings include malnutrition, paren-
bmjgh-2020-003508 Methods Families with children aged 6–36 months tal stress and poor mental health, violence and poor
were recruited in 284 geographical clusters across three hygiene.
Handling editor Valery Ridde districts. Cluster-level randomisation (allocated 1:1 SM:UC)
►► Additional material is was used to prevent diffusion. SM was hypothesised to What are the new findings?
published online only. To view improve child development, reduce violence and increase ►► Sugira Muryango is a home-visiting-based parenting
please visit the journal online father engagement. Developmental outcomes were intervention that addresses known risks to child de-
(http://dx.doi.org/10.1136/ assessed using the Ages and Stages Questionnaire (ASQ- velopment in a low-resource setting.
bmjgh-2020-003508) ►► We reach the poorest families in Rwanda by lay-
3) and the Malawi Development Assessment Tool (MDAT)
and anthropometric assessments of growth. Violence was ering a parenting intervention into an ongoing
government-organised social protection programme.
Received 21 July 2020 assessed using questions from UNICEF Multiple Indicators
Revised 3 November 2020 Cluster Survey (MICS) and Rwanda Demographic and What do the new findings imply?
Accepted 23 December 2020 Health Surveys (DHS). Father engagement was assessed ►► This trial demonstrates that a scalable home-visiting
using the Home Observation for Measurement of the intervention delivered by lay workers to recipients
Environment. Blinded enumerators conducted interviews of a poverty-targeted social protection programme
and developmental assessments. can promote early-childhood development (ECD) and
Results A total of 541 SM families and 508 UC families prevent family violence when conducted within a ro-
were enrolled and included in the analyses. Study attrition bust system of supervision.
(2.0% children; 9.6% caregivers) was addressed by ►► We add to the evidence for the efficacy and effec-
hot deck imputation. Children in SM families improved tiveness of lay worker delivered community-based
more on gross motor (d=0.162, 95% CI 0.065 to 0.260), approaches to promotion of ECD using a strong,
communication (d=0.081, 95% CI 0.005 to 0.156), well-powered design.
problem solving (d=0.101, 95% CI 0.002 to 0.179) and
personal-social development (d=0.096, 95% CI −0.015
© Author(s) (or their to 0.177) on the ASQ-3. SM families showed increased INTRODUCTION
employer(s)) 2021. Re-use father engagement (OR=1.592, 95% CI 1.069 to Families living in extreme poverty require
permitted under CC BY-NC. No 2.368), decreased harsh discipline (incidence rate ratio,
commercial re-use. See rights attention to risks to healthy child devel-
IRR=0.741, 95% CI 0.657 to 0.835) and intimate partner
and permissions. Published by opment, such as low stimulation, poor
BMJ.
violence (IRR=0.616, 95% CI:0.458 to 0.828). There were
hygiene and nutrition, violence and neglect.
no intervention-related improvements on MDAT or child
For numbered affiliations see
growth.
Violence is a major factor affecting healthy
end of article. development, but family violence is rarely
Conclusion Social protection programmes provide a
Correspondence to means to deliver ECD intervention. addressed in early- childhood development
Dr Theresa S Betancourt; Trial registration number NCT02510313. (ECD)-focused interventions.1 Moreover,
Theresa.betancourt@bc.edu the typical focus of ECD interventions is on
Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 1

BMJ Global Health
mothers,2 limiting the ability to work with the full home
did not examine outcomes related to children’s devel-
environment, including male caregivers. Despite an opmental milestones.16 Finally, a quasi- experimental
increasing call to move beyond focusing on mothers in study using matching of counties across two provinces
initiatives to promote caregivers’ capabilities to provide in China found positive effects of a parenting interven-
enriched environments for children,3 few intervention that integrated components from the nurturing
tion studies have focused on integrating promotion of care framework with violence prevention on suspected
father engagement2 and family violence prevention4 developmental delay, and found that these effects were
into programmes that focus on common ECD elements partially accounted for by improvements in positive disci-
such as nutrition, early learning and nurturing care. pline, cognitive stimulation and child growth.17 Looking
ECD interventions are broadly defined as program- at the previous literature, evidence from even the more
matic efforts aimed at supporting healthy child devel- rigorous trails has been somewhat mixed. For example,
opment across domains of growth, developmental mile- two studies including the CRT from South Africa had
stones, cognition, language and socioemotional devel- significant findings in only one area (ie, ECD or violence
opment. Violence prevention may include education reduction).14 15
on harmful effects of direct and observed violence on Programmes targeting ECD and violence reduction have
children, and introduction of nonviolent strategies for yet to be evaluated through rigorous, large-scale studies
resolving parental disputes and disciplining children. employing randomised designs to add to the evidence of
Previous home-visiting interventions conducted in low impact on child outcomes. Government social protection
and middle- income countries demonstrate the value programmes present a potentially innovative platform
of integrated approaches that link parenting interven- for integrating home-visiting interventions to promote
tions to other ECD- promoting programmes including ECD interventions, improving nurturing care, health and
economic strengthening programmes such as cash nutrition, and reducing violence, thus providing a vehicle
transfer programmes,5 6 nutritional interventions,7–9 for targeting poor and vulnerable populations, and
primary healthcare10 and maternal mental health inter- combining income support with a focus on behaviours
ventions.11 However, evidence regarding interventions and investments in human capital.5 The Rwandan
that combine a focus on improving ECD outcomes and government established a social protection programme
reducing violence against children is sparse or of low called the ‘Vision 2020 Umurenge Programme’ in 2007
quality. A 2018 systematic review identified six studies to address poverty and human-capital related disparities
up to 2014.12 Only two of these studies involved rando- by offering direct support to poor families with young
misation, and both the randomised trial (n=45) and the children and public works- based support.18 Moreover,
cluster randomised trial (CRT) (n of clusters=5) lacked in 2017, the government established the National Early
sufficient numbers for randomisation to assure balance Childhood Development Programme to coordinate and
across treatment and control conditions. We located an expand an array of ECD initiatives, including ECD centres
additional five studies from January 2015 to May 2020 and home-visiting interventions for the most vulnerable
that featured both ECD and child violence outcomes.13–17 families. In this paper, we evaluate the effectiveness of the
Among these recent studies, the rigour of the study design Sugira Muryango (SM) (Strengthen the Family) interven-
varied considerably. Two trials were randomised, but had tion, a relatively brief (12 sessions delivered weekly over
insufficient enrolment to assume balance through rando- 3 months) comprehensive home- visiting intervention
misation alone, and no stratification or matching was for families with young children to promote ECD and
employed.13 14 More specifically, one small randomised reduce family violence as implemented by community-
controlled trial of the Triple P-positive parenting inter- based coaches and delivered in combination with Rwan-
vention in Indonesia observed significant improvements da’s social protection programme compared with those
in parenting practices, parental stress and child behav- receiving usual care (UC) and social protection. We
ioural problems but did not examine developmental have previously reported intervention effects on care-
milestones.13 A randomised trial from Liberia found that giver behaviours immediately postintervention.19 This
a parenting intervention successfully reduced harsh disci- paper reports on results from a 12-month follow-up. In
pline and improved caregiver–child interactions, but did line with our theory of change (figure 1), we hypothesise
not observe any improvements in child health or devel- that participation in SM will be associated with improved
opment.14 Two studies used a cluster randomised design, child development outcomes 1 year after the intervention
each with 24 clusters. One of these was a home-visiting ended because previously reported changes in caregiver
programme enrolling pregnant women in South Africa, behaviours have succeed in building a healthy, stimu-
and used matching at the cluster level. This study found lating and safe environment that supports children’s
that the intervention had a significant effect on child achievement of developmental milestones. Primary
health, growth and language, but did not find significant outcomes examined at 12-month follow-up are therefore
intervention effects on violence in the home.15 The other children’s developmental milestones, father engagement
CRT, which was set in Jamaica, was a preschool-based in childcare and violence reduction (harsh discipline,
programme that did not use matching and found positive maternal victimisation to intimate partner violence and
intervention effects on teacher and child behaviours, but paternal perpetration of violence). Given the emphasis
2 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

