Download StudyGeneral

Study Overview

Title:
Empirical evaluation of the Ethiopian Health Extension Program (HEP)
Study is 3ie funded:
No
Study ID:
RIDIE-STUDY-ID-5a940d821a20a
Initial Registration Date:
02/26/2018
Last Update Date:
02/09/2018
Study Status:
Ongoing
Location(s):
Ethiopia
Abstract:

This study is a retrospective evaluation of the impact of Ethiopia's Health Extension Program (HEP) in rural areas. The HEP was launched in 2003 as innovative approach to deliver health education, change health behaviors, provide family health services and improve disease prevention and control. Initially launched in rural areas, the HEP trained health extension workers (HEW) and deployed two HEWs to newly constructed health posts in kebeles (villages) of about 3,000 - 5,000 individuals. We combine existing household data (the 2005 and 2011 Ethiopia Demographic and Health Surveys) with newly collected data on the presence of HEW and health posts, as well as HEW training and operations in rural areas of four regions. We estimate difference-in-difference models to examine the change in outcomes as the HEP scaled up between 2003 and 2011.

Registration Citation:

Bauhoff, S. and Tolera, T., 2018. Empirical evaluation of the Ethiopian Health Extension Program (HEP). Registry for International Development for Impact Evaluations (RIDIE). Available at: 10.23846/ridie136

Categories:
Health, Nutrition, and Population
Additional Keywords:
Ethiopia, community health workers, maternal and child health
Secondary ID Number(s):

Principal Investigator(s)

Name of First PI:
Sebastian Bauhoff
Affiliation:
Center For Global Development
Name of Second PI:
Taye Tolera
Affiliation:
Armauer Hansen Research Institute & Ethiopian Ministry of Health

Research Partner

Name of Partner Institution:
Armauer Hansen Research Institute
Type of Organization:
Government agency (eg., statistics office, Ministry of Health)
Location:
Ethiopia
Intervention

Intervention Overview

Intervention:

Ethiopia’s Health Extension Program (HEP) launched in 2003 as an innovative approach to achieving the goal of improved access to under-served populations, at low costs. The HEP focuses on prevention, health promotion, behavioural change communication, and basic curative care. It includes expanding the physical infrastructure through the construction of health posts, and expanding frontline human resources through the training and deployment of health extension workers (HEW). Specifically, the HEP trains HEWs in hygiene and sanitation; family health services; disease prevention and control; and health education and communication. Workers are then deployed in pairs to rural kebeles (villages) to serve and interact with communities of 3,000-5,000 people. Since its launch in the four agrarian regions in 2003, HEP has been scaled up to pastoral communities in 2006 and urban areas in 2009. It had deployed about 34,000 HEWs in 2009/10.

Theory of Change:

The HEP is expected to increase the number of trained health extension workers and health posts in rural communities. Their activities are expected to improve household's knowledge and access to services targeted by the program. This is expected to improve health knowledge and health behavior, increase utilization of key services, and ultimately improve health outcomes.

Multiple Treatment Arms Evaluated?
No

Implementing Agency

Name of Organization:
Ministry of Health / Government of Ethiopia
Type of Organization:
Public Sector, e.g. Government Agency or Ministry

Program Funder

Name of Organization:
Ministry of Health / Government of Ethiopia
Type of Organization:
Public Sector, e.g. Government Agency or Ministry

Intervention Timing

Intervention or Program Started at time of Registration?
Yes
Start Date:
01/01/2003
End Date:
Evaluation Method

Evaluation Method Overview

Primary (or First) Evaluation Method:
Difference in difference/fixed effects
Other (not Listed) Method:
Additional Evaluation Method (If Any):
Other (not Listed) Method:

Method Details

Details of Evaluation Approach:

We use a difference-in-difference analysis that combines existing outcomes data (using retrospective data from the Demographic and Health Surveys, DHS) with newly collected administrative data on the presence of health workers and health posts in those rural districts/villages in the four agrarian regions that were sampled in the 2011 Ethiopia DHS (and sometimes also in the 2005 DHS). We match the outcomes and treatment data based on their geography. We estimate fixed-effects difference-in-difference models to estimate the impact of the HEW, controlling for household and individual characteristics. We examine the average effect of the HEP across the four regions; region-specific effects (by interacting the treatment variable with region-indicators); and effects across wealth quintiles (by interacting the treatment variable with the DHS-provided wealth quintile). We will examine the average effect of the HEP across the four regions; region-specific effects (by interacting the treatment variable with region-indicators); and effects across wealth quintiles (by interacting the treatment variable with the DHS-provided wealth quintile).

Outcomes (Endpoints):

We consider several categories of primary outcomes: (1) Knowledge/attitudes and health behavior, such as awareness of HIV, TB and contraception, (2) Utilization of health care services, including ante-natal and delivery care, and family planning services (3) Household-level outcomes, such as access to improved water sources or toilet facilities (4) Health outcomes, such as infant and under-5 mortality, anthropometric outcomes (stunting, wasting, underweight) and anemia To reduce the number of tests, where possible our primary measures are indices for each category that are simple averages of several individual binary measures. We also examine the individual measures of each index in secondary analyses and use (for the group of measures that are used for each index) the Benjamini, Krieger, and Yekutieli (2006) method to control the false discovery rate (FDR). See the attached document for details on the individual outcome measures and covariates, and how they are constructed.

Unit of Analysis:
Children under 5 and women aged 15-49
Hypotheses:
Unit of Intervention or Assignment:
Kebele (village)
Number of Clusters in Sample:
210 woredas (the number of kebeles is unknown at this time)
Number of Individuals in Sample:
Approximately 8,770 children over the period 2000-2011
Size of Treatment, Control, or Comparison Subsamples:
About 4,300 children or 50-60% of the sample. The actual number is unknown because the timing of the treatment (posting of the workers) is unknown until the data collection has been completed.

Supplementary Files

Analysis Plan:
Data

Outcomes Data

Description:
Ethiopia Demographic and Health Surveys 2005 and 2011. These are publicly available household surveys.
Data Already Collected?
Yes
Data Previously Used?
Yes
Data Access:
Restricted -- Access requires a formal approval process
Data Obtained by the Study Researchers?
Data Approval Process:
Registration and approval via DHS.
Approval Status:
Yes-obtained approval and have received the data

Treatment Assignment Data

Participation or Assignment Information:
No
Description:
We collect several kinds of treatment data, for each kebele (village) in a woreda (district). This includes annual measures (in December of each year) of: (a) the count of HEWs who are "on the job"; (b) the count of "active and functioning" health posts; (c) quality of the health posts' infrastructure, e.g., electricity; (d) training and services provided by the HEW across various clinical areas.
Data Obtained by the Study Researchers?
No
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Data Analysis

Data Analysis Status:

Study Materials

Upload Study Materials:

Registration Category

Registration Category:
Prospective, Category 1: Data for measuring impacts have not been collected
Completion

Completion Overview

Intervention Completion Date:
Data Collection Completion Date:
Unit of Analysis:
Clusters in Final Sample:
Total Observations in Final Sample:
Size of Treatment, Control, or Comparison Subsamples:

Findings

Preliminary Report:
Preliminary Report URL:
Summary of Findings:
Paper:
Paper Summary:
Paper Citation:

Data Availability

Data Availability (Primary Data):
Date of Data Availability:
Data URL or Contact:
Access procedure:

Other Materials

Survey:
Survey Instrument Links or Contact:
Program Files:
Program Files Links or Contact:
External Link:
External Link Description:
Description of Changes:

Study Stopped

Date:
Reason: