Implications of ethiopian productive safety net programme on household dietary diversity and women’s body mass index: A cross-sectional study

Food and Nutrition Research, Volume 62, Year 2018

Interpretation

Introduction: Poor nutritional status of women remains a critical problem in Ethiopia. Nutrition for women matters not only for the public health relevance of breaking the intergenerational cycle of malnutrition but for its high return in other sectors such as education and health. The Ethiopian Productive Safety Net Programme (PSNP) is a program that protects chronically food-insecure households against food insecurity through cash or food transfer. However, its effect on food access and women’s body mass index (BMI) has remained unexplored. Objective: This study was intended to assess differences in household dietary diversity (HDD) and women’s BMI and associated factors among PSNP and non-PSNP households. Methods: This community-based cross-sectional study was carried out in the Kombolcha District of Eastern Ethiopia from July 1 to 28, 2015. HDD and women’s BMI were compared. Ordinal logistic regression was used to identify factors associated with women’s BMI. Result: The prevalence of undernutrition was 27.3% (95% confidence interval [CI]: 23.8–30.9) and 20.2% (95% CI: 17.1–23.5) for women from PSNP and non-PSNP households, respectively. PSNP membership had a significant effect on HDD and minimal effect on women’s BMI. Ordinal logistic regression yielded significant associations for medium wealth status, with an odds ratio (OR) of 0.533 (95% CI: 0.339–0.837), uptake of better health care services compared to previous year with an odds ratio (OR) of 0.647 (95% CI: 0.429–0.974) and reduction in selling assets for the sake of buying food with an OR of 1.575 (95% CI: 1.057–2.349). Conclusion and recommendation: There was high magnitude of chronic energy deficiency among PSNP and non-PSNP households, at 27.3 and 20.2%, respectively, and it was associated with economic status and health care utilization, suggesting the need to promote profitable income-generating activities and nudging for minimum health care as a condition for transfer. A community-based cross-sectional study was carried out in the Kombolcha District of Eastern Ethiopia from July 1 to 28, 2015. This period overlapped with failed spring (mid-February to May) rain that affected crop production from the first harvest that would provide 20% of food production followed by the end of 6 months of PSNP cash transfer (28). The district contains 19 kebeles (smallest administrative units in Ethiopia next to districts), out of which 10 are non-beneficiary and 9 kebeles (total of 2,375 households) benefit from cash transfers. This translates to about 9,752 people who receive cash in exchange for participating in public works and 1,409 people with direct support. For this study, five PSNP and six non-PSNP kebeles were selected randomly, and only public works participants were included in the study. Though fairness and transparency is the core principle of PSNP client selection, there are inclusion and exclusion errors. Corrupt officials, clan politics, and quota allocation were the main causes of inclusion and exclusion errors (22, 29). To obtain data with a low bias estimate, firstly, the data collection was carefully planned to include the same variables by using similar data collection tools and procedures for beneficiary and non-beneficiary households. Secondly, outcomes related to program participation were identified using key PSNP-related variables (livestock ownership, household landholding, access to government health post, asset depletion and food aid, and asset losses) that identify outcomes related to women’s nutrition and other related variables. This information was obtained from the kebele food security task force (KFSTF), which has seven members, including a health extension worker. Thirdly, to attain comparable access to market systems, similar livelihood zones known for khat and vegetable production were selected. These livelihood zones had similar agro-ecology and production patterns of these commercial crops; the participants had common livelihood strategies and comparable access to markets, including distance from the market. In this district, cash was provided because the markets functioned well. Information about women was collected during the mother’s interview for eligible children aged 6 months to 5 years (information on children being processed in another publication). Hence, participants were selected from five randomly selected PSNP and six non-PSNP kebeles. Women eligible for a child interview were identified from lists obtained from the district PSNP office compiled by KFSTF and respective kebele health extension workers. Non-PSNP kebeles have similar KFSTFs that follow the same procedure to identify food insecure clients. Both PSNP and non-PSNP household lists are finally ascertained by social networks leaders called gare (groups containing 25–30 women). In order to minimize handout expectations and a spillover effect of the transfer, women from