Background: Substantial improvements have been observed in coverage and access to maternal health services in Ethiopia. However, the quality of care has been lagging behind. Therefore, this study aimed to assess the level of quality of Option B+ PMTCT in Northern Ethiopia. Methods: A facility based survey was conducted from February to April 2016 in Northern Ethiopia. Twelve health facilities were enrolled in the study. Mixed method approach was used in line with Donabedian (Input- Process-Output) service quality assessment model. Data of 168 HIV positive mothers & their infant were abstracted from registers, and follow up charts. During the Option B+ service consultation, a total of 60 sessions were involved for direct observation. Of which, 30 clients and 12 service providers were subjected for exit and in-depth interview respectively. Facilities were categorized rendering good service quality based on predetermined quality judgment criteria. Reasons of good and bad service quality were thematically fitted with each quality component based on emerging themes (TM1-TM3), and categories (CA1-CA6). Results: Of the total 12 study health facilities, 2(16.7%) were achieved the desired level of service quality based on the three quality components. The input quality was better and judged as good in 33.3% health facilities. However; process and output service quality were realized in one – fourth of them. Conclusion: Insignificant numbers of facilities fulfilled the aspired level of service quality. Quality of care was found influenced by multiple inputs, processes, and output related barriers and facilitators. Comprehensive Program monitoring is needed based on three quality components to improve the overall service quality.
The study aimed to assess the level of quality of OptionB+PMTCT and to explore reasons for good and bad service quality. This study was a facility based cross-sectional survey conducted between February to April 2016. Mixed method approach was used involving both quantitative and qualitative data collection methods. Donabedian model was used [19] as depicted in the figure below (Fig. (Fig.11). The study was conducted in Mekelle zone, Tigray of Northern Ethiopia, 802KMs from Addis Ababa, the capital city of Ethiopia. It is among the top three high HIV prevalence areas in Tigray region [22]. According to the projected national census in 2018 [23], the total population was projected to 320,000. A total of 12 health facilities have been provided Option B+ under MNCH continuum of care. Of which three of them were in hospitals. All health facilities (nine health centers and three hospitals) providing Option B+ under MNCH platforms were involved in the study. Study participants were HIV positive mother & their infants. They were newly diagnosed women and enrolled under Option B+ package aiming “test and treat” strategy. Besides, who were under continuous follow up for more than one year prior to data collection in MNCH clinic for the purpose of complete clinical history. To come up with the final sample, review of all patient records and follow up charts were conducted which fulfilled predetermined eligibility criteria. First, a total of 219 HIV positive mother & infant pairs were reviewed. However, 48 study participants were excluded since they were transferred from other facilities and did not have a complete clinical history. Similarly, three newly diagnosed HIV positive women spent only six months in HIV chronic care were not included in the study. Hence, the total participants enrolled for the study were 168. Convenience sampling method was used to recruit participants for qualitative study, until information saturation was obtained [24]. Observation and interviews were conducted during & after service consultation. A total of 60 sessions were involved for direct observation. Of which, 30 HIV positive women and 12 service providers were subjected for exit and in-depth interview respectively to identify reasons for good and bad service quality in respective quality components. Health workers have been providing Option B + PMTCT in MNCH clinic for more than one year were included in the study. Quantitative data were collected by four midwives and master’s level supervisor who had an experience in public health data collection. For input quality, 47 indicators were adopted from the national guidelines [20, 21]. Facility inventory was conducted to ensure the availability of essential equipments, drugs and supplies. See list of input quality verification variables (Additional file 1). A total of 43 process related indicators were articulated for assessing process quality [16, 17, 25]. Non-participant observation was conducted using a checklist to observe service adheres to standard practices. The intention was to observe 10 sessions from each facility during each round of data collection. However, this was not possible since too few women came for Option B+PMTCT services during the data collection period; indeed, 60 clients in which five for each facility were participated. In addition, some process related variables were obtained from recorded review. See list of process quality verification variables (Additional file 1). Output quality was assessed using 13 items adopted from national guideline [20, 21]. Data were abstracted retrospectively from logbooks and/or records and individual medical records for 168 mother infant pairs using a data abstraction tool. Besides, 12 HIV exposed infant follow up charts were reviewed about the utilization of the Option B+ service package in MNCH clinic. Confidentiality around patient records was protected; communication was made with only authorized person. Patient codes were used and documents were locked. See list of output quality verification variables (Additional file 1). Overall service quality was assessed by combining input, process, and output service quality items. Facilities were categorized rendering good input service quality, if the average weighted score of input quality performance standards is 100% [10], and 90% or more for process, output, and overall quality performance standards [10, 20]. See the score of each variable for respective quality components (Additional file 1). Qualitative data collection was guided by principal investigator (corresponding author) who had an experience on qualitative data collection. Both interviews were conducted in Tigrigna (local language), and audio recording was involved. Exit interviews were conducted with mothers who were provided oral consents for participation at the end of service consultation. After 30 interviews were involved with clients, information saturation was obtained; no further interviews were conducted since no new information was being generated. Interviews were guided by using a pre-tested flexible interview guide with probes to identify client views on service utilization. Interviews were lasted within the range of 30–40 min. In-depth interviews were held by appointment with service providers during the study. Twelve key informant interviews were conducted until information saturation was obtained. The interviews were guided by using a pre-tested, flexible interview guide with probes to identify their role in providing PMTCT services and reasons for good and bad service quality. Interviews were lasted within 15–25 min. To enhance data quality, training was provided for data collectors and supervisor with the objective of the study, the nature of the data collection tools, and ways of approaching during interview, observe, record and chart review. During the data collection period there was a strict supervision scheme. Completed questionnaires were checked on a daily basis by supervisor and principal investigator. Quantitative data were coded, cleaned, and entered into EPI info version 7 and then exported and analyzed using SPSS version 21 software. Univariate data analysis was conducted to estimate the prevalence of variables for respective quality components. The Findings were presented in tables (Tables (Tables1,1, ,2,2, ,33 and and4)4) and figure (Fig. 2). Health facilities not fulfilling 100% of input service quality performance verification indicators in Mekelle zone, Tigray, Northern Ethiopia (n = 12) Health facilities not fulfilling 90% of process service quality performance verification indicators in Mekelle zone, Tigray, Northern Ethiopia (n = 12) Health facilities not fulfilling 90% of output service quality performance verification indicators in Mekelle zone, Tigray, Northern Ethiopia (n = 12) Summery of themes (TMs) and categories (CAs) developed based on KIIs data in respective quality components in Mekelle zone, Tigray, Northern Ethiopia Summary of Option B + PMTCT service quality in studied health facilities in Mekelle Zone, Northern Ethiopia. [Note: (0/3): number of facilities not achieved any of the three quality components; (1/3): number of facilities achieved any one of the three quality components; (2/3): number of facilities achieved two of the three quality components; (3/3): number of facilities achieved all three quality components] Qualitative data were analyzed manually using the content thematic approach [24]. Data analysis was done by the first author in collaboration with a second author (experienced qualitative researcher). This involved reading script several times, translating transcripts from local language (Tigrigna) to English, identifying themes (TM1-TM3 …) and categories (CA1-CA6). The main study themes were fitted with three quality components, whilst categories were motivators and barriers for good and bad service quality in each identified themes. All authors were involved in discussions of study themes, & categories. This process facilitated researcher triangulation. Direct quotations were presented reflecting reasons for good and bad service quality. Input dimension: this dimension was used to assess the availability of human resources, materials, drugs, equipment, and supplies needed for Option B+ PMTCT service provision [17]. Process dimension: this dimension used to reflect how service providers adhere to service standards during a service consultation of Option B+PMTCT service in MNCH unit [18]. Output dimension: used to evaluate the ultimate service result of Option B+PMTCT service and patient satisfaction level [17]. Overall quality: this particular dimension was determined by combining predetermined three quality components; input, process, and output [25]. DBS result turnaround time: The date for DBS sample collection from HIV-exposed infants to the date when HIV screening results was arrived at health facility [26].