Inequities in skilled attendance at birth in Namibia: A decomposition analysis

BMC Pregnancy and Childbirth, Volume 11, Year 2011

Interpretation

Background: The fifth Millennium Development Goal (MDG5) aims at improving maternal health. Globally, the maternal mortality ratio (MMR) declined from 400 to 260 per 100000 live births between 1990 and 2008. During the same period, MMR in sub-Saharan Africa decreased from 870 to 640. The decreased in MMR has been attributed to increase in the proportion of deliveries attended by skilled health personnel. Global improvements maternal health and health service provision indicators mask inequalities both between and within countries. In Namibia, there are significant inequities in births attended by skilled providers that favour those that are economically better off. The objective of this study was to identify the drivers of wealth-related inequalities in child delivery by skilled health providers.Methods: Namibia Demographic and Health Survey data of 2006-07 are analysed for the causes of inequities in skilled birth attendance using a decomposable health concentration index and the framework of the Commission on Social Determinants of Health.Results: About 80.3% of the deliveries were attended by skilled health providers. Skilled birth attendance in the richest quintile is about 70% more than that of the poorest quintile. The rate of skilled attendance among educated women is almost twice that of women with no education. Furthermore, women in urban areas access the services of trained birth attendant 30% more than those in rural areas. Use of skilled birth attendants is over 90% in Erongo, Hardap, Karas and Khomas Regions, while the lowest (about 60-70%) is seen in Kavango, Kunene and Ohangwena. The concentration curve and concentration index show statistically significant wealth-related inequalities in delivery by skilled providers that are to the advantage of women from economically better off households (C = 0.0979; P < 0.001).Delivery by skilled health provider by various maternal and household characteristics was 21 percentage points higher in urban than rural areas; 39 percentage points higher among those in richest wealth quintile than the poorest; 47 percentage points higher among mothers with higher level of education than those with no education; 5 percentage points higher among female headed households than those headed by men; 20 percentage points higher among people with health insurance cover than those without; and 31 percentage points higher in Karas region than Kavango region.Conclusion: Inequalities in wealth and education of the mother are seen to be the main drivers of inequities in the percentage of births attended by skilled health personnel. This clearly implies that addressing inequalities in access to child delivery services should not be confined to the health system and that a concerted multi-sectoral action is needed in line with the principles of the Primary health Care. © 2011 Zere et al; licensee BioMed Central Ltd. In measuring equity in a health outcome or access to health interventions, the following are required: - indicator of the health intervention of interest (delivery by skilled health providers) - a variable (stratifier) capturing socio-economic status against which the distribution is to be assessed (wealth); and - a measure of socio-economic inequality to quantify the degree of inequity in the indicator variable of interest. A concentration index (C) is used to measure wealth-related inequalities in the observed use of delivery services by skilled health providers. The concentration index of a health care variable y (utilization of delivery services by trained health providers) can be defined using the concentration curve that links the cumulative proportion of individuals ranked by wealth to the corresponding cumulative proportion of y (use of delivery services by trained health providers). The concentration curve plots shares of the health care variable (y) against quantiles of the measure of socio-economic status (asset-based wealth index) [17]. The concentration index is defined as twice the area between the concentration curve and the line of equality and assumes values between -1 and +1. A negative value of the concentration index denotes inequity in skilled care at birth that is to the advantage of the lower wealth quintiles implying that women of lower socio-economic status are delivered by skilled health providers more than their counterparts who are wealthier. In this case the concentration curve lies above the line of equality. On the other hand, a positive concentration index implies inequality in the use of delivery services by skilled providers that favours women who are wealthier (the concentration curve lies below the line of equality). When the value of the concentration index is zero, there are no wealth related inequalities in the use of delivery services by skilled providers. The concentration curve overlaps with the 45-degree line. From individual level data, the concentration index can be computed using the following formula [18]: Where hi is the health variable of interest (delivery by skilled health providers); μ is the mean of hi; Ri is the fractional rank of individual i in the distribution of socio-economic position; and (; i = 1 for the poorest and i = n for the richest). Wagstaff et al. [19] demonstrated that the concentration index of a health variable is additively decomposable to the concentration indices of the determinants of that health variable. In other words, the concentration index of the health variable of interest can be expressed as the sum of the contributions of the various determinants of that variable, together with unexplained residual component. In decomposing the concentration index of delivery by skilled providers, the following steps are pursued: 1. Regressing the health variable against its determinants: Where: yi = 1 if the delivery was conducted by a skilled health provider; xk: a set of exogenous determinants of delivery by trained health provider; βk: coefficient of determinant xk; and εi: random error term. The dependent variable (delivery by skilled health personnel) is a binary variable with