Quality of Care in Family Planning Services in Ethiopia

Background Annually approximately 303,000 maternal deaths occurred globally in 2015, of which 99% occurred in Low-and Middle-income countries (LMIC). Estimates suggest that Ethiopia is one of the ten countries that account for about 60% of these maternal deaths. Studies demonstrate that use of Family Planning (FP) can contribute to reducing maternal deaths. While a proportion of women are using FP services, more than 214 million reproductive aged women in LMIC who desired to avoid or postpone pregnancy are not using any modern contraceptive methods. This trend in the use of FP services is also found in Ethiopia. Along with socio-demographic and cultural factors, the quality of care in FP services has been shown to influence the use of FP services. However, few studies have been conducted on the quality of care in FP services in Africa, and little evidence has been reported on the factors determining quality of care in FP services. While a few small studies that have assessed quality of care in FP services in Ethiopia, no study has identified systematically the factors associated with quality of care or the influence of facility type on quality of care. Aim The overall aim of the study is to identify the determinants and differences in quality of care in FP services. The overall study consisted of four interrelated studies each with their own inter-related but specific aim. These were: to assess and explore factors associated with quality of care in FP services in Africa (Study 1); to identify the client and facility-level determinants of quality of care in FP services in Ethiopia (Study 2); to compare the quality of FP services in private and public primary health care facilities and users’ characteristics of these facilities (Study 3), and since the use of FP guidelines was a factor in quality of care, the last study’s aim was to understand provider perspectives on barriers to and facilitators for FP guidelines use in FP services (Study 4). Methods This study employed a mix of study designs using three different methods and analytic approaches. Firstly, informed by a published systematic review protocol, a systematic review was conducted to evaluate the best available evidence on factors affecting the quality of care in FP services in Africa (Study 1). The systematic review included studies published between 1990 and 2016 in English. Quantitative studies reporting on factors affecting quality of care and qualitative studies exploring client and provider experiences and/or perceptions of the factors that determine quality of care in FP services were considered for inclusion in the review. For quantitative studies, client satisfaction was used as a measure of quality of care and was assessed in three ways: First, it was assessed using proxy questions reflecting client satisfaction on a range of issues such as waiting time, privacy for not being seen or heard by others, cleanliness of the facility, and costs of the services. Then clients’ responses for these questions were aggregated into a single variable using principal component analyses to create a measure either as a continuous or binary outcome in terms of satisfied or not satisfied; Secondly, using a likert scale in ten categories with the higher scale indicating greater satisfaction and then creating a binary variable using the mean as a cutoff point (i.e those who scored below the mean regarded as less satisfied while those who scored equal to or above the mean regarded as highly satisfied). Thirdly, using client’s overall satisfaction and then creating a binary outcome comprising satisfied or not satisfied. I performed separate synthesis of the two evidence types: the quantitative data were summarised in narrative and tables; the qualitative findings were synthesised using meta-aggregation. Secondly, informed by the findings of the systematic review, I conducted a quantitative analysis of secondary data (Study 2 and 3). More specifically, for Study 2, a multilevel mixed-effects modelling was conducted using the Ethiopian Services Provision Assessment (ESPA+) 2014 data to identify the client and facility-level determinants of quality of care in FP services as measured by client satisfaction in Ethiopia. In the analysis, while client and facility-level variables were considered as independent variables, client satisfaction was considered as an outcome variable. Client satisfaction was measured using clients’ exit interview responses to questions about the health service quality, assessed by the problems encountered by clients during their visit to health facilities for FP services. Clients’ responses on eleven questions that reflect clients’ perceptions of the quality of FP services were aggregated into an index using principal component analysis. Then, the aggregated index was dichotomised using the median score as cutoff point. Finally, a binary outcome of client satisfaction was devised as “more satisfied” if a score was greater or equal to the median cutoff point, and as “less satisfied” if a score was less than the median cutoff point. For Study 3, a combination of facility-based data from the ESPA+ 2014 dataset and community-based data from the Ethiopian Demographic and Health Survey (EDHS) 2016 dataset were employed. In the analysis, structural variables that reflected the material structure such as facility’s infrastructure (basic amenities) and availability of equipment and supplies and human resources such as health provider availability and trained provider availability were compared in public versus private primary health care facilities. The structural variables that reflected organisational structure such as presence of a quality assurance system and supervision in the past six months, availability of FP guidelines/protocols, and availability of a range of modern contraceptive methods were also compared in public versus private primary health care facilities. Survey logistic regression analysis was conducted to compare the structural quality of services, and a chi-square test was used to compare the characteristics of clients’/users’ who accessed FP services from these facilities. In Study 4, qualitative method was used to understand health providers’ perspectives on the use of FP guidelines in FP services. This study used in-depth interviews guided by a semi-structured interview guide. Twenty one participants were recruited from nine health facilities including two hospitals, five health centres, and two health posts in Gondar and Bahir Dar City administrations, Amhara region, Northwest Ethiopia. The audio-recorded interviews and notes taken were translated and transcribed into English by the lead author and entered into NVivo 11TM for data analysis and management. Thematic analysis according to the approach described by Braun and Clarke was employed for data analysis. Results In Study 1, the systematic review found few studies (eight quantitative studies and three qualitative studies) had been undertaken in