BMJ Global Health
Figure 1 Sugira Muryango theory of change. Reprinted with remission from Betancourt et al24-https://www.ncbi.nlm.nih.gov/
pmc/articles/PMC7201751/.
on father engagement and violence reduction, we also provided cash for intensive manual labour, or the newer
hypothesise that such effects will be sustained 1 year after expanded public works, which provided cash for more
the intervention ended. Measures related to child growth flexible, less strenuous work and thus all belonged to
were included as primary outcomes in the clinical trial the most extreme level of poverty in the government’s
registration, but later redefined as secondary outcomes in household-ranking system, Ubudehe 1; (2) had one or
the study protocol in recognition that even programmes more child(ren) aged 6–36 months and (3) were willing
that focus on and provide nutritional supplementation to engage in a home-visiting intervention. The age range
struggle to achieve improvements on child growth. of 6–36 months was defined to supplement existing ECD
service for children in Rwanda, which are most widely
available for children aged 5 years and older. Exclusion
METHODS
criteria for caregivers were a severe crisis such as psychosis
Study design
or suicide attempts or severe cognitive impairment which
We used a stratified CRT design to test SM’s effect on
would prevent the caregiver from responding to the inter-
children’s motor, cognitive, language and social develop-
views. All adult caregivers gave written informed consent
ment, and physical growth, as well as family-level violence
for themselves and their eligible children. The primary
and father engagement in care among families enrolled
caregiver was defined as the caregiver who self-identified
in the social protection programme. The use of a CRT
as knowing the child best, most often the mother.
over simple randomised assignment of families was to
reduce the threat of diffusion of elements of SM within
the setting of Rwandan villages. The CRT was conducted Randomisation and masking
within three Rwandan districts selected based on the Families were enrolled between 9 February 2018 and 27
presence of the social protection programme and paying April 2018 (see figure 2 for timeline). Government staff
attention to other ongoing interventions in order to in Nyanza, Ngoma and Rubavu districts provided lists of
minimise overlap with other large-scale ECD interven- households that were eligible for the government’s public
tions that could bias results. The study protocol can be works programme. Geographically defined clusters were
found with the clinical trials registration cited above. created as non-overlapping and comprising at least 30
Both coaches and enumerators were blind to the family’s families identified as eligible for the classic public works
intervention status at the time of baseline assessments. programme or at least ten families identified as eligible
Enumerators were also not informed about the family’s for the newer expanded public works programme, with
assignment status during the two post-intervention assess- some clusters containing both ≥30 classic public works
ments. and ≥10 expanded public works- eligible households.
Clusters were formed from one or more contiguous
Participants villages so that one coach could provide services to all
Families were eligible for inclusion in the study if they: eligible families in that cluster maintaining a caseload of
(1) were eligible for either the classic public works, which five families per coach. Within the same cluster, villages

BMJ Global Health
Figure 2 Study timeline.
were selected to be as proximate as possible and as distal works only) and geographical sector. By strata, clusters
from other clusters as possible, such that after random were assigned random numbers and placed on a ranked
assignment to SM or UC, risk of diffusion to UC fami- list. On the randomly ranked list, the first half of clusters
lies would be minimised. Given the limited reach of were assigned to SM. In case of an uneven number of
expanded public works programming at the time of the clusters per strata, randomisation was used to round the
CRT, 100% of clusters containing at least 10 expanded number assigned to SM up or down. After cluster assign-
public works families were sampled for participation ment, households were invited to participate in the study.
in the study. Clusters containing classic public works We retained clusters if at least five families in the classic
families (including combined clusters) were randomly public works strata or at least one family in the expanded
sampled for inclusion in the CRT until the target sample public works strata enrolled and had at least one child
size of ≥1040 households. Randomisation was completed aged 6–36 months. The final sample encompassed 198
at the cluster level within strata defined by public works clusters: 48 expanded PW- only clusters, 38 expanded
type (expanded public works only, combined expanded PW/classic PW clusters, and 112 classic PW-only clusters
public works/classic public works and classic public were retained (figure 3). Given the ongoing roll-out of
Figure 3 Cluster sampling strategy and flow chart of participants in the Sugira Muryango trial. Although each cluster had a
50% chance of being assigned to receive Sugira Muryango, we were not guaranteed an equal number of Sugira Muryango and
usual care clusters because randomisation occurred within relatively small strata that sometimes contained an odd number of
clusters. ePw, expanded public works; cPW, classic public works.

BMJ Global Health
the expanded public works programme during the design coaches at a pace of about one module per week. Each
phase for the CRT, it was not possible to have expanded module takes an average of 60 min.
public works families make up half of our sample, thus, Home- visiting modules facilitated by the coaches
calculations were based on an assumption of 91 expanded involved participation of female and male caregivers (as
public works clusters and 104 classic public works clusters present) in interaction with their child(ren) whereby
with five households per cluster assigned to SM and UC ‘serve and return’ (ie, responsive) interactions were
conditions. After the calculation, further adjustments pointed out and coached to encourage responsive and
were made, by adding combined clusters of expanded stimulating parent–child interactions. Other caregivers
public works clusters and classic public works randomly and children in the household were welcome to partic-
allocated to condition. Adding more classic public works ipate. Sessions were facilitated in participating families’
clusters allowed us to maintain power to test SM versus homes, unless contraindicated due to privacy concerns.
UC on the primary hypotheses. The CRT enrolled a total Three- month and 6- month booster sessions occurred
of 1049 households. Randomisation by cluster allocated from 18 November 2018 to 7 December 2018 and 4
n=508 families to UC and n=541 families to SM. The March to 30 March 2019, respectively. The aim of each
overall average number of households per cluster was booster visit was to reconnect with families, identify and
6.2 (SD: 2.4), with an average number of households per address ongoing challenges, and engage caregivers in an
cluster of 6.0 (SD: 2.2) for UC families and 6.3 (SD: 2.6) ‘active play’ session as modelled in each home-visiting
for SM families. session (described below); each booster visit was approx-
imately 1 hour.
Intervention component The coaches delivering SM were selected from the local
The intervention was implemented from 7 May 2018 to community using a three-step process: (1) nomination
14 September 2018. The intervention was staggered by from community members; (2) a phone screening and
district to ensure it started within 2 weeks of baseline data (3) an intensive in-person interview in which applicants
collection. SM is a home-visiting-based intervention that were given a curriculum vignette of a challenging family
uses psychoeducation and active coaching of caregivers situation and asked to explain the vignette, instructed
to promote responsive caregiving, nutrition, hygiene and to explain an image shown from the curriculum, which
nonviolent interactions among household members (see depicted family unity, nutrition, father engagement and
figure 1 and online supplemental table 1 for overview of early stimulation, and administered a writing test to
intervention content). Active participation and engage- capture demographic information and answer brief ques-
ment of both female and male caregivers (as present in tions to identify any prior experience they may have had
the home) in childcare and household-related decision delivering community- based programmes (see online
making was encouraged throughout the programme. supplemental table 2 for detail on training, supervision
Each visit included a 15-min active play session, in which and incentive practices). Prior to the start of the interven-
caregivers received live feedback on parent–child interaction, coaches participated in a 3-week training (120 hours),
tions to support and enhance responsive care following followed by a multilayered approach to supervision that
the UNICEF and the WHO Care for Child Development included in-person supervision during the first 3 weeks
package20 and Nurturing Care Framework.21 Flexible of programme delivery, weekly telephone supervision
scheduling and messaging during the visits encouraged (approximately 12 hours total) and monthly in-person
engagement of male caregivers in play and nurturing group supervision. Additionally, weekly in-person peer
care along with female caregivers. The coaches also support groups, facilitated by lead coaches, were held to
helped families navigate formal resources such as govern- complement supervision strategies and promote transfer
ment programmes to promote child health and nutrition of audiorecordings for fidelity monitoring using lead
including supplemental nutrition for malnourished chil- coaches’ computers (approximately 15 hours total).
dren and informal supports such as those from neigh- Coaches received training on confidentiality and risk of
bours and extended family to address issues such as family harm protocols. Coaches received a monthly stipend of
conflict and housing insecurity. The core evidence-based RWF28 000 (caseload of five households).
parenting curriculum, which makes up the backbone of Families in both the intervention households and the
the SM intervention, was originally developed and tested UC condition were eligible for the social protection
in HIV/AIDS- affected Rwandan families with school- public works programme and services as usual from the
aged children, which included nonviolent parenting and Rwandan government and its partners. There was no
caregiver conflict resolution strategies.22 During previous incentive associated with participation in SM but all fami-
pilot studies,19 23 a non- HIV/AIDS- specific version of lies (SM and UC) received a stipend (RWF5000 equal to
the SM home- visiting programme with an ECD focus 3 kg of rice) after each data collection time point. Data
was developed by integrating UNICEF/WHO Care for collection was carried out by trained local enumerators
Child Development20 materials with additional framing working for an independent research firm, who were
consistent with the WHO and UNICEF Nurturing Care blind to the intervention status of the families. Enumer-
Framework.21 SM comprises 12 modules (see online ators signed confidentiality agreements embedded in
supplemental table 1 for details) that are delivered by their contracts and were trained to identify cases of risk

BMJ Global Health
of harm. Assessments were conducted in Kinyarwanda in collection using a weighing bag as a standard weight, to
the family’s home, except anthropometric measurements check that the scales read the same weight at the begin-
which were taken at the local health centre. All question- ning of every day. If the measurements differed in any
naires were developed from pilot intervention research way, the team checked the equipment and changed
and were forward- translated and back-translated from batteries if needed to make sure the scales were accu-
English to Kinyarwanda following standard WHO proce- rate. Height was measured in centimetres to the nearest
dures.23 Data were entered on Android tablets. Baseline 0.1 cm. Weight was measured in kilograms to the nearest
assessments were conducted between 23 April 2018 and 1 1 g. Middle upper arm circumference was measured in
June 2018, with immediate postintervention assessments centimetres to the nearest 0.1 cm. Standardisation was
conducted between 13 August 2018 and 30 September done using WHO Anthro Survey Analyser software.29
2018, and 12- month follow- up assessments conducted Father engagement was measured using an item from
between 19 August 2019 and 30 September 2019. the Home Observation for Measurement of the Envi-
ronment Inventory,30 namely ‘father spends time every
Outcome measures day caring for the child’ with response options ‘yes/no’
The primary caregiver provided information about the reported by the primary caregiver. Violent and nonvio-
household, including family composition and assets. Per lent discipline practices were assessed using the UNICEF
our conceptual model (figure 1), the post- treatment Multiple Indicator Cluster Survey Child Development
(3 month) outcomes focused on change in caregiver prac- and Child Disciplinary modules, as reported by the
tices, including parent–child interactions, diet, health/ primary caregivers.31 Exposure to violent disciplinary
hygiene, family functioning and family violence.24 The practices included being shouted or screamed at, called
12-month primary outcomes, assessed here, focused on demeaning names, shaken, spanked, slapped or beaten.
changes in children’s motor, language, cognitive and These exposures were summed to create a continuous
social development, as well as violence reduction, and score. Intimate partner violence was assessed by the
father’s engagement in caregiving. Secondary outcomes Rwanda Demographic and Health Survey’s Domestic
were child stunting (ie, child height for age equal to or Violence Module among caregivers who reported being
more than 2 SD below age and sex norms) and related currently married, cohabitating or in a relationship.32 We
anthropometric assessments. report experiences of physical or sexual abuse victimisa-
Child motor, language, cognitive and social devel- tion among female caregivers and perpetration of abuse
opment was assessed using the Ages and Stages Ques- among male caregivers within the last 3 months. Adverse
tionnaire (ASQ-3),25 and the Malawi Developmental events were defined as severe malnutrition or illness of
Assessment Tool (MDAT).26 The ASQ-3 is a series of the child, caregiver suicidality, reports of severe violence
age-specific questionnaires designed to screen for devel- or the death of a caregiver or child.
opmental delay of children in the areas of gross motor
skills, fine motor skills, communication and problem Sample size calculation
solving. We administered the ASQ-3 through interviews Data from two previous pilot studies were used to estimate
with the primary caregiver. No developmental cut-offs for power and determine sample size for a 0.18 minimum
the ASQ-3 have been validated for use in Rwanda, thus, detectable standardised effect size (d) on child develop-
we used continuous scoring using standard guidelines for ment and violence reduction outcomes for the postinter-
obtaining norm-referenced z-scores using means and SD vention and 12-month follow-up period assuming power
from reference samples in South Africa and Zambia.27 of 0.8 and a standard two-tailed alpha level of p<0.05 using
The MDAT is an observational, task-based measure used Optimal Design V.3.01. Estimated intraclass correlation
to asses child development across domains of gross motor, for geographical clusters was 0.03 for parent–child inter-
fine motor, language and socioemotional development. actions based on published pilot data,19 23 and a baseline
The MDAT was developed for use in Malawi with the aim survey conducted in collaboration with UNICEF.33
to be culturally appropriate for use in rural Africa.26 The
MDAT assessment was performed by trained enumera- Statistical analysis
tors. We obtained norm-referenced z-scores using means Effectiveness of SM was determined based on significant
and SD from the Sanitation Hygiene Infant Nutrition differences in the slope of the response variable for SM
Efficacy study from Zimbabwe.28 families compared with UC. Data were analysed by fitting
Children’s anthropometric growth was assessed using a linear mixed effect model for continuous outcomes
measures of standardised height- for-
age (HAZ), stan- where the primary predictors were treatment group
dardised weight-for-age (WAZ), standardised weight-for- (SM vs UC), time as a continuous variable, and their
height (WHZ) and middle upper arm circumference two-way interaction all of which were included as fixed
(MUAC). In coordination with community health workers effects. The model had three levels of nesting: children/
and local leaders, enumerators conducted child growth caregivers are measured within measurement waves,
assessments in centralised locations using locally sourced and measurement waves are nested within randomisa-
MUAC tapes and height boards and scales purchased in tion cluster or region. Since we expect region-level and
the USA. The scales were calibrated every day before data time-level effects, subject-specific slopes and intercepts

