Missed opportunities in antenatal care for improving the health of pregnant women and newborns in Geita district, Northwest Tanzania

Background: Despite the significant benefits of early detection and management of pregnancy related complications during antenatal care (ANC) visits, not all pregnant women in Tanzania initiate ANC in a timely manner. The primary objectives of this research study in rural communities of Geita district, Northwest Tanzania were: 1) to conduct a population-based study that examined the utilization and availability of ANC services; and 2) to explore the challenges faced by women who visited ANC clinics and barriers to utilization of ANC among pregnant women. Methods: A sequential explanatory mixed method design was utilized. Household surveys that examined antenatal service utilization and availability were conducted in 11 randomly selected wards in Geita district. One thousand, seven hundred and nineteen pregnant women in their 3rd trimester participated in household surveys. It was followed by focus group discussions with community health workers and pregnant women that examined challenges and barriers to ANC. Results: Of the pregnant women who participated, 86.74% attended an ANC clinic at least once; 3.62% initiated ANC in the first trimester; 13.26% had not initiated ANC when they were interviewed in their 3rd trimester. Of the women who had attended ANC at least once, the majority (82.96%) had been checked for HIV status, less than a half (48.36%) were checked for hemoglobin level, and only a minority had been screened for syphilis (6.51%). Among women offered laboratory testing, the prevalence of HIV was 3.88%, syphilis, 18.57%, and anemia, 54.09%. In terms of other preventive measures, 91.01% received a tetanus toxoid vaccination, 76.32%, antimalarial drugs, 65.13%, antihelminthic drugs, and 76.12%, iron supplements at least once. Significant challenges identified by women who visited ANC clinics included lack of male partner involvement, informal regulations imposed by health care providers, perceived poor quality of care, and health care system related factors. Socio-cultural beliefs, fear of HIV testing, poverty and distance from health clinics were reported as barriers to early ANC utilization. Conclusion: Access to effective ANC remains a challenge among women in Geita district. Notably, most women initiated ANC late and early initiation did not guarantee care that could contribute to better pregnancy outcomes. The study was conducted at Geita district, one of the six districts in Geita region Northwest Tanzania. The district has 35 wards with a district hospital, 5 health centers, and 38 dispensaries. Eleven wards (31%) were randomly selected for inclusion in the study, namely: Lwamgasa, Nyaruyeye, Bukoli, Nyarugusu, Nyakamwaga, Butundwe, Chigunga, Nyamiluluma, Bukondo, Nzera, and Lwenzera. Random selection was accomplished by first alphabetically ordering the wards and numbering them from 1 to 35. Then a random numbers table was used to select eleven wards. Based on the Tanzania Demographic and Health Survey (TDHS) 2015/16, Geita performed poorly on components of ANC and more than a half (52.3%) of pregnant women delivered at home [7]. This study is part of a large cohort study that is investigating utilization of maternal and child health services, pregnancy outcomes, birth-related complications, and maternal and child mortality and morbidity up to 4 months postpartum. The study utilized a sequential explanatory mixed method approach. This method has two phases. The first phase is characterized by the collection and analysis of quantitative data. The second phase involves the collection and analysis of qualitative data. This method uses the qualitative findings to further explain and interpret the findings of the quantitative data and was selected as it allowed exploring in detail the challenges pregnant women faced when accessing ANC services and the barriers to utilization of ANC services. A household survey using a cross sectional design involving pregnant women in the third trimester was conducted between September 2016 and August 2017. Based on 2015 total deliveries for the entire district (36,101), we assumed that approximately 10,000 (28%) pregnant women would be residing in the study area with at least 20% (2000) of them being in their third trimester of pregnancy. The study reached 1805 (90%) of the expected pregnant women in the selected wards, of whom 1719 participated in the survey. Of those who were contacted but not included in our final sample, 81 were not residents of a study area, two had had miscarriages, one woman was in the second trimester of her pregnancy based on her last menstruation period, and two women refused to participate because they were not feeling well. The household survey consisted of a face-to-face interview. During the conduct of the household survey interview, only pregnant women in their 3rd trimester who voluntarily consented were invited to participate in the study. In the situation that a pregnant woman was not at home during the time the household survey was conducted, that household was re-visited up to three times to see if the woman was interested in participating. Village leaders, community health workers, and traditional birth attendants assisted in the identification of households with pregnant women. Trained research assistants who were registered nurses, and intern medical doctors conducted the household survey. The research assistants were not health care providers at health facilities in the study areas (they were from college/university and were not employed as health care providers); hence, interview bias was expected to be minimal. The principal investigator (EK) supervised the conduct of the survey. For the household survey, a structured pretested questionnaire was utilized to capture baseline information