non-PSNP beneficiary households were entirely selected from non-beneficiary kebeles. Pregnant women and direct support beneficiaries were excluded from this study. A structured pretested questionnaire was used to assess socioeconomic and demographic characteristics of the households. Nursing students who could speak a local language (Afaan Oromo) were trained to collected data. The tool was pretested on 20 households to determine its suitability to local accent, format, wording, and order. In addition, periodic checking of the weighing scale and repeated measurement were used to assure the data quality. Ethical clearance was obtained from the Haramaya University College of Health and Medical Science Institutional Health Research Ethics Review Committee. The objective of the study, known benefits, and risks of participant involvement in the research were communicated. Informed written and signed consent was obtained from women before commencing the study. The primary outcome of this study was women’s BMI. The secondary outcome was Household Dietary Diversity Score (HDDS). In the statistical analyses, the factor considered as a potential confounder was maternal age. Factors considered as potential effect modifiers were the sex of head of household and PSNP beneficiary status. BMI is a proxy indicator of energy status (undernutrition), calculated as weight (kg) divided by the square of height (m2). Women’s height was measured to the nearest 0.5 cm without shoes, feet flat, heels together, legs straight using a portable wooden height-measuring board with a sliding head bar following standard anthropometric techniques. Heights <145 cm were classified as stunted. Weight was measured repeatedly to the nearest 100 g using an electronic scale (SECA, Hamburg, Germany). A BMI of 17–18.4 indicates marginal energy deficiency, 16 to <17 moderate energy deficiency, and that of less than 16.0 indicates severe chronic energy deficiency. A BMI of ≥25 signifies overweight, and >30 signifies obesity. Even though a global database on women nutrition is not available, a BMI of 20–25 kg/m2 is recommended for good health and is associated with normal fertility. A weight for height equivalent to a BMI of 18 kg/m2 or lower is considered too low for successful reproductive ability (30). The HDD score is a measure of the total number of different food groups consumed in the last 24 hours by household members with a well-grounded construction of diet quality and accuracy, cross-checked with incomes. HDDS ranges from 0 to 12, the higher the better, and it is a good indicator of both quantity and quality. It is included in the acute food insecurity reference table for household group classification of the Integrated Food Security Phase Classification (IPC). HDDS does not have established categorical cutoffs and is analyzed only as a scale measure. A face-to-face interview was used to administer the tool. For households with unusual food intake in the previous 24 hours, another appointment was made for the interview. Due emphasis was placed on acquiring a response with minimal social desirability bias (31–33). Household wealth is a proxy measure of household income for long-term wealth. Principal components analysis was run using 38 items comprising productive assets, livestock, household goods, and consumer durables. It was used as a continuous variable, and each household was classified as being in the lowest, middle, or highest asset category. Analysis was performed on data that were already available for child wasting. Excluding 52 women, the final sample size was 623 women from PSNP and 635 non-PSNP (total 1,258). This sample size is sufficient for the analysis of the data to produce results with sufficient statistical precision. Data were entered in EpiData 3.1 and the software package SPSS version 23 for Windows was used for statistical analysis. To examine whether associations differed across groups, stratification was done based on PSNP and wealth index. Descriptive statistical analysis was conducted to describe the characteristics of participants. For constructing wealth index based on 38 items, the selection of each factors was based on the rotated component matrix of greater than 0.5. One-way Analysis of Variance (ANOVA) was conducted. The independent-samples t-test was used to compare mean HDDS across PSNP and other variables. In order to check whether the assumptions of Multivariate analysis of variance (MANOVA) were met, preliminary assumption testing for normality, linearity, univariate and multivariate outliers, homogeneity of variance–covariance matrices, and multicollinearity were conducted. No significant violation was found. Further, an ordinal logistic regression model was used for prediction of women’s BMI (dependent variable). The odds ratio (OR) was used as the primary measure of strength and direction of the relationship between each independent variable and the women’s BMI values, which were categorized into underweight (BMI<18.4), normal (BMI 18.5–24.9), and overweight (BMI ≥25). In this analysis, OR less than 1 indicated a negative relationship.

AI Digest

Study Justification:

– Poor nutritional status of women in Ethiopia is a critical problem with implications for public health, education, and health sectors.
– The Ethiopian Productive Safety Net Programme (PSNP) aims to protect food-insecure households, but its effect on food access and women’s body mass index (BMI) has not been explored.
– Understanding the differences in household dietary diversity (HDD) and women’s BMI between PSNP and non-PSNP households is important for identifying potential interventions to improve nutrition.

Study Highlights:

– The prevalence of undernutrition was 27.3% for women from PSNP households and 20.2% for women from non-PSNP households.
– PSNP membership had a significant effect on HDD and a minimal effect on women’s BMI.
– Factors associated with women’s BMI included medium wealth status, uptake of better health care services compared to the previous year, and reduction in selling assets for food.

Study Recommendations:

– Promote profitable income-generating activities to improve economic status and food access.
– Encourage minimum health care utilization as a condition for transfer to improve women’s nutrition.
– Nudge for better health care services and discourage selling assets for food.

Key Role Players:

– Government officials responsible for implementing and monitoring the PSNP.
– Health extension workers to provide health care services and education.
– Community leaders and social networks leaders to ensure fairness and transparency in beneficiary selection.

Cost Items for Planning Recommendations:

– Funding for income-generating activities and training programs.
– Budget for improving health care services and infrastructure.
– Resources for monitoring and evaluation of the PSNP implementation.
– Costs associated with awareness campaigns and community engagement activities.

Strength of Evidence

The strength of evidence for this abstract is 7 out of 10.
The evidence in the abstract is moderately strong. The study design is a community-based cross-sectional study, which provides valuable insights into the topic. The sample size is sufficient for statistical analysis. The study compares household dietary diversity and women’s BMI between PSNP and non-PSNP households, and identifies factors associated with women’s BMI. However, the abstract does not provide information on the representativeness of the sample or the generalizability of the findings. To improve the strength of the evidence, future studies could consider using a larger and more diverse sample, and employing a longitudinal design to assess the long-term effects of the PSNP program on women’s nutrition. Additionally, providing information on the limitations of the study and potential sources of bias would enhance the transparency and reliability of the findings.

