Transparency for Development (T4D): Evaluating the impact of community transparency and accountability initiatives on Maternal and Neonatal Health outcomes

How to cite

Levy, Dan; Creighton, Jessica; Arkedis, Jean; Tolmie, Courtney; Fung, Archon; Kosack, Stephen. 2015. "Transparency for Development (T4D): Evaluating the impact of community transparency and accountability initiatives on Maternal and Neonatal Health outcomes." AEA RCT Registry. AEARCTR-0000655. https://doi.org/10.1257/rct.655-5.0

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

Dan Levy* , Jessica Creighton , Jean Arkedis , Courtney Tolmie , Archon Fung , Stephen Kosack

* Primary investigator

Registry ID
AEARCTR-0000655
DOI
https://doi.org/10.1257/rct.655-5.0
First registered
Last updated
Status
On going
Trial dates
2015-02-09 – 2019-09-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/655

Abstract

Recent decades have seen vast expansions of public health services across much of the developing world. But the quality of these services is often lacking, limiting their ultimate impact on the world’s poor. While many factors contribute to service delivery problems, one important factor is governance. The incentives that shape frontline service provider performance, training, oversight, and management all play an important role in the quality of healthcare provided by any health system. Transparency and accountability (T/A) techniques—such as social audits, public expenditure tracking surveys, citizen report cards, absenteeism studies, and community scorecards—are increasingly seen as one potential solution to some of these problems in that they allow communities to identify breakdowns and hold responsible agents or decision makers to account.

Yet the evidence about the effectiveness of these interventions is frustratingly mixed. Some studies of T/A interventions show tremendously positive effects on community empowerment and health and education outcomes, while others show no impact. Further, evidence on successful interventions provides few insights into why these interventions worked, what aspects of the context played a role, and whether they truly empowered poor communities or were largely irrelevant to the deeper problems of power inequity, institutional failure, or social conflict that often foster slow and uneven development. One consequence of this evidence gap is that reformers, policymakers, donors, and other practitioners lack clear guidance on whether any T/A interventions can ultimately empower poor populations and improve service delivery to them and on what sorts of T/A interventions should be pursued in different contexts.

The Transparency for Development (T4D) project is designed to investigate the questions of whether well-designed T/A interventions improve health outcomes and under what conditions. The overriding goal is to generate actionable evidence for practitioners, researchers, and other stakeholders working to improve health, accountability, and citizen participation.

The Transparency for Development project will:

1. Design, pilot, and rigorously evaluate a series of T/A interventions across several countries and contexts carefully selected to allow maximum potential for producing generalizable results about whether, where, and why T/A interventions improve development outcomes
2. Develop and widely disseminate the project’s findings along with concrete recommendations for donors, policymakers, practitioners, and other stakeholders on how to apply the findings to future work in transparency and accountability.

Intervention

The intervention comprises four activities: (1) information gathering; (2) identification and development of community activists; (3) facilitation of community meetings to share information and develop an action plan; and (4) follow-up

The intervention is designed to give community members access to information related to MNH in order to motivate them to plan and conduct social action to improve maternal and neonatal health in their community.

Experimental design

Two RCTs (one in Indonesia and one in Tanzania). Each includes 100 treatment and 100 control villages. See pre-analysis plan for a full description.

Design details

Randomization method
Randomization done by a computer
Randomization unit
Indonesia: one eligible village is randomly selected per health facility catchment area. Within each village, subvillages are randomly selected. Eligible respondents are then randomly selected from these selected subvillages. Tanzania: one or two eligible villages are randomly selected per health facility catchment area. Within each village, subvillages are randomly selected. Eligible respondents are randomly selected from these subvillages.
Clusters
200 villages per country for a total of 400 villages
Observations
Baseline = 5,400 women who have given birth in the past year (2,400 in Indonesia and 3,000 in Tanzania). Endline = 12,000 women who have given birth in the past year (6,000 in Indonesia and 6,000 in Tanzania).
Arms
100 control villages and 100 treatment villages (per country), for a total of 200 control and 200 treatment villages
Primary outcomes
- Health service outcomes (increased utilization and improved facility interaction) - Health outcomes - Citizen empowerment and participation

Cited by

No citation data available for this registration yet.

Keywords