BMJ Global Health
Table 1 Estimated difference-in-difference coefficients for continuous, binary and count outcomes
Growth models
(Average change over time in Baseline to
Sugira Muryango compared postintervention Baseline to 12-month
with usual care) difference† follow-up difference†
Difference-in- P
Outcome difference* (95% CI) value Marginal effect (95% CI) Marginal effect (95% CI)
Child development outcomes
ASQ-3 Gross motor z-score 0.294 (0.118 to 0.470) 0.001 0.298 (0.150 to 0.446) 0.597 (0.300 to 0.893)
ASQ-3 Fine Motor z- score 0.004 (−0.119 to 0.127) 0.95 0.061 (−0.043 to 0.165) 0.122 (−0.086 to 0.330)
ASQ-3 Communication z- score 0.139 (0.009 to 0.268) 0.034 0.129 (0.020 to 0.238) 0.259 (0.041 to 0.476)
ASQ-3 Prob. Solving z- score 0.159 (0.035 to 0.282) 0.012 0.189 (0.085 to 0.293) 0.379 (0.171 to 0.587)
ASQ-3 Social Emotional z- score 0.148 (0.023 to 0.273) 0.020 0.122 (0.017 to 0.228) 0.245 (0.0342 to 0.455)
MDAT Gross motor z- score −0.003 (−0.101 to 0.095) 0.951 0.028 (−0.055 to 0.111) 0.055 (−0.111 to 0.221)
MDAT fine motor z- score 0.070 (−0.058 to 0.198) 0.283 0.097 (−0.012 to 0.205) 0.193 (−0.023 to 0.410)
MDAT language z- score −0.022 (−0.126 to 0.080) 0.668 −0.016 (−0.103 to 0.072) −0.032 (−0.206 to 0.143)
MDAT socioemotional z- score 0.073 (−0.016 to 0.164) 0.108 0.060 (−0.016 to 0.136) 0.121 (−0.031 to 0.272)
Father engagement
Father engagement in childcare‡ 1.591 (1.069, 2.368) 0.022 1.565 (1.091, 2.244) 2.449 (1.191, 5.037)
Child growth outcomes
Height-for-age −0.019 (-0.067 to 0.029) 0.44 −0.033 (−0.074 to 0.007) −0.067 (−0.148 to 0.0144)
Weight-for-age −0.031 (−0.071 to 0.009) 0.139 −0.038 (−0.072 to –0.003) −0.075 (−0.143 to –0.007)
Weight-for-height −0.016 (−0.069 to 0.037) 0.551 −0.015 (−0.059 to 0.029) −0.030 (−0.119 to 0.0592)
Middle upper arm circumference −0.057 (−0.113 to –0.001) 0.046 −0.036 (−0.083 to 0.011) −0.072 (−0.167 to 0.0227)
Violence and safety
Harsh discipline § 0.741 (0.657 to 0.835) <0.001 0.774 (0.688 to 0.870) 0.632 (0.510 to 0.783)
Victimisation (female 0.616 (0.458 to 0.828) 0.001 0.616 (0.425 to 0.893) 0.442 (0.238 to 0.820)
caregivers)§, ¶
Perpetration (male caregivers)§, ** 0.604 (0.325 to 0.110) 0.110 0.897 (0.708 to 1.134) 0.842 (0.559 to 1.266)
*Assesses the significance of the ‘difference-in-difference’ or ‘time-by-treatment’ interaction coefficient.
†Difference in marginal means estimates between Sugira Muryango and usual care at each time point.
‡Among households with a father (N=524), Displayed as OR.
§From MICS: UNICEF’s Multiple Indicator Cluster Survey, displayed as incidence rate ratio.
¶Among female caregivers reporting a current intimate partner at baseline (n=523), displayed as incidence rate ratio.
**Among male caregivers reporting a current intimate partner at baseline (n=450), displayed as incidence rate ratio.
ASQ-3, Ages and Stages Questionnaire; MDAT, Malawi Development Assessment Tool.
were modelled as random effects nested within rando- were conducted in StataMP V.16.34 Under intention-to-
misation clusters. Additionally, the type of public works treat assumptions, we replaced lost cases using multiple
programme the family participated in (a stratifier) was imputation via the HOTDECK plug- in module for
modelled as a fixed effect. Binary outcomes were esti- Stata35–38 by randomly selecting five cases matched on
mated using generalised linear mixed models with sex, public works programme type, household structure,
a binomial distribution and logistic link, and count age and treatment group for child outcomes; caregiver
outcomes were analysed using multilevel mixed-effects cases were matched on public works programme type,
negative binomial models. Other details of the model age, educational achievement and treatment group. This
are the same as the linear model described above. We is preferred to carrying the last observation forward, a
defined significance as observing p<0.05 (two tailed) for method now widely understood to be problematic.39
the interaction term between treatment and time. We Quality assurance checks were conducted during data
also report adjusted effect sizes as the marginal effects collection using proprietary audit algorithms to review
from the mixed effects models (marginal effect for survey metadata and flag unusual submissions for further
continuous outcomes and ORs for binary outcomes) for investigation by a trained data manager and senior field
the baseline to immediate postintervention and baseline supervisor. A random sample of 10% of all surveys were
to 12-month follow-up assessments (table 1). Analyses automatically recorded and reviewed by a supervisor to

BMJ Global Health
ensure enumerators followed appropriate data collection 0.657 to 0.835). Among those in dual-caregiver house-
protocols. holds, we found that female caregivers receiving SM
showed a greater decrease in intimate partner violence
Baseline equivalence and modification of intervention effects (IRR=0.616, 95% CI 0.458 to 0.828). However, there
by child sex was no significant reduction of male caregiver reported
Per Sex and Gender Equity in Research guidelines, we perpetration of violence towards their partner.
explored baseline equivalence and potential differen- Adverse events were reported to SM staff by coaches
tial intervention effects by child sex by rerunning all and enumerators within 24 hours of identification.
models including a main effect of gender, a treatment- From baseline to the 12-month follow-up assessment,
by-gender interaction, a time point-by-gender interac- 28 risk of harm cases were reported among interven-
tion and a treatment-by-gender-by-time point interaction families while 27 occurred in the UC group (details
tion. in online supplemental file 1). These households were
retained in the analyses under intention to treat.
RESULTS Effects of child sex

Baseline data were collected on 1084 children and 1498 Analyses examining baseline equivalence and interven-
caregivers and intimate partners. Loss to-follow-up at tion effects by child sex did not reveal any differences in
12 months (2.0% for children; 9.6% for caregivers) and intervention-related improvements in male versus female
item-level missing data (0.1%–2.5%) were low across the children.
outcomes. Table 2 provides sample descriptive statistics.
Caregivers ranged in age from 18 to 84 years at baseline
and were most often the biological mother (n=950), the DISCUSSION
biological father (n=433), or a grandparent (n=96). A The primary hypotheses of greater improvements in
total of 70% (n=1048) of the included caregivers were child development, violence reduction and father
primary caregivers and 30% were secondary caregivers. engagement among families receiving the intervention
Among the primary caregivers, 98.9% were female. compared with UC was supported for violence reduction,
Among the secondary caregivers 98.4% were male. At father engagement and developmental outcomes meas-
baseline, 64% of the caregivers were married or cohab- ured by the parent-reported ASQ-3. For family violence
itating, 61% of the families reported high levels of food (ie, maternal reports of intimate partner violence and
insecurity and 48% of the children were stunted as harsh discipline) and father engagement the hypothesis
defined by a standardised HAZ score below −2 SD of of greater improvement was supported at both points
the reference population at baseline. of intervention follow-up. We did not see significantly
Results from the linear growth models and marginal greater improvements on the MDAT, however, and
effects are shown in table 1. Raw means and SD are there was no evidence of greater improvement in the
reported in the online supplemental table 3; intra-class secondary anthropometric growth outcomes. We have
correlations for main outcomes are displayed in online previously shown immediate post-treatment (3 months)
supplemental table 4. effects of SM on responsive caregiving, caregiver engage-
We observed differences in change over time ment in play activities with the child, decreased use of
between SM and UC on the ASQ-3. Intervention fami- violent discipline and improvements in the mental
lies improved more across the domains of gross motor health of male and female caregivers.24 We also observed
(β=0.294, 95% CI: 0.118 to 0.470), communication immediate pretreatment to post- treatment increases
(β=0.139 95% CI 0.009 to 0.268), problem solving in caregiver behaviours to support healthy child devel-
(β=0.159, 95% CI 0.035 to 0.282) and personal-social opment including increased dietary diversity, hygiene
development (β=0.148, 95% CI 0.023 to 0.273). We did behaviours and healthcare seeking for sick children
not see any significant group differences in change on among participants in SM relative to families in UC.24
fine motor development. Our theory of change (figure 1) would predict that
We did not observe any significant group differences in such changes in parenting behaviours, in turn, would
changes on the MDAT across any of the domain scores: be lead to improvements in long-term ECD and growth
gross motor, fine motor, language and socioemotional outcomes. The combination of significant caregiver
development. behavioural change immediately post- treatment, and
Regarding child growth, at 12-month follow-up, we improvement in developmental milestones and sustain-
did not observe any significant intervention effects on ment of violence reduction at 12-month postintervention
HAZ, WAZ, WHZ or MUAC (all 95% CIs crossed zero). follow-up supports this theoretical foundation of SM. By
Among families with a father in the household, father linking an ECD programme into an existing social protec-
engagement increased more in families receiving SM tion infrastructure, SM was able to identify and reach the
compared with UC (OR=1.592, 95% CI 1.069 to 2.368). most vulnerable population of caregivers in Rwanda,
Families receiving SM also showed a greater decrease those classified as being most vulnerable according to the
in the use of harsh discipline (IRR=0.741, 95% CI government’s poverty ranking system. The linkage with