Abstract

A community-based cross-sectional study was carried out in the Kombolcha District of Eastern Ethiopia from July 1 to 28, 2015. This period overlapped with failed spring (mid-February to May) rain that affected crop production from the first harvest that would provide 20% of food production followed by the end of 6 months of PSNP cash transfer (28). The district contains 19 kebeles (smallest administrative units in Ethiopia next to districts), out of which 10 are non-beneficiary and 9 kebeles (total of 2,375 households) benefit from cash transfers. This translates to about 9,752 people who receive cash in exchange for participating in public works and 1,409 people with direct support. For this study, five PSNP and six non-PSNP kebeles were selected randomly, and only public works participants were included in the study. Though fairness and transparency is the core principle of PSNP client selection, there are inclusion and exclusion errors. Corrupt officials, clan politics, and quota allocation were the main causes of inclusion and exclusion errors (22, 29). To obtain data with a low bias estimate, firstly, the data collection was carefully planned to include the same variables by using similar data collection tools and procedures for beneficiary and non-beneficiary households. Secondly, outcomes related to program participation were identified using key PSNP-related variables (livestock ownership, household landholding, access to government health post, asset depletion and food aid, and asset losses) that identify outcomes related to women’s nutrition and other related variables. This information was obtained from the kebele food security task force (KFSTF), which has seven members, including a health extension worker. Thirdly, to attain comparable access to market systems, similar livelihood zones known for khat and vegetable production were selected. These livelihood zones had similar agro-ecology and production patterns of these commercial crops; the participants had common livelihood strategies and comparable access to markets, including distance from the market. In this district, cash was provided because the markets functioned well. Information about women was collected during the mother’s interview for eligible children aged 6 months to 5 years (information on children being processed in another publication). Hence, participants were selected from five randomly selected PSNP and six non-PSNP kebeles. Women eligible for a child interview were identified from lists obtained from the district PSNP office compiled by KFSTF and respective kebele health extension workers. Non-PSNP kebeles have similar KFSTFs that follow the same procedure to identify food insecure clients. Both PSNP and non-PSNP household lists are finally ascertained by social networks leaders called gare (groups containing 25–30 women). In order to minimize handout expectations and a spillover effect of the transfer, women from non-PSNP beneficiary households were entirely selected from non-beneficiary kebeles. Pregnant women and direct support beneficiaries were excluded from this study. A structured pretested questionnaire was used to assess socioeconomic and demographic characteristics of the households. Nursing students who could speak a local language (Afaan Oromo) were trained to collected data. The tool was pretested on 20 households to determine its suitability to local accent, format, wording, and order. In addition, periodic checking of the weighing scale and repeated measurement were used to assure the data quality. Ethical clearance was obtained from the Haramaya University College of Health and Medical Science Institutional Health Research Ethics Review Committee. The objective of the study, known benefits, and risks of participant involvement in the research were communicated. Informed written and signed consent was obtained from women before commencing the study. The primary outcome of this study was women’s BMI. The secondary outcome was Household Dietary Diversity Score (HDDS). In the statistical analyses, the factor considered as a potential confounder was maternal age. Factors considered as potential effect modifiers were the sex of head of household and PSNP beneficiary status. BMI is a proxy indicator of energy status (undernutrition), calculated as weight (kg) divided by the square of height (m2). Women’s height was measured to the nearest 0.5 cm without shoes, feet flat, heels together, legs straight using a portable wooden height-measuring board with a sliding head bar following standard anthropometric techniques. Heights <145 cm were classified as stunted. Weight was measured repeatedly to the nearest 100 g using an electronic scale (SECA, Hamburg, Germany). A BMI of 17–18.4 indicates marginal energy deficiency, 16 to 30 signifies obesity. Even though a global database on women nutrition is not available, a BMI of 20–25 kg/m2 is recommended for good health and is associated with normal fertility. A weight for height equivalent to a BMI of 18 kg/m2 or lower is considered too low for successful reproductive ability (30). The HDD score is a measure of the total number of different food groups consumed in the last 24 hours by household members with a well-grounded construction of diet quality and accuracy, cross-checked with incomes. HDDS ranges from 0 to 12, the higher the better, and it is a good indicator of both quantity and quality. It is included in the acute food insecurity reference table for household group classification of the Integrated Food Security Phase Classification (IPC). HDDS does not have established categorical cutoffs and is analyzed only as a scale measure. A face-to-face interview was used to administer the tool. For households with unusual food intake in the previous 24 hours, another appointment was made for the interview. Due emphasis was placed on acquiring a response with minimal social desirability bias (31–33). Household wealth is a proxy measure of household income for long-term wealth. Principal components analysis was run using 38 items comprising productive assets, livestock, household goods, and consumer durables. It was used as a continuous variable, and each household was classified as being in the lowest, middle, or highest asset category. Analysis was performed on data that were already available for child wasting. Excluding 52 women, the final sample size was 623 women from PSNP and 635 non-PSNP (total 1,258). This sample size is sufficient for the analysis of the data to produce results with sufficient statistical precision. Data were entered in EpiData 3.1 and the software package SPSS version 23 for Windows was used for statistical analysis. To examine whether associations differed across groups, stratification was done based on PSNP and wealth index. Descriptive statistical analysis was conducted to describe the characteristics of participants. For constructing wealth index based on 38 items, the selection of each factors was based on the rotated component matrix of greater than 0.5. One-way Analysis of Variance (ANOVA) was conducted. The independent-samples t-test was used to compare mean HDDS across PSNP and other variables. In order to check whether the assumptions of Multivariate analysis of variance (MANOVA) were met, preliminary assumption testing for normality, linearity, univariate and multivariate outliers, homogeneity of variance–covariance matrices, and multicollinearity were conducted. No significant violation was found. Further, an ordinal logistic regression model was used for prediction of women’s BMI (dependent variable). The odds ratio (OR) was used as the primary measure of strength and direction of the relationship between each independent variable and the women’s BMI values, which were categorized into underweight (BMI<18.4), normal (BMI 18.5–24.9), and overweight (BMI ≥25). In this analysis, OR less than 1 indicated a negative relationship.

