Using verbal and social autopsies to explore health-seeking behaviour among HIV-positive women in Kenya: A retrospective study

BMC Women's Health, Volume 14, No. 1, Year 2014

Ukuhunyushwa

Background: There is limited understanding of the factors that influence decisions to seek HIV care and treatment services in community settings. The aim of this study was to explore the socio-cultural and health system factors affecting health-seeking behaviour among deceased women in Kenya who were living with HIV at the time of death.Methods: Out of a total of 796 deaths for which a caregiver was available to provide information, retrospective data were drawn from verbal and social autopsies administered to caregivers of 218 women who had died of AIDS-related illnesses aged 15 to 49 years. Information was collected on essential elements of the care-seeking process from the onset of severe illness episodes and analysed using qualitative and quantitative techniques.Results: Results from the quantitative data showed that poor women were less likely to access formal health services (OR = 0.2; p < 0.001) compared to non-poor women. The qualitative data showed that socioeconomic status, poor knowledge and understanding of AIDS-related illness, distance to facility and transportation costs, medical pluralism, stigma, low HIV risk perception, lack of family support and health care system barriers contributed to delays/constraints in seeking care.Conclusions: The findings highlight important issues that have implications for addressing challenges faced by women living with HIV, including non-adherence to treatment regimen and late diagnosis of HIV. Provision of transportation subsidies as part of the national social safety-net strategy can help in addressing financial constraints associated with transportation costs among poor women living with HIV. © 2014 Njuki et al.; licensee BioMed Central Ltd. The study that provided the data for this paper was part of a project implemented by the Population Council which aimed at evaluating the effect of the output-based aid (OBA) voucher program (being implemented in six districts in the country) on reproductive health behaviours and outcomes [25]. The objective of the voucher program is to significantly reduce maternal and neonatal mortality by increasing the number of health facility deliveries and improving access to appropriate health services—including reproductive health—for the poor through incentives for increased demand and improved service provision. Death audits were conducted to determine the prevalence of maternal and newborn mortality and understand health-seeking behaviour prior to the occurrence of deaths in the OBA study sites. Data were collected between June and December 2010 and involved a cross-sectional survey of deaths of women aged 15–49 years living within a 5 km radius of accredited OBA facilities in three districts (Kisumu, Kiambu and Kitui). The primary sampling unit for the verbal autopsy was the sub-location. In each district, 14 sub-locations were randomly selected from among those within the stipulated distance (within 5 km radius) to accredited facilities. Three enumeration areas/villages were then randomly selected from each of the sampled sub-locations. In the selected enumeration areas, all deaths that occurred among women aged 15–49 years were identified through the death notification register that is maintained by the local government administration. In Kenya, the civil registration process includes a chief (who heads a location) who is tasked with registering all deaths occurring in that particular location. The chiefs are supported by assistant chiefs and village elders to ensure all deaths are registered and a death notification card is issued as a requirement for acquiring a burial permit. To collect the retrospective data, we used verbal and social autopsies. The World Health Organization (WHO) standard verbal autopsy questionnaires and a social autopsy tool were administered to caregivers of women aged 15–49 years who had died between 1996 and 2010. Only one caregiver was interviewed for each death. Verbal autopsy is a research method that helps determine probable causes of death in cases where there was no medical record or formal medical attention given and involves conducting interviews with next of kin or other caregivers [26]. Social autopsy, on the other hand, refers to an interview process with a next of kin or other caregivers and aims at identifying social, behavioural, and health systems contributors to deaths in the community [27]. Both verbal and social autopsies provide evidence for informing health care programmers and policymakers in designing and implementing initiatives for improving maternal and child health [27]. A total of 819 deaths were identified. However, because some caregivers could not be traced, our analyses focused on 796 deaths where verbal and social autopsy interviews were conducted with an immediate caregiver of the deceased. Interviews were conducted in Kiswahili or a local language. The caregiver was selected by the family members on the basis of her/his participation in the health and care seeking process for the deceased. All the 796 deaths were coded by two trained medical doctors to determine the causes of death using the International Classification of Disease (ICD)-10. Where the doctors did not agree on the causes of death, they met to arrive at a consensus, and if they still did not agree after the meeting, the cause of death was recorded as unknown. Of the 796 deaths, 218 were coded as HIV/AIDS-related deaths. It is worth noting that a total of 64 deaths met the World Health Organization (WHO) definition of maternal death due to direct or indirect causes. Out of the 64 maternal deaths, 11 (17%) were caused by AIDS-related complications. Overall, AIDS-related complications accounted for nearly a third of the maternal deaths. Information was collected on a wide range of issues, including socio-demographic characteristics, essential elements of the care-seeking process such as recognition of the symptoms and signs associated with illness, whether adequate care was provided, whether and what type of outside-the-home care was sought (informal, formal, or both), cause and place of death. Social autopsy interviews explored health and care-seeking behaviour of the deceased including any preventive care received, the diagnostic procedures followed, the type and timing of any treatment provided within or outside the home, any barriers/delays encountered during care seeking, and the quality of health care provided (from the client’s perspective). Data analysis was conducted separately for the verbal and social autopsy. For quantitative data derived from the verbal autopsy, descriptive statistics were used to characterize the study sample, including primary caregiver’s relationship, place where care was sought, number of contacts with formal health services, place of death, household socioeconomic status, marital status, education and age. Quantitative data were analysed using STATA® version 10. Cross-tabulations with Chi-square and Fisher’s exact tests were done to test the associations between key variables of interest. Multivariable logistic regression analysis was conducted to examine the relationship between having contact with formal health services and socioeconomic status after controlling for confounding factors. Regression analysis was also performed to examine the relationship between the use of more than one source of medical care (medical pluralism) and socioeconomic status. Socioeconomic status was assessed using a poverty grading tool that consisted of eight items on household assets and amenities, expenditure or income, and access to health services customized to each district to identify OBA voucher clients from the community. For each item, the score ranged from 1 to 3 with the maximum score for all the items being 24. Women scoring between eight (which is the minimum) and 16 points on the poverty grading tool were deemed poor. Qualitative interviews were tape recorded, transcribed verbatim and translated into English. There was, however, no back-translation due to financial constraints. We adopted the grounded theory methods in data analysis which involved both inductive and deductive approaches. Analysis was done by two researchers to ensure reliability in the coding and results. Following coding, a full list of themes was available for categorization within a hierarchical framework of main and sub-themes. The thematic framework was then systematically applied to all of the interview transcripts. We looked for patterns and associations of the themes and compared and contrasted within and between the different regions and age groups. The qualitative transcripts were managed and analysed using QSR NVivo 9 Software © (International Pty 2007, Australia). Analysis of the data was guided by the pathway to survival model [28,29] and the three-delay framework [30]. Written informed consent for participation in the study was obtained from informants. Ethical and research clearance was obtained from the Institutional Review Board of the Population Council, the Ethics Review Committee of the Kenya Medical Research Institute (KEMRI), the National Council for Science and Technology (NCST), and the Ministries of Health (Public Health and Medical Services).

