Menu
Menu
Menu
Menu
Background: Disparities in perinatal health care occur worldwide. If the UN Millennium Development Goals in maternal and child health are to be met, this needs to be addressed. This study was conducted to facilitate our understanding of the changing use of maternity care services in a semi-urban community in Entebbe Uganda and to examine the range of antenatal and delivery services received in health care facilities and at home. Methods: We conducted a retrospective community survey among women using structured questionnaires to describe the use of antenatal services and delivery care. Results: In total 413 women reported on their most recent pregnancy. Antenatal care attendance was high with 96% attending once, and 69% the recommended four times. Blood pressure monitoring (95%) and tetanus vaccination (91%) were the services most frequently reported and HIV testing (47%), haematinics (58%) and presumptive treatment for malaria (66%) least frequently. Hospital clinics significantly outperformed public clinics in the quality of antenatal service. A significant improvement in the reported quality of antenatal services received was observed by year (p < 0.001). Improvement in the range and consistency of services at Entebbe Hospital over time was associated with an increase in the numbers who sought care there (p = 0.038). Although 63% delivered their newborn at a local hospital, 11% still delivered at home with no skilled assistance and just under half of these women reported financial/transportation difficulties as the primary reason. Less educated, poorer mothers were more likely to have unskilled/no assistance. Simple newborn care practices were commonly neglected. Only 35% of newborns were breastfed within the first hour and delayed wrapping of newborn infants occurred after 27% of deliveries. Conclusion: Although antenatal services were well utilised, the quality of services varied. Women were able and willing to travel to a facility providing a good service. Access to essential skilled birth attendants remains difficult especially for less educated, poorer women, commonly mediated by financial and transport difficulties and several simple post delivery practices were commonly neglected. These factors need to be addressed to ensure that high quality care reaches the most vulnerable women and infants. © 2007 Tann et al; licensee BioMed Central Ltd.
The survey was carried out amongst the community of Entebbe Municipality and Katabi Subcounty in central Uganda. The study area comprises 9 parishes divided into 47 wards (each with an elected local executive council) and 5 governmental army barracks. The geographical terrain of the study area is diverse. Entebbe Municipality forms a peninsula projecting into Lake Victoria and is located around 40 km Southwest of the Ugandan capital, Kampala. It is transected by Entebbe International Airport. Katabi Subcounty borders Entebbe and extends either side of the Entebbe-Kampala road. The Katabi community combines the semi-urban population within close proximity to the road and the relatively isolated rural fishing communities residing on small peripheral peninsulas extending out into the waters and marshland of Lake Victoria (Figure (Figure11). Study area. Entebbe General Hospital, the major provider of governmental health services in the area, is located centrally within Entebbe town. In 2002, in accordance with Ministry of Health policy, a programme for prevention of mother-to-child HIV transmission (PMTCT) was established by Entebbe Hospital in collaboration with a research study examining the effect of helminth infections on the response to immunisations in childhood (clinical trial registration: ISRCTN32849447). In addition to training, the ongoing research study employed counsellors to facilitate entry into the PMTCT programme and a laboratory technician to perform rapid antibody testing for HIV and syphilis. Women attending the clinic are routinely offered iron and folate supplementation and intermittent presumptive treatment for malaria in pregnancy (IPTp). Surveillance data from the hospital estimates the prevalence of HIV amongst women attending for antenatal care to be around 12%, while that of active syphilis was less than 2% [9]. A number of other health facilities in the area offer maternity services. These include small government health centres (public clinics); private delivery homes owned and run by retired midwives, and private hospitals. We conducted a cross-sectional retrospective community survey to identify experiences of maternity care and practice for each pregnancy within the past five years. Consent to conduct the survey was obtained from local community leaders and participating women gave verbal consent to be involved. The study was reviewed and approved by the ethics committee at the London School of Hygiene and Tropical Medicine, the Ugandan Virus Research Institute (UVRI) Science and Ethics Committee and the Ugandan National Council for Science and Technology. Wards were chosen as the sampling unit and census statistics (Ugandan Bureau of Statistics) were used to estimate the relative population size of each ward. A sample of 40 wards was randomly selected, by probability proportional to size, and we aimed to select 16 households within each ward, expecting that on average two thirds of these households would have at least one woman who had given birth over the last 5 years. Maps were available for the study area with the location of major roads and landmarks but not the location of individual households. Each ward was divided into segments of equal geographical size (approximately 70 m2) and four segments were randomly selected from each ward (ranging between 99 and 641 segments per ward). The centre point of the segment was used to define the starting point for the sampling of 4 households. The first household was the household closest to the start point. Each subsequent household was selected as the closest to the one before it. The head of selected households was approached by a pair of interviewers and asked about the presence of any women in the household who had had a pregnancy within the past five years. No exclusion criteria were applied and all women who reported a pregnancy in the past five years were invited to respond. If any such person was identified, verbal consent was obtained before administration of the survey questionnaire. Where such a woman was resident but not present at that time, interviewers returned to the property on at least one occasion at a later date. Where a woman was unavailable due to working or travel, interviewers made every effort to return at a time when she might be expected to be at home (such as the evening or weekend). We were unable to use antenatal cards and delivery records to crosscheck information because women do not routinely keep these after delivery and they are often retained by health facilities. Structured questionnaires were designed to identify the antenatal, delivery and postpartum experiences of women. Questionnaires were administered in the local language by pairs of interviewers, with all pairs containing at least one female interviewer. Answers were predefined with tick boxes or were described under 'other'. The questionnaire and sampling strategy within households were pre-tested in a ward not selected for inclusion into the survey. Women were asked to answer questions with regard to each pregnancy experienced within the past five years, starting with the most recent. Information sought regarding antenatal care during each pregnancy included the frequency of antenatal visits, the primary place of attendance and the type of services that they received there. In addition, women were asked about their delivery experience for each pregnancy including the place of delivery, what assistance they received during labour, hygiene practices at birth and care of the infant soon after delivery. Data were entered and checked using Excel (Microsoft) and analysed using Stata version 8 (StataCorp, Texas). Analysis relating to pregnancies was restricted to the most recent pregnancy to minimise recall bias and to avoid using a complicated hierarchical analysis to take into account correlation within woman. A negligible number of households contained more than one eligible woman, so the only level of clustering taken into account in the analysis was that within wards. This was done using the complex survey commands within stata. Initial comparisons were made using simple tables (svytab). To assess associations between demographic or socio-economic variables and antenatal or delivery/perinatal care practices, binary outcome variables were developed and associations examined using logistic regression models (svylogit) to adjust for possible confounding variables. To obtain an overall assessment of the quality of antenatal services and postnatal practices at delivery, scores were developed for this study. One point was given for each service received of the seven listed in table table11 for antenatal care, and one point for each of the postnatal practices considered to be beneficial among the six listed for postnatal practices. Delivery practices were not included in the scores as it was felt that these might be inaccurately recalled; further, over 90% of women reported that appropriate procedures had been used, so that these parameters would not discrimate well between comparison groups in an overall assessment. For each score a binary variable was created with values below the median defined as low score, including and above the median as high score. Antenatal services and delivery/perinatal practices during the most recent pregnancy among 413 Ugandan women
N/A
