Scaling up an early childhood intervention by integrating into health services in rural Bangladesh

How to cite

Baker-Henningham, Helen. 2020. "Scaling up an early childhood intervention by integrating into health services in rural Bangladesh." AEA RCT Registry. AEARCTR-0006536. https://doi.org/10.1257/rct.6536-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Helen Baker-Henningham*

* Primary investigator

Registry ID
AEARCTR-0006536
DOI
https://doi.org/10.1257/rct.6536-1.1
First registered
Last updated
Status
Completed
Trial dates
2018-01-01 – 2019-07-31
Countries
Bangladesh ()
Registry record
http://www.socialscienceregistry.org/trials/6536

Abstract

We have previously demonstrated that parent-training sessions conducted by government health workers, as part of their existing duties, in primary health care clinics in rural Bangladesh leads to large benefits to the level of stimulation in the home and to the development of disadvantaged young children. However, in previous studies, the research team trained and supervised the government workers. For the intervention to be implemented at scale, we need a feasible and effective model for the training and supervision of clinic health staff using the government supervisors and trainers. In this feasibility study, we evaluated the effectiveness of utilising government supervisors to train and supervise clinic staff to implement the intervention. We evaluated this approach using two measures: observed quality of parent-training sessions and the parenting practices of participating mothers. A key issue when integrating early childhood development programmes into existing services is that staff may be overburdened. Hence, we also measured burn-out among clinic staff.

The study was conducted in sixteen unions in rural Bangladesh. Eight unions (24 clinics) were randomly assigned to an intervention group and eight unions (24 clinics) to a control group. In intervention clinics, twenty-four mothers of children aged 6-36 months, living within a thirty minute walk from the clinic were invited to participate in fortnightly play sessions, delivered by government health staff in the clinic, in groups of four mother/child dyads. The effect of the play sessions was evaluated on the level of stimulation in the home by maternal report. The evaluation sample consisted of 384 mother/child dyads (192 intervention and 192 control). Burn out of health staff was also assessed in both intervention and control clinics. We also measured the quality of the parenting sessions in intervention clinics through observation.

Intervention

Government staff in community clinics conduct fortnightly parenting sessions with groups of four mother/child dyads for six months. The parenting sessions involve supporting mothers to play and talk with their child using low-cost play materials.

Experimental design

The study is two-armed cluster randomized trial. Sixteen unions in rural Bangladesh were randomly assigned to an intervention (n=8) or control (n=8) group. Each union has three community clinics, giving a total of 48 clinics (24 intervention, 24 control).
The participants are mothers of children aged 6-30 months living within a thirty-minute walking distance from a participating community clinic in rural Bangladesh and government health staff attached to those clinics.
In intervention clinics, mother/child dyads are invited to attend fortnightly play sessions, delivered by government staff in the community clinic for six months. Mother/child dyads assigned to the control group receive no play sessions.
Twenty-four mother/child dyads from 24 intervention clinics were recruited into the intervention (to give a total of 576 mother/child dyads). A random sample of twelve mothers from two randomly selected clinics in each of the eight regions were included in the evaluation sample (n=192).
In the unions allocated to the control group, two clinics in each region were randomly selected and twelve mothers of children aged 6-30 months, living within a 30-minute walk from the clinic were recruited (n=192).
All health staff in the 48 clinics participating in the study were recruited (65 intervention, 65 control). In intervention clinics only, these staff implemented parenting sessions. All health staff answered questionnaires on professional burnout.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Unions
Clusters
16 unions were randomized to intervention or control. Each union has 3 health clinics, giving a total of 48 clinics (24, intervention, 24 control).
Observations
576 mother/child dyads, 65 clinic health staff, 24 government health clinics, 19 government supervisors were involved in intervention implementation  48 community clinics, 125 clinic health staff, 384 mother/child dyads were included in the evaluation
Arms
A subsample of mother/child dyads were selected for inclusion in the evaluation. Two clinics from each union and 12 mother/child dyads from each clinic were randomly selected to participate in the evaluation giving a total of 192 mother/child dyads per group.
Primary outcomes
Stimulation in the home
Secondary outcomes
Clinic staff burn-out

Papers from this trial

  1. Mehrin SF, Salveen NE, Kawsir M, Grantham-McGregor S, Hamadani JD, Baker-Henningham H. (2022) Scaling-up an early childhood parenting intervention by integrating into government health care services in rural Bangladesh: A cluster randomized controlled trial. Child: Care, Health and Development, 1-10. https://doi.org/10.1111/cch.13089 https://doi.org/10.1111/cch.13089

Cited by

No citation data available for this registration yet.

Keywords