I-AI Digest

Study Justification:

This study aimed to explore the factors that influence health-seeking behavior among HIV-positive women in Kenya who have died. The study is important because there is limited understanding of these factors, which can have implications for non-adherence to treatment and late diagnosis of HIV. By identifying the barriers and constraints faced by these women, the study can inform interventions and policies to improve their access to care and support.

Highlights:

– The study found that poor women were less likely to access formal health services compared to non-poor women.
– Socioeconomic status, poor knowledge of AIDS-related illness, distance to facility and transportation costs, medical pluralism, stigma, low HIV risk perception, lack of family support, and health care system barriers were identified as factors contributing to delays in seeking care.
– The study suggests that providing transportation subsidies as part of the national social safety-net strategy can help address financial constraints associated with transportation costs among poor women living with HIV.

Recommendations:

– Implement transportation subsidies as part of the national social safety-net strategy to address financial constraints associated with transportation costs among poor women living with HIV.
– Improve knowledge and understanding of AIDS-related illness among the community to reduce delays in seeking care.
– Address stigma and low HIV risk perception through awareness campaigns and community engagement.
– Strengthen the health care system to remove barriers and improve access to care for HIV-positive women.

Key Role Players:

– Ministry of Health: Responsible for implementing and coordinating interventions and policies related to HIV care and treatment.
– Non-governmental organizations (NGOs): Involved in providing support services and advocacy for HIV-positive women.
– Community leaders and influencers: Play a crucial role in raising awareness, reducing stigma, and promoting health-seeking behavior.
– Health care providers: Responsible for delivering quality care and support to HIV-positive women.

Cost Items:

– Transportation subsidies: Budget allocation for providing financial support to cover transportation costs for poor women living with HIV.
– Awareness campaigns: Funding for community engagement activities, including workshops, seminars, and media campaigns to improve knowledge and understanding of AIDS-related illness.
– Health system strengthening: Investment in infrastructure, training, and capacity building to remove barriers and improve access to care for HIV-positive women.
– Support services: Funding for NGOs to provide counseling, psychosocial support, and other services to HIV-positive women.

Amandla Obufakazi

The strength of evidence for this abstract is 8 out of 10.
The evidence in the abstract is strong because it includes both qualitative and quantitative data. The study sample size is large, and the data collection methods are well-described. However, to improve the evidence, the abstract could provide more details on the specific findings and their implications for addressing challenges faced by women living with HIV. Additionally, it would be helpful to include information on the limitations of the study and any potential biases in the data collection process.