Materials

N/A

Innovations

Based on the provided description, here are some potential innovations that could improve access to maternal health:

1. Mobile Health (mHealth) Applications: Develop mobile applications that provide pregnant women with access to important health information, such as nutrition guidelines, prenatal care schedules, and reminders for medication or doctor appointments. These apps can also include features for tracking maternal health indicators, such as BMI, and provide personalized recommendations.

2. Telemedicine Services: Implement telemedicine services that allow pregnant women in remote or underserved areas to consult with healthcare professionals through video calls or phone consultations. This can help overcome geographical barriers and provide timely advice and support for maternal health concerns.

3. Community Health Workers: Train and deploy community health workers who can visit pregnant women in their homes to provide education, support, and monitoring of maternal health. These workers can help identify risk factors, provide guidance on nutrition and healthy habits, and refer women to appropriate healthcare facilities when needed.

4. Cash Transfer Programs: Expand and improve cash transfer programs, like the Ethiopian Productive Safety Net Programme (PSNP), to ensure that pregnant women have access to adequate nutrition and healthcare. This can help alleviate financial barriers to maternal health services and improve overall maternal health outcomes.

5. Public-Private Partnerships: Foster collaborations between public and private sectors to improve access to maternal health services. This can involve partnering with private healthcare providers to offer affordable or subsidized services, leveraging technology and innovation from the private sector, and implementing public awareness campaigns to promote maternal health.

6. Maternal Health Vouchers: Introduce voucher programs that provide pregnant women with access to essential maternal health services, such as prenatal care, delivery, and postnatal care. These vouchers can be distributed through community health centers or local organizations and can help reduce financial barriers to accessing quality care.

7. Maternal Health Education Programs: Develop and implement comprehensive maternal health education programs that target women, families, and communities. These programs can focus on raising awareness about the importance of maternal health, promoting healthy behaviors during pregnancy, and addressing cultural or social barriers that may hinder access to care.

8. Transportation Support: Establish transportation support systems, such as community-based transportation networks or partnerships with local transportation providers, to ensure that pregnant women can easily access healthcare facilities for prenatal visits, delivery, and emergency care.

9. Maternal Health Hotlines: Set up toll-free hotlines staffed by trained healthcare professionals who can provide information, advice, and support to pregnant women. These hotlines can be available 24/7 and can help address concerns, provide guidance on emergency situations, and connect women to appropriate healthcare services.

10. Maternal Health Monitoring Systems: Implement digital health solutions, such as electronic health records and data analytics, to improve the monitoring and tracking of maternal health indicators. These systems can help identify trends, gaps in care, and areas for improvement, ultimately leading to more targeted interventions and improved access to maternal health services.

AI Innovations Description

Based on the information provided, the study conducted in the Kombolcha District of Eastern Ethiopia aimed to assess the differences in household dietary diversity (HDD) and women’s body mass index (BMI) between households participating in the Ethiopian Productive Safety Net Programme (PSNP) and non-PSNP households. The study found that there was a high prevalence of undernutrition among both PSNP and non-PSNP households, with 27.3% and 20.2% respectively. The study also identified factors associated with women’s BMI, including medium wealth status, uptake of better healthcare services compared to the previous year, and reduction in selling assets for the sake of buying food.

Based on these findings, the study recommends several actions to improve access to maternal health and address the issue of undernutrition among women. These recommendations include:

1. Promoting profitable income-generating activities: Since economic status was found to be associated with undernutrition, promoting income-generating activities can help improve household income and food security, thereby reducing the prevalence of undernutrition.

2. Nudging for minimum healthcare as a condition for transfer: The study found that better healthcare utilization was associated with lower odds of undernutrition. Therefore, implementing measures to ensure that healthcare services are accessible and utilized by women can contribute to improving their nutritional status.