BMJ Global Health
Table 2 Characteristics of study participants at enrolment. Continuous variables reported as means (SD)
Classic public works (cPW) Expanded public works (ePW)
Sugira Sugira Muryango
Muryango+cPW cPW only +ePW ePW only
Households (n=1049) n=374 n=374 n=167 n=134
High food insecurity 239 (63.9%) 229 (61.2%) 104 (62.3%) 70 (52.2%)
Children (n=1084) n=386 n=384 n=173 n=141
Average age in months 21.0 (8.14) 21.8 (8.6) 20.8 (8.2) 22.3 (8.4)
Health status and well-being
Stunted (standardised height-for-age<2) 184 (47.7%) 178 (46.4%) 85 (49.1%) 75 (53.2%)
Wasted (standardised weight-for-height<2) 13 (3.4%) 9 (2.3%) 8 (4.6%) 2 (1.4%)
Underweight (standardised weight-for-age<2) 63 (16.3%) 71 (18.5%) 30 (17.3%) 27 (19.1%)
Screens positive, disability or developmental delay 110 (28.6%) 111 (29.0%) 57 (32.9%) 38 (27.1%)
Disciplinary practices
Any violent punishment 184 (47.7%) 180 (47.0%) 83 (48.0%) 59 (41.8%)
Caregivers (n=1498) n=555 n=564 n=211 n=168
Primary caregiver 374 (67.4%) 374 (66.3%) 166 (78.7%) 134 (79.8%)
Female 371 (66.9%) 372 (65.9%) 166 (78.7%) 134 (79.8%)
Average age in years 34.5 (9.7) 35.7 (10.3) 36.3 (10.6) 37.5 (12.7)
(Range) (18-79) (19-75) (18-79) (18-84)
Marital Status
Single, separated, divorced, widowed 171 (30.8%) 166 (29.4%) 117 (55.5%) 91 (54.2%)
Relationship with child
Biological mother 341 (61.4%) 338 (59.9%) 152 (72.0%) 119 (70.8%)
Biological father 179 (32.3%) 183 (32.4%) 44 (20.9%) 27 (16.1%)
Adoptive mother 2 (0.4%) 1 (0.2%) 0 (0.0%) 0 (0.0%)
Stepfather 1 (0.2%) 4 (0.7%) 0 (0.0%) 5 (3.0%)
Stepmother 1 (0.2%) 0 (0.0%) 0 (0.0%) 0 (0.0%)
Aunt/uncle 3 (0.5%) 2 (0.4%) 0 (0.0%) 0 (0.0%)
Grandparents 28 (5.0%) 36 (6.4%) 15 (7.1%) 17 (10.1%)
Educational Attainment
No school/don't know 112 (20.2%) 132 (23.4%) 60 (28.4%) 38 (22.6%)
<6 years 275 (49.5%) 252 (44.7%) 97 (46.0%) 88 (52.4%)
≥6 years primary 88 (15.9%) 89 (15.8%) 26 (12.3%) 21 (12.5%)
Secondary/vocational school 80 (14.4%) 91 (16.1%) 28 (13.3%) 21 (12.5%)
Health and safety
Maternal victimisation violence, last 3 months* 78 (39.8%) 73 (35.3%) 15 (29.4%) 15 (36.6%)
Paternal perpetration violence, last 3 months* 38 (21.2%) 41 (22.3%) 10 (23.3%) 4 (12.5%)
Binary variables reported as frequency (%).
*Among mothers (n=495) and fathers (n=438) who are married or cohabitating.
a social protection programme may introduce several counselling and/or supplements, and cash for work that
other benefits. For example, integration of a parenting can be directed towards food, and other vital househod
intervention within social protection systems may result expenses, all of which impact a family’s ability to provide
in higher participation rates and stronger social buy-in a safe and healthy environment for a child.
around the parenting programme. Moreover, combined The use of active coaching allowed SM to involve
ECD and social protection programmes may benefit all family members, including fathers and other male
from several synergistic effects because social protection caregivers present in the household in nurturing care
programmes tend to provide important benefits such as as reflected in increased father engagement. Indeed,
access to health insurance and healthcare, nutritional we note that father engagement is one of the strongest

BMJ Global Health
effects of home-visiting and parenting interventions.42–44
effects of SM, especially postintervention where fathers
in SM families were 2.5 times more likely to have engaged The observed effects on the parent-reported ASQ-3, in
with their child in the past 24 hours, compared with the absence of similar effects on observational measures,
control fathers. Flexible scheduling and messaging about may suggest that parenting interventions change or sensi-
the importance of fathers in ensuring a nurturing and tise parents’ perception of their child’s behaviours and
safe environment for young children to grow and thrive abilities, and thus affect the parents’ ability to provide
resulted in both high module attendance by fathers and parent reports over time. Moreover, although interven-
a significant relative increase in fathers’ involvement in tion and control families were exposed to the same test
childcare. This adds to the important and growing litera- batteries, parents receiving the intervention may feel
ture on ways to engage fathers in ECD interventions and more encouraged to report on their children’s develop-
childcare responsibilities. Previous work from Uganda mental milestones because they know they received an
found that use of father-only sessions, exploitation of intervention. Interestingly, a similar discrepancy between
men’s pre-existing motivation to improve their children’s parent report vs an observer rating was found in an eval-
behaviour, and interactive delivery helped to engage uation of Peru’s home- visiting Cuna Mas programme
fathers40 wherein a programme delivery fidelity measure was
SM also maintained effects on reduced violence, found to correlate significantly with the ASQ-3 scores,
including reduced violent discipline and reduced but not with the Bayley scales,45 a gold-standard observa-
maternal reports of intimate partner violence observed tional measure administered by selected highly trained
immediately postintervention and again 1 year after psychology graduates.41
the intervention ended.24 We did not retain the reduc- SM also did not lead to improvements in child growth.
tion of paternal- reported perpetration of intimate While the 95% CI crosses zero for all estimated effects
partner violence observed immediately postintervention. on growth outcomes, we saw a small negative trend
Violence reduction is a key aim of SM and is achieved towards lower MUAC in the SM children at the 12-month
through a combination of psychoeducation and coaching follow-up, but this was not significant. Several factors may
on topics ranging from mindfulness meditation tech- explain the absence of positive intervention effects on
niques, conflict resolution, alternative strategies to harsh children’s anthropometric outcomes. At baseline, 48% of
discipline, the harmful effect of harsh discipline, the the children were stunted, which is known to decrease the
benefits of responsive care and father engagement in opportunity to change a child’s growth trajectory, because
childcare. We believe that the use of active coaching and the effects of early malnutrition are difficult to reverse.46
instruction was central in establishing persistent changes Although the curriculum addresses child feeding, nutri-
in caregiver behaviours. The inclusion of violence reduction and hygiene with the aim of improving children’s
tion was motivated by previous identification of violence health and growth, a behavioural intervention with active
as a key challenge to ECD among young children in coaching on nurturing care may not be enough to cause
Rwanda.33 During the intervention, coaches became measurable improvements in anthropometric growth,
aware of several incidents of violence in the families particularly among extremely vulnerable households and
they served and worked closely with their supervisor and already malnourished children. For example, parental
village officials to assess and resolve such situations and to training in diet among parents in extreme poverty
link families with existing resources. without additional direct nutrition support may not be
Key strengths of the study compared with prior adequate to overcome past and present malnutrition.
related studies include the rigorous, well-powered CRT In future research, the impact of SM on child growth
design and randomisation, including the blinding of the may be intensified if combined with other programmes,
enumerators to treatment condition and geographical such as child-focused conditional cash transfers, agricul-
separation of SM locations from UC. The use of a strat- tural and food supplementation, health, water, sanitisa-
ified randomised clustered design with larger numbers tion and hygiene programmes. Future research should
of both clusters and families is a marked improvement also test the proposed mechanistic pathways of change
over studies recently reviewed systematically12 and found in child outcomes via changes in caregiver behaviours
in our own review of more recent studies.13–17 put forward in our theory of change. Moreover, future
With regard to cognitive developmental outcomes, the analyses looking at potential moderators of interven-
results from the ASQ-3 and MDAT were not consistent. tion outcomes (beyond child sex) should be explored
The MDAT was new to the Rwandan setting; it is possible to understand whether certain subpopulations benefit
that revision of items and other refinements may be more or less from the intervention.
necessary to ensure its sensitivity to change over time in Our results are consistent with other studies that have
that environment. Although both the ASQ-3 (caregiver examined child development and parental behaviour
report) and MDAT (direct assessment) were originally outcomes in the context of combined ECD and govern-
developed as screening tools to detect developmental ment social protection programmes. In Colombia, an
delay, not growth trends, the ASQ-3 has been widely used ECD programme that delivered psychosocial stimula-
in previous intervention studies in settings as diverse as tion (and micronutrient supplementation) to beneficia-
Rwanda,23 China17 and Peru41 and found to be sensitive to ries of the Familias en Acción conditional cash transfer

BMJ Global Health
programme yielded promising short- term impacts on explored further. Supervision in these systems would need
parenting behaviours and child cognitive develop- to be addressed to support the workforce to deliver and
ment.5 Specifically, the parenting intervention, which sustain quality in delivery of the programme which likely
consisted of weekly home visits by beneficiary- elected requires a different set of competencies. Testing imple-
female community leaders over an 18-month period, was mentation strategies for scaling out the intervention and
associated with statistically significant improvements in innovations in structures for sustainment including atten-
cognition (effect size 0.26 SD) and receptive language tion to quality improvement models to support sustained
(effect size 0.22 SD). Similarly, results from a randomised high-quality supervision will be critical future research.
controlled trial of a home-based parenting intervention A number of study limitations should be noted. First,
delivered by local NGOs to households participating in given the large sample size and focus on vulnerable fami-
Niger’s unconditional cash transfer programme (Niger lies, some families enrolled in this study may also have
Safety Nets) also lend support for integrated approaches been eligible for, and participated in, other interventions
to supporting the holistic development of children beyond SM and the public works programme. During the
living in poverty. Among families who participated in selection of target districts, we spoke with a broad range
the ECD intervention, a combination of monthly home of implementors and funders of related programmes to
visits and small group meetings, they found that chil- map ongoing interventions and we selected the targeted
dren had moderate improvements in socioemotional districts based on minimal overlap with other interven-
development, and parents interacted with their chil- tions. Still, we cannot rule out the possibility that some
dren around activities such as storytelling, counting or families may have participated in other interventions that
drawing more frequently and were less likely to use harsh could influence results, although this should not differ
disciplinary practices.46 Overall, these along with current between SM and control families. We also note that this
findings suggest that integrated ECD and social protec- study tested the linkages of an ECD intervention with a
tion programmes hold promise for improvements in social protection programme. Synergistic effects of ECD
child cognitive development and parenting behaviours. intervention and social protection may mean that find-
Evidence for anthropometric gains in combined ECD and ings from the current study do not necessarily gener-
social protection programmes, however, remains elusive. alise to poor families who do not have access to social
The potential for combined ECD and social protection protection programmes. A second limitation is that key
such as cash transfer programmes to have synergistic outcomes related to violence in the home and child devel-
effects remains inconclusive and understudied since most opment assessed on the ASQ is based on parental report.
studies use social protection as a platform for identifying This always introduces some potential for reporting bias.
their beneficiaries and don’t include an ECD-only arm. In particular, participation in the study and intervention
SM, a brief, comprehensive ECD intervention delivered may have sensitised parents to observing changes in these
by lay coaches, successfully targeted and reached fami- outcomes.
lies living in extreme poverty and provided an oppor-
tunity for parenting support to complement income
support through two public works programmes. SM is CONCLUSION
designed to be deployment-focused, relatively brief and This CRT integrated SM into Rwanda’s social protection
delivered by lay workers with strong fidelity monitoring programme as a platform for targeting poor and vulner-
and quality improvement- focused supervision. Scaling able families with the potential for transitioning the
out the programme to reach large numbers of families SM programme to scale. In particular, the programme
in extreme poverty might best be achieved through the had a large and persistent effect on engaging fathers in
utilisation of a pre-existing government workforce. In childcare. Moreover, we also saw smaller but significant
Rwanda, such a work force includes the inshuti z’umury- effects on violence reduction including intimate partner
ango (friends of the family) child protection worker and violence and harsh discipline as well as child develop-
other community health workforces. Linkage of ECD ment. With continued attention to quality and fidelity,
programmes to the existing social protection platform, evidence-based interventions such as SM can be scaled
as explored in this study, has several advantages; it targets out across settings, providing the government of Rwanda
the most disadvantaged, supports equity, and coordinates and others with a critical tool for helping to break inter-
with other interventions (eg, food supplementation, generational cycles of poverty and violence.
conditional cash transfers). Combined provision of social
Author affiliations
protection and ECD intervention may also have several 1
School of Social Work, Boston College, Chestnut Hill, Massachusetts, USA
synergistic effects because even greater gains, espe- 2
Lynch School of Education and Human Development, Boston College, Chestnut
cially on anthropometric growth may be possible if the Hill, Massachusetts, USA
3
programme were integrated explicitly with a programme Women's Study Research Center, Brandeis University, Waltham, Massachusetts,
providing nutritional supplementation. Such synergistic USA
4
Academic Research Services, Boston College, Chestnut Hill, Massachusetts, USA
effects will be important to unpack in future research. 5
Department of Global Health and Population, Harvard T.H. Chan School of Public
From an implementation standpoint, coordination Health, Boston, Massachusetts, USA
across sectors (health and social protection) need to be 6
The World Bank, Washington, District of Columbia, USA