Abstract

The study that provided the data for this paper was part of a project implemented by the Population Council which aimed at evaluating the effect of the output-based aid (OBA) voucher program (being implemented in six districts in the country) on reproductive health behaviours and outcomes [25]. The objective of the voucher program is to significantly reduce maternal and neonatal mortality by increasing the number of health facility deliveries and improving access to appropriate health services—including reproductive health—for the poor through incentives for increased demand and improved service provision. Death audits were conducted to determine the prevalence of maternal and newborn mortality and understand health-seeking behaviour prior to the occurrence of deaths in the OBA study sites. Data were collected between June and December 2010 and involved a cross-sectional survey of deaths of women aged 15–49 years living within a 5 km radius of accredited OBA facilities in three districts (Kisumu, Kiambu and Kitui). The primary sampling unit for the verbal autopsy was the sub-location. In each district, 14 sub-locations were randomly selected from among those within the stipulated distance (within 5 km radius) to accredited facilities. Three enumeration areas/villages were then randomly selected from each of the sampled sub-locations. In the selected enumeration areas, all deaths that occurred among women aged 15–49 years were identified through the death notification register that is maintained by the local government administration. In Kenya, the civil registration process includes a chief (who heads a location) who is tasked with registering all deaths occurring in that particular location. The chiefs are supported by assistant chiefs and village elders to ensure all deaths are registered and a death notification card is issued as a requirement for acquiring a burial permit. To collect the retrospective data, we used verbal and social autopsies. The World Health Organization (WHO) standard verbal autopsy questionnaires and a social autopsy tool were administered to caregivers of women aged 15–49 years who had died between 1996 and 2010. Only one caregiver was interviewed for each death. Verbal autopsy is a research method that helps determine probable causes of death in cases where there was no medical record or formal medical attention given and involves conducting interviews with next of kin or other caregivers [26]. Social autopsy, on the other hand, refers to an interview process with a next of kin or other caregivers and aims at identifying social, behavioural, and health systems contributors to deaths in the community [27]. Both verbal and social autopsies provide evidence for informing health care programmers and policymakers in designing and implementing initiatives for improving maternal and child health [27]. A total of 819 deaths were identified. However, because some caregivers could not be traced, our analyses focused on 796 deaths where verbal and social autopsy interviews were conducted with an immediate caregiver of the deceased. Interviews were conducted in Kiswahili or a local language. The caregiver was selected by the family members on the basis of her/his participation in the health and care seeking process for the deceased. All the 796 deaths were coded by two trained medical doctors to determine the causes of death using the International Classification of Disease (ICD)-10. Where the doctors did not agree on the causes of death, they met to arrive at a consensus, and if they still did not agree after the meeting, the cause of death was recorded as unknown. Of the 796 deaths, 218 were coded as HIV/AIDS-related deaths. It is worth noting that a total of 64 deaths met the World Health Organization (WHO) definition of maternal death due to direct or indirect causes. Out of the 64 maternal deaths, 11 (17%) were caused by AIDS-related complications. Overall, AIDS-related complications accounted for nearly a third of the maternal deaths. Information was collected on a wide range of issues, including socio-demographic characteristics, essential elements of the care-seeking process such as recognition of the symptoms and signs associated with illness, whether adequate care was provided, whether and what type of outside-the-home care was sought (informal, formal, or both), cause and place of death. Social autopsy interviews explored health and care-seeking behaviour of the deceased including any preventive care received, the diagnostic procedures followed, the type and timing of any treatment provided within or outside the home, any barriers/delays encountered during care seeking, and the quality of health care provided (from the client’s perspective). Data analysis was conducted separately for the verbal and social autopsy. For quantitative data derived from the verbal autopsy, descriptive statistics were used to characterize the study sample, including primary caregiver’s relationship, place where care was sought, number of contacts with formal health services, place of death, household socioeconomic status, marital status, education and age. Quantitative data were analysed using STATA® version 10. Cross-tabulations with Chi-square and Fisher’s exact tests were done to test the associations between key variables of interest. Multivariable logistic regression analysis was conducted to examine the relationship between having contact with formal health services and socioeconomic status after controlling for confounding factors. Regression analysis was also performed to examine the relationship between the use of more than one source of medical care (medical pluralism) and socioeconomic status. Socioeconomic status was assessed using a poverty grading tool that consisted of eight items on household assets and amenities, expenditure or income, and access to health services customized to each district to identify OBA voucher clients from the community. For each item, the score ranged from 1 to 3 with the maximum score for all the items being 24. Women scoring between eight (which is the minimum) and 16 points on the poverty grading tool were deemed poor. Qualitative interviews were tape recorded, transcribed verbatim and translated into English. There was, however, no back-translation due to financial constraints. We adopted the grounded theory methods in data analysis which involved both inductive and deductive approaches. Analysis was done by two researchers to ensure reliability in the coding and results. Following coding, a full list of themes was available for categorization within a hierarchical framework of main and sub-themes. The thematic framework was then systematically applied to all of the interview transcripts. We looked for patterns and associations of the themes and compared and contrasted within and between the different regions and age groups. The qualitative transcripts were managed and analysed using QSR NVivo 9 Software © (International Pty 2007, Australia). Analysis of the data was guided by the pathway to survival model [28,29] and the three-delay framework [30]. Written informed consent for participation in the study was obtained from informants. Ethical and research clearance was obtained from the Institutional Review Board of the Population Council, the Ethics Review Committee of the Kenya Medical Research Institute (KEMRI), the National Council for Science and Technology (NCST), and the Ministries of Health (Public Health and Medical Services).