values of 1 (delivered by skilled provider) and 0 (otherwise). The linear probability model (LPM) in Equation 2 above has been used in order to satisfy the linearity assumption of the decomposition analysis, although the estimates are inefficient and the probability of delivery by skilled health providers may not fall within the conventional values of 0 ≤ p ≤ 1 and has heteroskedastic errors [20]. However, the estimated probabilities from the LPM model have been constrained within the conventional values and a comparison with a probit model has not shown significant variations between the coefficients of the LPM and the marginal (or average) effects of the probit regression derived using the dprobit Stata command [17]. Furthermore, to adjust for heteroskedasticity, the predicted values from the regression model have been saved and used as weights to run weighted least squares (WLS) using the "aweight" option in Stata [21] 2. Calculating concentration indices for the health variable and for its determinants (and generalized concentration index of the error term): For any linearly additive regression model of the health variable of interest (yi) such as Equation 2 above, the concentration index for y, can be written as: Where: Cy: concentration index of skilled care at birth (i.e. concentration index of yi); : mean value of determinant xk; μ: mean of the outcome variable yi - that is the mean of deliveries by skilled health providers ck: concentration index of determinant xk GCε: residual component that captures wealth-related inequality in skilled care at birth that is not accounted for by systematic variation in determinants across wealth groups. The term in parenthesis in Equation 3 above expresses the impact of each determinant on the probability of delivery by skilled health providers. In other words, it denotes the elasticity (ηk) of the outcome variable (delivery by skilled health providers) with respect to the determinant xk evaluated at the mean value of yi (delivery by skilled health providers). The concentration index of delivery by skilled health providers is thus a weighted sum of the inequality in each of its determinants, with the weights equal to the elasticities of the determinants: Report of the WHO Commission on the Social Determinants of Health revitalized the need for sustained and concerted efforts to achieve health equity through action on the social determinants of health. The Commission's social determinants framework takes a holistic view of inequities in health and health care within and between countries. Inequities in health/healthcare are caused by the unequal distribution of power, income, goods and services nationally and internationally (Figure (Figure1)1) [22]. Commission on Social Determinants of health conceptual framework. The social determinants of health are the circumstances in which people are born, grow up, live, work and age, and the systems put in place to deal with illness. These circumstances are in turn influenced by a wider set of forces: economics, social policies and politics [22]. The social determinants framework suggests that interventions to address health inequities have to be geared towards: 1. The circumstances of daily living, which include: differential exposure to health risks in early life, the social and physical environments and work associated with social stratification; and health care responses. 2. Structural drivers including the nature and degree of social stratification; biases, norms and values within society; global and national economic and social policy; and processes of governance at all levels. As observed in Figure Figure1,1, the health system is an important social determinant of health influenced by and influencing the other social determinants. However, the health system is not the only social determinant of health. The effect of the each of the factors in Figure Figure11 in the genesis and perpetuation of health/health care inequities may vary from one country to another. It is therefore important to try to identify the effect of the various social determinants of health on health outcomes and access to health care in order to design evidence-based interventions and policy instruments. Data from the Namibia Demographic and Health Survey 2006-07 was used for this study. The data is available on the MEASURE DHS website for registered users. In the linear probability model of the determinants of delivery by skilled health providers and the decomposition analysis the following variables have been used: 1. Dependent variable: delivery by skilled health providers, which takes a value of 1 if the delivery has been attended by skilled health providers and a value of zero otherwise. 2. Independent variables: • Region; • Place of residence - urban/rural; • Wealth as computed from the asset indices; • Education of mother in years of schooling completed • Head of household - a dummy where female household assumes a value of one; and • Insurance coverage - a dummy with a value of one if the woman has insurance coverage. In NDHS 2006-07, a representative two-stage probability sample of 10,000 households was selected. The first stage consisted of selection of 500 primary sampling units (PSUs) from a sampling frame of 3,750 PSUs with probability proportional to size; the size being the number of households in the 2001 Population Census. The second stage involved the systematic selection of 20 households in each PSU [14]. The demographic and health surveys do not contain data on household income or consumption expenditure. Instead wealth index is used as a proxy. The wealth index is based on household ownership of consumer goods (such as radio, television); dwelling characteristics; type of drinking water source; toilet facilities and other characteristics related to the household's socio-economic status. The asset indices are constructed using the method of principal component analysis (PCA) [14]. Studies have shown a close relationship between asset ownership and consumption expenditure in developing countries [23] and that household asset is a good indicator of the long-run economic status of households [24] Data was analyzed using STATA 10 statistical software and MS Excel.