a small proportion of African countries. While the quantitative studies were undertaken in Egypt, Kenya, Senegal, Ethiopia, Ghana, Tanzania, Namibia, the qualitative studies were undertaken in Kenya and Uganda. The studies were performed between 1995 and 2016. The limited evidence, assessed as being moderate to high quality suggested that quality of care in FP services was influenced by a range of client, provider and facility factors, as well as structural and process aspects of the facilities. Amongst the process factors, shorter client waiting time, presence of competent healthcare providers, provision/prescription of injectable methods, maintenance of privacy, and confidentiality were the most commonly identified factors positively associated with quality of care in FP services. For factors related to structure, good quality of stock was the most commonly identified factor positively associated with quality of care in FP services. In terms of the facility related factors, quality of care was associated with facility ownership in that privately-owned facilities and availability of FP guidelines were positively associated with better quality of care. The qualitative component of the systematic review pointed to additional factors associated with quality of care in FP services including access related factors such as ‘pre-requisites to be fulfilled by the clients and cost of services, provider workload, and providers’ behaviour. In Study 2, while both client and facility-level factors were shown to be associated with quality of care in FP services in Ethiopia, nearly one-third (32.8%) of the differences in the quality of care were attributed to the health facility level factors. At the client-level; provision of information on potential side effects of contraceptive method and number of history and physical assessments performed were positively associated with client satisfaction, long client waiting times (between 30 minutes to two hours) was negatively associated with client satisfaction. At the facility-level; facilities being in an urban location, and having FP guidelines/protocols for their providers were positively associated with client satisfaction. In Study 3, private health facilities appear relatively more deficient in terms of some important aspects of structural aspects of quality of services such as availability of trained staff, access to FP guidelines/protocols and access to a range of contraceptives than public health facilities. Private health facilities are better equipped with basic infrastructure component of the structural quality of services such as functional cell phones and water supply and equipment than public health facilities. Women who accessed FP services from private facilities were different from those who accessed these services in public facilities. They were more likely to reside in an urban area, to be Muslim, have a job, and have no or fewer number of children than women accessing FP services from public health facilities. In Study 4, healthcare providers identified a number of barriers affecting use of FP guideline. These included: lack of knowledge and lack of or inadequate access to guideline; lack of up-to-date information in the guideline, providers’ behaviour including limited reading of literature including clinical guidelines and religious beliefs/values against FP services provision; lack of support and supervision from managers; insufficient health workforce; and lack of or inadequate training about FP guideline. Healthcare providers also identified a few facilitators to FP guideline use including ease of access; managers who championed their use; and provision of training about FP guideline. Conclusion Overall, the findings showed that quality of care in FP services was influenced by multiple factors related to FP services clients, healthcare providers, and the health facility characteristics. In Ethiopia, the factors affecting quality of care in FP services were related to structure and process of care provision. The findings have also indicated that the structural quality of services in FP services were different between public and private health facilities. Moreover, the findings demonstrated the characteristics of women accessing FP services in private facilities were different from the characteristics of women accessing FP services in public facilities. The findings have also pointed to a set of factors affecting use of FP guidelines including lack of knowledge and lack of or insufficient access to the guidelines, providers’ personal religious beliefs, relying on prior knowledge and tradition rather than protocols and guidelines, insufficient health workforce, and lack of support from managers, and inadequate training on use of guidelines. These results provide important evidence for policy makers and stakeholders to develop effective strategies to help to further improve the quality of care in FP services in Ethiopia and thereby improve the uptake of FP services in that country. Moreover, the results showed that actions are needed at different levels targeting health systems and health facilities. Further studies are also required to explore the healthcare providers’ and managers’ views of factors affecting quality of care in FP services. Strengths of the study The thesis has a number of strengths. The thesis employed multiple but interrelated study deigns including systematic reviews, secondary data analysis, and qualitative interviews. The thesis examines determinants of quality of care from broader (Africa-level) to specific geographic locations (Amhara region). The thesis uses nationally representative datasets obtained from surveys conducted using standardised methodologies and data collection instruments. The qualitative study explored the barriers and facilitators of FP guidelines use for the first time in Ethiopia. Unique contributions of the thesis The findings of thesis has several contributions for policy and further research. In terms of policy implications, the thesis suggests that improving quality of care in FP services in Ethiopia and other LMIC requires improving structural and process components of quality of care. Moreover, it was found that structural components of quality of care could influence not only the outcome of quality of care but also the process of care provision in FP services. The thesis also suggests that, with its limitations, the Donabedian model of quality of care can serve as a lens through which quality of care in FP services could be measured. In terms of further research, the thesis indicates that: 1) exploring factors affecting quality of care from health providers and health managers’ viewpoint as it is useful to identify additional factors related to the healthcare system, 2) further studies are needed to explore how structural and process components of quality of care affecting long term outcomes of quality of care such as reduction in fertility and maternal mortality.Thesis (Ph.D.) -- University of Adelaide, School of Public Health, 201