including socio-demographic characteristics, gestational age, parity, gravidity, obstetric history, immunization status, intermittent preventive treatement status, deworming status, and the use of iron and folate supplements. The questionnaire also captured information on birth preparedness, an anticipated place for delivery, social support after delivery, money saving for any emergency during pregnancy, and the purchase baby’s items, which are included as part of the health education provided through ANC services (seeAdditional file 1). For women who had attended ANC clinics, responses to the household survey questions were crosschecked with their antenatal cards, which are provided to pregnant women who attend ANC clinics. During the household survey, the pregnant women’s blood pressure was checked, weight and height were measured, an identification card for follow-up purposes was provided, and counselling regarding the utilization of ANC services was provided. The household surveys were conducted in the homes of the pregnant women unless otherwise specified by the participant. In those rare cases (only four women), the interview was scheduled at a convenient location identified by the participant, usually at their farm or at a friend’s house. A case study was conducted to explore in depth barriers to utilization of antenatal care services among women in the study area. Women and community health workers from six wards namely Lwenzera, Nzera, Nyarugusu, Bukoli, Lwamgasa, and Bukondo were invited to participate in focus group discussions. Using the results from the analyses of the household survey data, study wards were selected based on the criterion of having a significant number of women who initiated ANC late or who never attended ANC. Community health workers (CHWs) were also invited to participate because of the nature of their work in bridging the community with the health care system. Hence, participants were purposively selected for the focus group discussions (FGDs). FDGs provided the women with a safe environment in which they could present their views and opinions and provided the participants with the opportunity to see that the challenges and barriers that they experienced were similar those of other women. The FGDs also promoted open discussion and sometimes disagreement, and allowed us to observe the group dynamics. Finally, FGDs allowed us to observe whether there was consensus of group members on the challenges faced by women who visited ANC clinics and the barriers to utilization of ANC among pregnant women. We conducted six FGDs with women who had recently delivered babies, many of whom had not utilized ANC and delivery services. Thirty-five women from the selected wards participated in the FGDs with an average of five to six women in each group. Six focus groups were also conducted with CHWs, one in each of the six wards. On average, five to six CHWs, both male and female participated from each ward making 32 CHWs. A semi-structured interview guide was used to facilitate discussion among participants. The key issues explored were: 1) What do you know about maternal and child health services available to you during pregnancy or to pregnant women? 2) What has been your experience with ANC services in your community? and 3) Why do some pregnant women not utilize the ANC services available in their communities or are late in utilizing these services? The research team conducted the FDGs in the Swahili language and where necessary in the local language (Sukuma). Members of the research team who were fluent in Sukuma assisted with translation during discussions. None of the research team members involved in conducting the FGDs provided directed medical care to the participants through the local community health centers. Voice recordings were transcribed, translated into English, and back translated into Swahili to ensure content consistency. Field notes were taken by EK and some of the FGDs were supervised by DD. For the purposes of confidentiality, privacy, and friendly environment, FGDs typically occurred at the village leaders’ offices or at primary schools. However, in some cases, (two focus groups), they took place at the local health clinic. Each FGD took approximately one hour and thirty minutes. ANC service utilization was the outcome of interest. Three levels were examined, namely no attendance at ANC services, attendance within the first trimester, and attendance in the second or third trimester. Epi-Data version 3.1 software was used for data entry with the double entry system feature to reduce data entry errors. This feature allows double entry of the same questionnaire data by two different clerks. During dataset validation, inconsistencies were resolved by reviewing the original questionnaires and editing accordingly. Cleaned data were exported and analyzed using STATA version 13 [27]. The 95% confidence intervals and p values reported were based on a 5% level of statistical significance. Chi-squared tests were used to examine associations between categorical variables. Qualitative data were transcribed and translated into English by two RAs fluent in Swahili and English and cross-checked by EK for discrepancies. Thematic analysis was conducted by EK and reviewed by DD. It involved familiarization with data, identification of the main themes, indexing, charting, mapping, and interpretation. Line by line coding was done manually and identified themes were compared with written field notes for convergence or divergence of ideas. The identified themes were used to gain a deeper understanding of the quantitative results. The data source triangulation was done by having group discussions with community health workers and women in order to confirm the perceived challenges and barriers. The contiguous approach was adapted for data integration at the interpretation and reporting level.