BMJ Global Health
7
FXB Rwanda, Kigali, Rwanda Data availability statement Data are available on request. Data can be made
8
Center for Mental Health, University of Rwanda, Kigali, Rwanda available on request and will be made publicly available via the World Bank’s data
catalog at a later date.
Acknowledgements The work was made possible by the collaboration of the Supplemental material This content has been supplied by the author(s). It has
University of Rwanda Center for Mental Health, the Ministry of Gender and Family not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been
Promotion, the National Child Development Agency (formerly the National Early peer-reviewed. Any opinions or recommendations discussed are solely those
Childhood Development Program and the National Commission for Children), the of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and
Local Administrative Entities Development Agency, and the advisory committee responsibility arising from any reliance placed on the content. Where the content
comprised of representatives from the Ministry of Local Government, National includes any translated material, BMJ does not warrant the accuracy and reliability
Commission for Children, Rwanda Biomedical Centre, Ministry for Gender and of the translations (including but not limited to local regulations, clinical guidelines,
Family Promotion, National Early Childhood Development Program and the terminology, drug names and drug dosages), and is not responsible for any error
University of Rwanda School of Nursing.Early Childhood Development Program and/or omissions arising from translation and adaptation or otherwise.
Contributors TSB conceptualised the study, obtained funding, led intervention Open access This is an open access article distributed in accordance with the
development, provided supervision and review of the manuscript. SKGJ had full Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which
access to the data and lead the analyses, interpreted results and lead the writing permits others to distribute, remix, adapt, build upon this work non-commercially,
of the manuscript. MPC conducted data analyses, created tables, and contributed and license their derivative works on different terms, provided the original work is
to data interpretation and writing and had full access to the data. SMM oversaw properly cited, appropriate credit is given, any changes made indicated, and the
intervention implementation, data collection and made significant intellectual use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
contributions to the manuscript content. RTB contributed to study concept and
design, statistical consultation and drafting of the manuscript. SG provided expert
advice and help with data imputation and analysis. JF oversaw the integrated
fidelity monitoring and supervision process and provided critical review of
manuscript content.AY contributed subject matter expertise and critical review of REFERENCES
manuscript content. LBR and BW contributed to policy dialog management with 1 Mikton C, MacMillan H, Dua T, et al. Integration of prevention of
government counterparts, study design, securing funding, data interpretation and violence against children and early child development. Lancet Glob
writing. EH oversaw intervention implementation and policy dialog management Health 2014;2:e442–3.
with local government. VS contributed to intervention adaptation, facilitated contact 2 Jeong J, Obradović J, Rasheed M, et al. Maternal and paternal
stimulation: Mediators of parenting intervention effects on
and dialog with local partners and made significant intellectual contributions to the
preschoolers’ development. J Appl Dev Psychol 2019;60:105–18.
intervention and manuscript content. 3 Knaul FM, Bustreo F, Horton R. Countering the pandemic of gender-
Funding World Bank Early Learning Partnership, Strategic Impact Evaluation Fund, based violence and maltreatment of young people: the Lancet
World Bank Japan Trust Fund, USAID Rwanda, Network of European Foundations Commission. Lancet 2020;395:98–9.
and ELMA Foundation. 4 Ward CL, Wessels IM, Lachman JM, et al. Parenting for lifelong
health for young children: a randomized controlled trial of a
Disclaimer The World Bank Group (WBG) authors contributed to study design and parenting program in South Africa to prevent harsh parenting and
writing of the manuscript. The findings, interpretations, and conclusions expressed in child conduct problems. J Child Psychol Psychiatry 2020;61:503–12.
this paper are entirely those of the authors. They do not necessarily reflect the views 5 Attanasio OP, Fernández C, Fitzsimons EOA, et al. Using the
of the WBG, its Board of Executive Directors, or the governments they represent. The infrastructure of a conditional cash transfer program to deliver a
scalable integrated early child development program in Colombia:
WBG does not guarantee the accuracy of the information included in this work. The cluster randomized controlled trial. BMJ 2014;349:g5785.
remaining funders, USAID Rwanda, ELMA Foundation, and the Network of European 6 Lachman J, Wamoyi J, Spreckelsen T, et al. Combining parenting
Foundations played no role in study design, data collection, data analysis, data and economic strengthening programmes to reduce violence against
interpretation, or writing of the report. children: a cluster randomised controlled trial with predominantly
male caregivers in rural Tanzania. BMJ Glob Health 2020;5:e002349.
Competing interests None declared. 7 Walker SP, Chang SM, Vera-Hernández M, et al. Early childhood
Patient and public involvement statement Design of the intervention was stimulation benefits adult competence and reduces violent behavior.
carried out with input and routine consultation from stakeholders at the Ministry Pediatrics 2011;127:849–57.
of Gender and Family Promotion and the National Early Childhood Development 8 Yousafzai AK, Obradović J, Rasheed MA, et al. Effects of responsive
stimulation and nutrition interventions on children’s development and
Program in Rwanda. We engaged with village leaders and the Ministry of Local growth at age 4 years in a disadvantaged population in Pakistan: a
Government (MINALOC), which runs the Poverty Reduction Strategy, the Vision longitudinal follow-up of a cluster-randomised factorial effectiveness
Umurenge Program, prior to and during the study implementation to discuss trial. Lancet Glob Health 2016;4:e548–58.
recruitment of participants and to inform them of the process and outcomes. 9 Hamadani JD, Huda SN, Khatun F, et al. Psychosocial stimulation
Representatives from the data collection firm, Laterite, and our collaborators at improves the development of undernourished children in rural
the University of Rwanda, were closely involved in developing and refining the Bangladesh. J Nutr 2006;136:2645–52.
research-questions, study design, selection of measures and assessment tools, 10 Powell C, Baker-Henningham H, Walker S, et al. Feasibility of
integrating early stimulation into primary care for undernourished
and ethical procedures. During implementation, we worked closely with our field-
Jamaican children: cluster randomised controlled trial. BMJ
based implementation partner, FXB-Rwanda, who provided training and supervision 2004;329:89.
to the community-based coaches to discuss ongoing implementation and emerging 11 Singla DR, Kumbakumba E, Aboud FE. Effects of a parenting
issues. At the end of the study, we communicated results at a local dissemination intervention to address maternal psychological wellbeing and child
event which included all key stakeholders, including the Government of Rwanda development and growth in rural Uganda: a community-based,
and other organisations that implement child development programs in Rwanda. cluster-randomised trial. Lancet Glob Health 2015;3:e458–69.
"The event included a facilitated discussion with program stakeholders, including 12 Efevbera Y, McCoy DC, Wuermli AJ, et al. Integrating early child
local government officials and study interventionists, as well as a presentation, led development and violence prevention programs: a systematic
review. New Dir Child Adolesc Dev 2018;2018:27–54.
by the Principal Investigator, on program outcomes, lessons learned and next steps.
13 Sumargi A, Sofronoff K, Morawska A. A Randomized-Controlled
Patient consent for publication Not required. trial of the triple P-Positive parenting program seminar series with
Indonesian parents. Child Psychiatry Hum Dev 2015;46:749–61.
Ethics approval Procedures and research protocols related trial (registration 14 Puffer ES, Green EP, Chase RM, et al. Parents make the difference:
number NCT02510313) received ethics approval from the Harvard T. H. Chan School a randomized-controlled trial of a parenting intervention in Liberia.
of Public Health (IRB16-1570) and Boston College Institutional Review Boards Glob Ment Health 2015;2:e15.
(19.017), as well as the Rwanda National Ethics Committee (no approval number 15 Tomlinson M, Rotheram-Borus MJ, le Roux IM, et al. Thirty-six-
provided), National Commission for Science and Technology (no approval number month outcomes of a generalist Paraprofessional perinatal home
provided), and the National Institute of Statistics of Rwanda (no approval number visiting intervention in South Africa on maternal health and child
health and development. Prev Sci 2016;17:937–48.
provided).
16 Baker-Henningham H, Walker S. Effect of transporting an evidence-
Provenance and peer review Not commissioned; externally peer reviewed. based, violence prevention intervention to Jamaican preschools on