3. Strengthening the implementation of the Ethiopian Productive Safety Net Programme: The study highlighted the need to address inclusion and exclusion errors in the selection process of PSNP beneficiaries. Strengthening the implementation of the program can ensure that it effectively targets and supports food-insecure households.

4. Enhancing dietary diversity: Since household dietary diversity was found to be associated with women’s nutritional status, interventions that promote diverse and nutritious diets can help improve women’s health and reduce the prevalence of undernutrition.

Overall, these recommendations aim to address the underlying factors contributing to undernutrition among women and improve access to maternal health in Ethiopia.

AI Innovations Methodology

Based on the provided description, here are some potential recommendations to improve access to maternal health:

1. Strengthen the Ethiopian Productive Safety Net Programme (PSNP): Enhance the effectiveness of the PSNP in addressing food insecurity and improving access to nutritious food for pregnant women and new mothers. This can be done by increasing the cash or food transfer provided to chronically food-insecure households and ensuring fair and transparent client selection processes.

2. Promote income-generating activities: Implement programs that promote profitable income-generating activities for women, particularly those from food-insecure households. This can help improve economic status and increase access to nutritious food, ultimately improving maternal health.

3. Improve access to healthcare services: Increase access to better healthcare services for women, including prenatal care, postnatal care, and nutrition counseling. This can be achieved by strengthening the healthcare infrastructure in the targeted areas and ensuring that women are aware of and utilize these services.

4. Nudge for minimum healthcare as a condition for transfer: Introduce a requirement for PSNP beneficiaries to utilize minimum healthcare services as a condition for receiving cash or food transfers. This can help improve women’s health outcomes by ensuring regular check-ups and access to necessary healthcare interventions.

To simulate the impact of these recommendations on improving access to maternal health, a methodology could be developed as follows:

1. Define the indicators: Identify key indicators that reflect access to maternal health, such as the percentage of pregnant women receiving prenatal care, the percentage of women with adequate nutrition during pregnancy, and the percentage of women accessing postnatal care.

2. Collect baseline data: Gather baseline data on the identified indicators before implementing the recommendations. This can be done through surveys, interviews, and data collection from healthcare facilities and relevant government agencies.

3. Implement the recommendations: Roll out the recommended interventions, such as strengthening the PSNP, promoting income-generating activities, and improving access to healthcare services. Ensure that these interventions are implemented consistently and monitored closely.

4. Collect post-intervention data: After a sufficient period of time, collect post-intervention data on the same indicators used for the baseline assessment. This can be done using the same methods as the baseline data collection.

5. Analyze and compare data: Analyze the baseline and post-intervention data to determine the impact of the recommendations on improving access to maternal health. Compare the indicators before and after the interventions to identify any significant changes or improvements.

6. Evaluate the results: Assess the findings of the data analysis to evaluate the effectiveness of the recommendations. Determine if the interventions have led to improved access to maternal health and identify any areas that may require further attention or adjustment.

7. Adjust and refine interventions: Based on the evaluation results, make any necessary adjustments or refinements to the interventions to further improve access to maternal health. This could involve scaling up successful interventions, addressing any identified challenges, or introducing additional strategies.

8. Monitor and track progress: Continuously monitor and track the progress of the interventions to ensure sustained improvements in access to maternal health. Regularly collect and analyze data to assess the ongoing impact and make any necessary adjustments to maintain positive outcomes.

By following this methodology, it is possible to simulate the impact of the recommendations on improving access to maternal health and make informed decisions for future interventions.

Author

Dima Health

Statistics:

Citations: 7

Identifiers:

DOI: 10.29219/fnr.v62.1574

ISSN: 16546628

Research Areas:

Community Interventions, Disparities, Food Security, Health System and Policy, Maternal Access, Maternal and Child Health, Noncommunicable Diseases, Quality of Care, Sexual and Reproductive Health, Social Determinants

Study Countries:

Ethiopia

Study Design:

Case-Control Study, Cross Sectional Study, Grounded Theory

Study Approach:

Quantitative

Participants Gender:

Female