AI Digest

Study Justification:

This study aims to address the inequities in skilled attendance at birth in Namibia. The Millennium Development Goal 5 focuses on improving maternal health, and it is important to understand the factors contributing to disparities in access to skilled health providers during childbirth. By identifying the drivers of wealth-related inequalities in childbirth attendance, this study can provide valuable insights for policymakers and healthcare providers to develop targeted interventions and policies to improve access to skilled birth attendants.

Highlights:

– The study found that 80.3% of deliveries in Namibia were attended by skilled health providers.
– There are significant wealth-related inequalities in access to skilled birth attendants, with the richest quintile having 70% more attendance than the poorest quintile.
– Education and urban residence also play a role in access to skilled birth attendants, with educated women and those in urban areas having higher rates of attendance.
– There are regional variations in access, with higher attendance rates in certain regions such as Erongo, Hardap, Karas, and Khomas.
– The concentration index, which measures wealth-related inequalities, showed statistically significant disparities favoring women from economically better off households.

Recommendations:

– Addressing inequalities in access to skilled birth attendants should not be limited to the health system alone. A multi-sectoral approach is needed, in line with the principles of Primary Health Care.
– Policies and interventions should focus on reducing wealth-related disparities, improving education opportunities for women, and increasing access to skilled birth attendants in rural areas and regions with lower attendance rates.

Key Role Players:

– Ministry of Health and Social Services: Responsible for implementing policies and interventions to improve access to skilled birth attendants.
– Ministry of Education: Plays a role in improving education opportunities for women, which can contribute to higher attendance rates.
– Regional Health Directorates: Responsible for implementing interventions at the regional level and addressing specific regional disparities.
– Non-governmental organizations (NGOs): Can provide support and resources for implementing interventions and raising awareness about the importance of skilled birth attendance.

Cost Items for Planning Recommendations:

– Training and capacity building for healthcare providers to ensure they have the necessary skills to provide skilled birth attendance.
– Infrastructure development and improvement in rural areas to increase access to skilled birth attendants.
– Education programs and initiatives to improve educational opportunities for women.
– Outreach and awareness campaigns to educate communities about the importance of skilled birth attendance and reduce cultural barriers.
– Monitoring and evaluation systems to track progress and ensure the effectiveness of interventions.

Strength of Evidence

The strength of evidence for this abstract is 8 out of 10.
The evidence in the abstract is strong, as it provides detailed information on the methodology used, the data analyzed, and the results obtained. However, to improve the evidence, it would be helpful to include information on the sample size and the statistical significance of the findings.