BMJ Global Health
teacher and class-wide child behaviour: a cluster randomised trial. 30 Caldwell BM, Bradley RH. Home observation for measurement of the
Glob Ment Health 2018;5:e7. environment: administration manual. Tempe, AZ: Family & Human
17 Zhou S, Zhao C, Huang X, et al. The effect of a community-based, Dynamics Research Institute, Arizona State University, 2003.
integrated and nurturing care intervention on early childhood 31 UNICEF. MICS5 questionnaire for children under five, 2013.
development in rural China. Public Health 2019;167:125–35. Available: http://mics.unicef.org/tools?round=mics5
18 Republic of Rwanda Local Administrative Entities Development 32 National Institute of Statistics of Rwanda, ORC Macro. Rwanda
Agency (LODA). Environmental and social management framework demographic and health survey 2005. Kigali, Rwanda, 2006.
for the social protection investment project financing operation, Available: https://dhsprogram.com/pubs/pdf/FR183/FR183.pdf
2018. Available: https://www.undp.org/content/undp/en/home/ 33 UNICEF, Imbuto Foundation. Early childhood development and
search.html?q=2020+Umurenge+Program+%28VUP%29%0A family services: baseline evaluation in 20 sites in Rwanda, 2015.
19 Betancourt TS, Franchett E, Kirk CM, et al. Integrating social Available: https://www.unicef.org/evaldatabase/files/ECD_and_F_
protection and early childhood development: open trial of a family Baseline_Evaluation_Rwanda.pdf 34 Statacorp. StataMP 16. 2019
home-visiting intervention, Sugira Muryango. Early Child Dev Care 34 Statacorp, StataMP 16. Release 16. College Station, TX: StataCorp
2020;190:219–35. LLC, 2019.
20 World Health Organization, UNICEF. Care for child development: 35 Reilly M. Data analysis using hot Deck multiple imputation. The
improving the care for young children, 2012. Available: https:// Statistician 1993;42:307–13.
www.who.int/maternal_child_adolescent/documents/care_child_ 36 Little RJA. Missing-Data adjustments in large surveys. J Bus Econ
development/en/ Stat 1988;6:287–96.
21 UNICEF, WHO. The world bank. nurturing care for early child 37 Lachin JM. Fallacies of last observation carried forward analyses.
development: a framework for helping children survive and Clin Trials 2016;13:161–8.
thrive to transform health and human potential. Geneva, 2018. 38 Mander A, Clayton D. HOTDECK: Stata module to impute missing
Available: https://apps.who.int/iris/bitstream/handle/10665/272603/ values using the hotdeck, 1999. Available: https://ideas.repec.org/c/
9789241514064-eng.pdf boc/bocode/s366901.html
22 Betancourt TS, Ng LC, Kirk CM, et al. Family-Based promotion 39 Siu GE, Wight D, Seeley J, et al. Men’s Involvement in a Parenting
of mental health in children affected by HIV: a pilot randomized Programme to Reduce Child Maltreatment and Gender-Based
controlled trial. J Child Psychol Psychiatry 2017;58:922–30. Violence: Formative Evaluation in Uganda. Eur J Dev Res
23 Barnhart DA, Farrar J, Murray SM, et al. Lay-worker delivered home 2017;29:1017–37.
visiting promotes early childhood development and reduces violence 40 Araujo MC, Dormal M, Rubio-Codina M. Quality of parenting
in Rwanda: a randomized pilot. J Child Fam Stud 2020;29:1804–17. programs and child development outcomes: the case of Peru’s Cuna
24 Betancourt TS, Jensen SKG, Barnhart DA, et al. Promoting parent- Mas. Washington D.C. Peru: Ministerio de Educación del Perú, 2018.
child relationships and preventing violence via home-visiting: a 41 Atukunda P, Muhoozi GKM, van den Broek TJ, et al. Child
pre-post cluster randomised trial among Rwandan families linked to development, growth and microbiota: follow-up of a randomized
social protection programmes. BMC Public Health 2020;20:621. education trial in Uganda. J Glob Health 2019;9:010431.
25 Squires J, Bricker D. Ages & stages questionnaires, third edition 42 Worku BN, Abessa TG, Wondafrash M, et al. Effects of home-
(ASQ-3): a parent-completed child-monitoring system, 2009. based play-assisted stimulation on developmental performances
Available: https://brookespublishing.com/product/asq-3/ of children living in extreme poverty: a randomized single-blind
26 Gladstone M, Lancaster GA, Umar E, et al. The Malawi controlled trial. BMC Pediatr 2018;18:29.
developmental assessment tool (MDAT): the creation, validation, 43 Mohammed H, El-Gibaly O, Monazea E, et al. The effect of a
and reliability of a tool to assess child development in rural African maternal training programme on early childhood development in
settings. PLoS Med 2010;7:e1000273. Egypt. East Mediterr Health J 2019;25:575–82.
27 Hsiao C, Richter L, Makusha T, et al. Use of the ages and stages 44 Bayley N. Bayley scales of infant and toddler development. (3rd ED.
questionnaire adapted for South Africa and Zambia. Child Care San Antonio, TX: The Psychological Corporation, 2006.
Health Dev 2017;43:59–66. 45 Rubio-Codina M, Attanasio O, Meghir C, et al. The socioeconomic
28 Gladstone MJ, Chandna J, Kandawasvika G, et al. Independent gradient of child development: cross-sectional evidence
and combined effects of improved water, sanitation, and from children 6–42 months in Bogota. J Human Resources
hygiene (wash) and improved complementary feeding on early 2015;50:464–83.
neurodevelopment among children born to HIV-negative mothers in 46 Premand P, Barry O. Behavioral change promotion, cash transfers
rural Zimbabwe: substudy of a cluster-randomized trial. PLoS Med and early childhood development: experimental evidence from a
2019;16:e1002766. government program in a low-income setting. Washington D.C.,
29 World Health Organization. WHO Anthro survey analyser, 2020. 2020. Available: https://openknowledge.worldbank.org/handle/
Available: https://www.who.int/childgrowth/software/en/ 10986/34385

BMJGH 2020 003508

Uploaded by

Copyright:

Available Formats

BMJGH 2020 003508

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BMJGH 2020 003508

Uploaded by

Copyright:

Available Formats

Original research

Effect of a home-­visiting parenting

To cite: Jensen SKG, ABSTRACT

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508  1

mothers,2 limiting the ability to work with the full home

2 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 3

4 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 5

6 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 7

RESULTS Effects of child sex

8 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 9

effects of home-­visiting and parenting interventions.42–44

10 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 11

12 Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 13

You might also like

Effect of a home-visiting parenting

Jensen SKG, et al. BMJ Global Health 2021;6:e003508. doi:10.1136/bmjgh-2020-003508 1

effects of home-visiting and parenting interventions.42–44