Izinto zokwakha

N/A

Emisha

The recommendation to improve access to maternal health services, particularly among poor women living with HIV, is to provide transportation subsidies as part of the national social safety-net strategy. This recommendation is based on the finding that poor women were less likely to access formal health services compared to non-poor women, and that transportation costs and distance to facilities were identified as barriers to seeking care. By providing transportation subsidies, financial constraints associated with transportation costs can be addressed, thereby improving access to maternal health services and addressing challenges such as non-adherence to treatment regimens and late diagnosis of HIV.

AI Innovations Description

Based on the information provided, the recommendation to improve access to maternal health is to provide transportation subsidies as part of the national social safety-net strategy. This recommendation is based on the finding that poor women were less likely to access formal health services compared to non-poor women. Transportation costs and distance to facilities were identified as barriers to seeking care. By providing transportation subsidies, financial constraints associated with transportation costs can be addressed, particularly among poor women living with HIV. This can help improve access to maternal health services and address challenges such as non-adherence to treatment regimens and late diagnosis of HIV.

AI Innovations Methodology

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

1. Define the study population: Identify the target population for the transportation subsidies, such as poor women living with HIV who are in need of maternal health services.

2. Select study sites: Choose specific locations or districts where the transportation subsidies will be implemented. Consider factors such as geographic diversity, availability of health facilities, and population density.

3. Randomize study participants: Randomly assign eligible women to either the intervention group (receiving transportation subsidies) or the control group (not receiving transportation subsidies). Ensure that the groups are comparable in terms of demographic characteristics, socioeconomic status, and health status.

4. Implement the intervention: Provide transportation subsidies to the intervention group, covering the costs associated with accessing maternal health services. This could include subsidies for public transportation, private vehicles, or other means of transportation.

5. Collect data: Use a combination of quantitative and qualitative methods to collect data on various indicators related to access to maternal health services. This could include information on the number of antenatal care visits, facility-based deliveries, postnatal care utilization, and overall satisfaction with the transportation subsidies.

6. Analyze the data: Compare the outcomes between the intervention and control groups to assess the impact of the transportation subsidies. Use statistical methods, such as chi-square tests or logistic regression, to determine if there are significant differences in access to maternal health services between the two groups.

7. Interpret the findings: Analyze the results to understand the effectiveness of the transportation subsidies in improving access to maternal health services. Consider factors such as the utilization rate of the subsidies, barriers that may still exist despite the subsidies, and any unintended consequences.

8. Make recommendations: Based on the findings, provide recommendations for policy and programmatic changes. This could include scaling up the transportation subsidies to a larger population, targeting specific subgroups within the population, or addressing other barriers to access identified during the study.

9. Monitor and evaluate: Continuously monitor and evaluate the implementation of the transportation subsidies to ensure their effectiveness and make any necessary adjustments. This could involve regular data collection, stakeholder consultations, and feedback mechanisms.

By following this methodology, researchers can assess the impact of transportation subsidies on improving access to maternal health services and provide evidence-based recommendations for policy and programmatic interventions.

Umbhali

Dima Health

Statistics:

Citations: 11

Identifiers:

DOI: 10.1186/1472-6874-14-77

Research Areas:

Community Interventions, Health System and Policy, Infectious Diseases, Maternal Access, Maternal and Child Health, Quality of Care, Sexual and Reproductive Health, Social Determinants, Technology and Innovations, Workforce

Study Countries:

Kenya

Study Design:

Cohort Study, Cross Sectional Study, Exploratory Study, Grounded Theory

Study Approach:

Mixed-methods, Qualitative, Quantitative

Participants Gender:

Female

Yabelana ngalokhu: