Objectives: This study assessed the completeness of child health records maintained and collected within community health information system in Ethiopia. Methods: A household listing was carried out in 221 enumeration areas in food insecure areas of Ethiopia to determine the presence of a child less than 24-months. This list of children was then compared against the information stored at the local health posts. A household survey was administered to a sample of 2155 households that had a child less than 24-months of age to assess determinants and consequences of exclusion from the health post registers. Results: Out of the 10,318 children identified during the listing, 36% were found from the health post records. Further analysis based on the household survey data indicated that health posts that had adopted nationally recommended recordkeeping practices had more complete records (p < 0.01) and that children residing farther from health posts were less likely to be found from the registers (p < 0.05). Mothers whose child was found from the registers were more likely to know a health extension worker (p < 0.01), had a contact with one (p < 0.01), and their child was more likely to have received growth monitoring (p < 0.05). Conclusions for Practice: The incompleteness of the data collected at the health posts poses a challenge for effective implementation of the national health extension program and various complementary programs in Ethiopia.
Ethiopia’s Health Extension Program (HEP) has been implemented by the Government of Ethiopia since 2002 and now covers nearly all woredas (districts) of the country. As an integrated program, the HEP focuses on four primary areas: hygiene and environmental sanitation, disease prevention and control, family health services and health education and communication (Workie and Ramana 2013). The HEWs stationed at the kebele (sub-district, lowest administrative unit) health posts play a key role in implementing the HEP activities. The health posts are built by the government and typically located at the kebele center. The infrastructure at the health posts is often rudimentary with many lacking access to reliable electricity, phone services and clean water (Bilal et al. 2011). There are typically two HEWs working in one health post and together these government employees are expected to reach approximately 5000 individuals (Lemma and Matji 2013). Each health post is supervised by a health center. Five health posts and one health center form a primary health care unit that serves ~ 25,000 people (Bilal et al. 2011). HEWs receive assistance from Health Development Army (HDA); a group of volunteers residing in the villages and supporting the HEWs. The Health Management Information System (HMIS) designed by the Ministry of Health plays a key role in facilitating data analysis to aid strategic planning. Since most Ethiopians reside in rural areas, the CHIS is a key component in the HMIS. Figure 1 shows the information flow within Ethiopia's Health Management Information System. HEWs collect data from their clients and store them at health posts. This is largely a paper based system, though efforts are under way to move to an electronic record keeping system in the near future (Advancing Partners and Communities 2019). Data from health posts are then sent in paper format to the health center once a month. The staff at the health centers compile the data and submit them to the woreda health office. At the woreda health office, the data form different health posts and centers are compiled and entered into a digital format. After this, these digitized data are sent to the zonal health department, which passes them on to the regional health bureau. The Federal Ministry of Health then gets the data from regional health bureaus. Information flow in Ethiopia's Health Management Information System. CHIS community health information system, woreda district. Source: Authors' construction based on FMOH (2013) At the health posts, the HEWs use three main record keeping instruments to collect data within the CHIS: Master Family Index, Family Folders and Individual Health Cards. The Master Family Index is a booklet listing all households residing in the kebele. Each household residing in the catchment area (kebele) should have a Family Folder at the health post (FMOH 2011). The Family Folder contains information on household demographics and other household characteristics relevant to health and nutrition; information on household's Water, Sanitation and Hygiene (WASH) infrastructure and the use of insecticide treated nets to prevent malaria. Inside the folder, there should be a Health Card for each household member who is five years or older. The information on the younger children are kept in the mother's Health Card until they turn five. This health card contains information on the provision of family planning services, immunization and growth monitoring, among others. As such, CHIS plays a central role in the HEWs work and implementation of the HEP at the community level. This assessment was carried out as a part of a larger evaluation of the Ethiopia's flagship safety net program—the Productive Safety Net Program (PSNP). The program currently operates in more than 300 woredas and supports about eight million people. Most PSNP beneficiaries receive cash or food payments against public works that take place over a six-month period, typically between January and June. Eligible households with limited labor capacity receive unconditional payments. The core objective of the PSNP has been to improve food security and prevent asset depletion in chronically food insecure areas of the country. Recently, the program added an increased emphasis on improving nutritional status of young children and their mothers. Therefore, the purpose of the evaluation was to assess the impact of PSNP on child and maternal nutrition. To this end, we conducted two baseline (or pre-intervention) and two endline (or post-intervention) surveys in PSNP woredas in the Ethiopian highland regions: Amhara, Oromia, Southern Nations, Nationalities, and People's Region (SNNP) and Tigray. The survey focused on households with a child less than 24-months of age. The first baseline survey was administered between the end of February and the beginning of April in 2017 and the second baseline in August 2017. The two endline surveys were administered during the same months in 2019. The current study uses the data from the first baseline survey in February-April 2017. Table Table11 provides an overview of the sampling and data collection procedures. The sampling for this survey was done in stages. First, 88 woredas were randomly selected from the full list of PSNP woredas in Amhara, Oromia, SNNP and Tigray regions. Three kebeles were randomly selected from each woreda. For census work purposes, the Central Statistical Agency (CSA) of Ethiopia divides all kebeles in the country into enumeration areas (EA) that are roughly of equal size, typically containing about 200 households. We used the CSA's EA maps to randomly select one EA from each kebele and then conducted a full listing of households residing in the EA. The primary purpose of this household listing was to form a sampling frame; to identify all households in the EA that had a child less than 24 months of age. Diagram of the sampling and data collection procedures EA enumeration area The listing and other data collection activities in each EA were carried out by a survey team consisting of one supervisor and four enumerators (22 supervisors and 88 enumerators were deployed in total). The enumerators carefully divided the EA into four areas and visited each household in the EA. A listing of one EA was typically done in one day. If nobody from the household was present at the time of the visit, the enumerators collected the required information from their neighbors or knowledgeable village guides assigned for this purpose. A total of 46,866 households residing in the 264 EAs were listed. Out of these, 10,318 households reported that they had at least one child less than 24 months of age. The presence of more than one child with less than 24 months in the same household was rare. In such cases, the enumerators were instructed to note the information of the youngest child of the household head and the spouse. Polygamy is not practiced in the localities that were surveyed. After the EA listing was completed, the survey team was left with two tasks. One enumerator took the list of names of the children identified in the EA listing to the kebele health post. At the health post, this enumerator and a HEW together compared this EA list to the information in the health information records available at the health post: Master Family Index, Family Folders and the individual Health Cards as well as any other material available at the health post. The remaining enumerators began the household level survey. This in-depth household survey was administered to 10 eligible households that were randomly selected from each EA using the information collected during the listing. A household was eligible if it had a child less than 24-months of age. We further stratified the sample so that roughly half of the selected households were PSNP beneficiaries and half of them were poor (based on their own assessment) but not benefitting from the PSNP. The survey collected information on basic household characteristics (household demographics, education attainment, wealth levels, etc.), exposure to health and nutrition services, health status, anthropometry, and so on. A total of 2635 households with a child less than 24 months of age were interviewed in 88 woredas and 264 kebeles in Amhara, Oromia, SNNP and Tigray. Of note is that the in-depth interview provided an opportunity to validate the information collected during household listing. Only in extremely rare cases (less than one percent), the household selected for the survey did not have a child in the specified age range (0–23 months). In such cases, a replacement household was drawn from the EA household list. This current study focused on a sub-sample of 221 kebeles in which the health post was also visited by the survey team. The enumerators were unable to visit all 264 health posts because some kebeles did not have a health post or because the health extension workers were absent due to training, leave or other reason. The number of households interviewed in these 221 kebeles was 2218. After removing households with missing observations, the final household sample used in the regression analysis is 2155 households. Given the focus on poor households, this household sample is not representative of the EAs or kebeles in which the sample was drawn. All data were collected via face-to-face interviews using structured questionnaires and verbal informed consent was obtained from all participants. Ethical approval was obtained from the Institutional Review Board of the International Food Policy Research Institute.
DIMA AI Care