Abstract

In measuring equity in a health outcome or access to health interventions, the following are required: – indicator of the health intervention of interest (delivery by skilled health providers) – a variable (stratifier) capturing socio-economic status against which the distribution is to be assessed (wealth); and – a measure of socio-economic inequality to quantify the degree of inequity in the indicator variable of interest. A concentration index (C) is used to measure wealth-related inequalities in the observed use of delivery services by skilled health providers. The concentration index of a health care variable y (utilization of delivery services by trained health providers) can be defined using the concentration curve that links the cumulative proportion of individuals ranked by wealth to the corresponding cumulative proportion of y (use of delivery services by trained health providers). The concentration curve plots shares of the health care variable (y) against quantiles of the measure of socio-economic status (asset-based wealth index) [17]. The concentration index is defined as twice the area between the concentration curve and the line of equality and assumes values between -1 and +1. A negative value of the concentration index denotes inequity in skilled care at birth that is to the advantage of the lower wealth quintiles implying that women of lower socio-economic status are delivered by skilled health providers more than their counterparts who are wealthier. In this case the concentration curve lies above the line of equality. On the other hand, a positive concentration index implies inequality in the use of delivery services by skilled providers that favours women who are wealthier (the concentration curve lies below the line of equality). When the value of the concentration index is zero, there are no wealth related inequalities in the use of delivery services by skilled providers. The concentration curve overlaps with the 45-degree line. From individual level data, the concentration index can be computed using the following formula [18]: Where hi is the health variable of interest (delivery by skilled health providers); μ is the mean of hi; Ri is the fractional rank of individual i in the distribution of socio-economic position; and (; i = 1 for the poorest and i = n for the richest). Wagstaff et al. [19] demonstrated that the concentration index of a health variable is additively decomposable to the concentration indices of the determinants of that health variable. In other words, the concentration index of the health variable of interest can be expressed as the sum of the contributions of the various determinants of that variable, together with unexplained residual component. In decomposing the concentration index of delivery by skilled providers, the following steps are pursued: 1. Regressing the health variable against its determinants: Where: yi = 1 if the delivery was conducted by a skilled health provider; xk: a set of exogenous determinants of delivery by trained health provider; βk: coefficient of determinant xk; and εi: random error term. The dependent variable (delivery by skilled health personnel) is a binary variable with values of 1 (delivered by skilled provider) and 0 (otherwise). The linear probability model (LPM) in Equation 2 above has been used in order to satisfy the linearity assumption of the decomposition analysis, although the estimates are inefficient and the probability of delivery by skilled health providers may not fall within the conventional values of 0 ≤ p ≤ 1 and has heteroskedastic errors [20]. However, the estimated probabilities from the LPM model have been constrained within the conventional values and a comparison with a probit model has not shown significant variations between the coefficients of the LPM and the marginal (or average) effects of the probit regression derived using the dprobit Stata command [17]. Furthermore, to adjust for heteroskedasticity, the predicted values from the regression model have been saved and used as weights to run weighted least squares (WLS) using the "aweight" option in Stata [21] 2. Calculating concentration indices for the health variable and for its determinants (and generalized concentration index of the error term): For any linearly additive regression model of the health variable of interest (yi) such as Equation 2 above, the concentration index for y, can be written as: Where: Cy: concentration index of skilled care at birth (i.e. concentration index of yi); : mean value of determinant xk; μ: mean of the outcome variable yi – that is the mean of deliveries by skilled health providers ck: concentration index of determinant xk GCε: residual component that captures wealth-related inequality in skilled care at birth that is not accounted for by systematic variation in determinants across wealth groups. The term in parenthesis in Equation 3 above expresses the impact of each determinant on the probability of delivery by skilled health providers. In other words, it denotes the elasticity (ηk) of the outcome variable (delivery by skilled health providers) with respect to the determinant xk evaluated at the mean value of yi (delivery by skilled health providers). The concentration index of delivery by skilled health providers is thus a weighted sum of the inequality in each of its determinants, with the weights equal to the elasticities of the determinants: Report of the WHO Commission on the Social Determinants of Health revitalized the need for sustained and concerted efforts to achieve health equity through action on the social determinants of health. The Commission's social determinants framework takes a holistic view of inequities in health and health care within and between countries. Inequities in health/healthcare are caused by the unequal distribution of power, income, goods and services nationally and internationally (Figure (Figure1)1) [22]. Commission on Social Determinants of health conceptual framework. The social determinants of health are the circumstances in which people are born, grow up, live, work and age, and the systems put in place to deal with illness. These circumstances are in turn influenced by a wider set of forces: economics, social policies and politics [22]. The social determinants framework suggests that interventions to address health inequities have to be geared towards: 1. The circumstances of daily living, which include: differential exposure to health risks in early life, the social and physical environments and work associated with social stratification; and health care responses. 2. Structural drivers including the nature and degree of social stratification; biases, norms and values within society; global and national economic and social policy; and processes of governance at all levels. As observed in Figure Figure1,1, the health system is an important social determinant of health influenced by and influencing the other social determinants. However, the health system is not the only social determinant of health. The effect of the each of the factors in Figure Figure11 in the genesis and perpetuation of health/health care inequities may vary from one country to another. It is therefore important to try to identify the effect of the various social determinants of health on health outcomes and access to health care in order to design evidence-based interventions and policy instruments. Data from the Namibia Demographic and Health Survey 2006-07 was used for this study. The data is available on the MEASURE DHS website for registered users. In the linear probability model of the determinants of delivery by skilled health providers and the decomposition analysis the following variables have been used: 1. Dependent variable: delivery by skilled health providers, which takes a value of 1 if the delivery has been attended by skilled health providers and a value of zero otherwise. 2. Independent variables: • Region; • Place of residence – urban/rural; • Wealth as computed from the asset indices; • Education of mother in years of schooling completed • Head of household – a dummy where female household assumes a value of one; and • Insurance coverage – a dummy with a value of one if the woman has insurance coverage. In NDHS 2006-07, a representative two-stage probability sample of 10,000 households was selected. The first stage consisted of selection of 500 primary sampling units (PSUs) from a sampling frame of 3,750 PSUs with probability proportional to size; the size being the number of households in the 2001 Population Census. The second stage involved the systematic selection of 20 households in each PSU [14]. The demographic and health surveys do not contain data on household income or consumption expenditure. Instead wealth index is used as a proxy. The wealth index is based on household ownership of consumer goods (such as radio, television); dwelling characteristics; type of drinking water source; toilet facilities and other characteristics related to the household's socio-economic status. The asset indices are constructed using the method of principal component analysis (PCA) [14]. Studies have shown a close relationship between asset ownership and consumption expenditure in developing countries [23] and that household asset is a good indicator of the long-run economic status of households [24] Data was analyzed using STATA 10 statistical software and MS Excel.

Materials

N/A

Innovations

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

1. Mobile health clinics: Implementing mobile health clinics that can travel to rural areas and provide skilled health personnel for deliveries. This would help address the disparity in access between urban and rural areas.

2. Telemedicine: Utilizing telemedicine technology to connect pregnant women in remote areas with skilled health providers. This would allow for virtual consultations and remote monitoring, ensuring access to quality care regardless of location.

3. Community health workers: Training and deploying community health workers who can provide basic maternal health services and education in underserved areas. This would help bridge the gap in access to skilled health providers.

4. Financial incentives: Implementing financial incentives for skilled health providers to work in rural or low-income areas. This could help attract and retain skilled professionals in areas where access is limited.

5. Education and awareness campaigns: Launching education and awareness campaigns to inform women about the importance of skilled attendance at birth and the available resources for maternal health. This could help increase demand for skilled health providers and encourage women to seek appropriate care.

6. Public-private partnerships: Establishing partnerships between public and private healthcare providers to improve access to maternal health services. This could involve leveraging private sector resources and expertise to expand services in underserved areas.

7. Strengthening health systems: Investing in the overall strengthening of health systems, including infrastructure, training, and supply chain management. This would help ensure that skilled health providers are available and equipped to provide quality maternal health services.

It’s important to note that these recommendations are based on the information provided and may need to be tailored to the specific context and needs of Namibia.

AI Innovations Description

The recommendation to improve access to maternal health based on the study “Inequities in skilled attendance at birth in Namibia: A decomposition analysis” is to address the wealth and education disparities that contribute to the inequities in accessing skilled health providers during childbirth.

To achieve this, a multi-sectoral approach is needed, in line with the principles of Primary Health Care. This means that efforts to improve access to maternal health services should not be limited to the health system alone, but should also involve other sectors such as education, social welfare, and economic development.

Specific actions that can be taken include:

1. Implementing targeted interventions to improve access to skilled health providers in rural areas, where access is lower compared to urban areas. This can involve increasing the number of skilled health providers in rural areas, providing incentives for health professionals to work in rural areas, and improving transportation infrastructure to facilitate access to health facilities.

2. Addressing the wealth-related inequalities by providing financial support or subsidies for maternal health services to women from economically disadvantaged households. This can help to reduce the financial barriers that prevent women from accessing skilled health providers during childbirth.

3. Promoting education and awareness about the importance of skilled attendance at birth, particularly among women with lower levels of education. This can be done through community-based education programs, antenatal care services, and partnerships with local organizations and community leaders.

4. Strengthening health insurance coverage to ensure that more women have access to affordable maternal health services. This can involve expanding health insurance programs, improving the affordability of premiums, and ensuring that insurance coverage includes maternal health services.

5. Conducting further research and monitoring to assess the impact of interventions and identify any additional barriers or disparities that need to be addressed. This can help to inform evidence-based policies and interventions to improve access to maternal health services.

By implementing these recommendations, it is possible to reduce the inequities in accessing skilled health providers during childbirth and improve maternal health outcomes in Namibia.

AI Innovations Methodology

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

1. Strengthening healthcare infrastructure: Investing in the development and improvement of healthcare facilities, particularly in rural areas, can help increase access to skilled health providers during childbirth.

2. Increasing availability of skilled health providers: Implementing strategies to train and deploy more skilled health providers, such as midwives and nurses, can help ensure that there are enough healthcare professionals available to attend to women during childbirth.

3. Promoting education and awareness: Implementing educational programs and awareness campaigns to educate women and communities about the importance of skilled attendance at birth can help increase demand for maternal health services.

4. Addressing socio-economic inequalities: Implementing policies and interventions that address socio-economic inequalities, such as providing financial support for low-income women to access maternal health services, can help reduce disparities in access to skilled health providers.

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

1. Define indicators: Identify key indicators to measure access to maternal health, such as the percentage of deliveries attended by skilled health providers or the maternal mortality ratio.

2. Collect baseline data: Gather data on the current status of access to maternal health services, including information on skilled attendance at birth, socio-economic status, and other relevant factors.

3. Develop a simulation model: Create a simulation model that incorporates the identified recommendations and their potential impact on access to maternal health. This model should consider factors such as population demographics, healthcare infrastructure, availability of skilled health providers, and socio-economic inequalities.

4. Input data and run simulations: Input the baseline data into the simulation model and run multiple simulations to assess the potential impact of the recommendations. Vary the parameters and assumptions to explore different scenarios and outcomes.

5. Analyze results: Analyze the simulation results to determine the potential impact of the recommendations on improving access to maternal health. Assess the changes in key indicators and identify any trade-offs or unintended consequences.

6. Refine and validate the model: Continuously refine and validate the simulation model based on new data and feedback. Incorporate additional factors or recommendations as needed to improve the accuracy and reliability of the simulations.

7. Communicate findings and make recommendations: Present the findings of the simulation analysis to relevant stakeholders, policymakers, and healthcare providers. Use the results to inform decision-making and guide the implementation of interventions to improve access to maternal health.

It is important to note that the methodology outlined above is a general framework and may need to be adapted based on the specific context and data availability.

Author

Dima Health

Statistics:

Citations: 73

Identifiers:

DOI: 10.1186/1471-2393-11-34

Research Areas:

Community Interventions, Disparities, Environmental, Health System and Policy, Maternal Access, Maternal and Child Health, Quality of Care, Social Determinants, Workforce

Study Countries:

Namibia

Study Design:

Cross Sectional Study, randomised-control-trial

Study Approach:

Quantitative

Participants Gender:

Male & Female