Background: Maternal health services have a potentially critical role in the improvement of reproductive health. In order to get a better understanding of adolescent mothers’needs we compared health seeking practices of first time adolescent and adult mothers during pregnancy and early motherhood in Wakiso district, Uganda. Methods: This was a cross-sectional study conducted between May and August, 2007 in Wakiso district. A total of 762 women (442 adolescents and 320 adult) were interviewed using a structured questionnaire. We calculated odds ratios with their 95% CI for antenatal and postnatal health care seeking, stigmatisation and violence experienced from parents comparing adolescents to adult first time mothers. STATA V.8 was used for data analysis. Results: Adolescent mothers were significantly more disadvantaged in terms of health care seeking for reproductive health services and faced more challenges during pregnancy and early motherhood compared to adult mothers. Adolescent mothers were more likely to have dropped out of school due to pregnancy (OR = 3.61, 95% CI: 2.40-5.44), less likely to earn a salary (OR = 0.43, 95%CI: 0.24-0.76), and more likely to attend antenatal care visits less than four times compared to adult mothers (OR = 1.52, 95%CI: 1.12-2.07). Adolescents were also more likely to experience violence from parents (OR = 2.07, 95%CI: 1.39-3.08) and to be stigmatized by the community (CI = 1.58, 95%CI: 1.09-2.59). In early motherhood, adolescent mothers were less likely to seek for second and third vaccine doses for their infants [Polio2 (OR = 0.73, 95% CI: 0.55-0.98), Polio3 (OR = 0.70: 95% CI: 0.51-0.95), DPT2 (OR = 0.71, 95% CI: 0.53-0.96), DPT3 (OR = 0.68, 95% CI: 0.50-0.92)] compared to adult mothers. These results are compelling and call for urgent adolescent focused interventions. Conclusion: Adolescents showed poorer health care seeking behaviour for themselves and their children, and experienced increased community stigmatization and violence, suggesting bigger challenges to the adolescent mothers in terms of social support. Adolescent friendly interventions such as pregnancy groups targeting to empower pregnant adolescents providing information on pregnancy, delivery and early childhood care need to be introduced and implemented.
This was a cross-sectional study conducted between May and August 2007 in Wakiso district, which has a total area of 2,704 Km2. The district lies in the central region of Uganda and has a population close to one million. It has two counties and one municipality, 17 sub-counties and 130 parishes with an average of four persons per household. In terms of health infrastructure the district has a total of 93 health facilities including two hospitals. The study population included first time mothers (adolescents aged 13–19 and adults 20–29 years) with a child less than one year. This was to ensure that pregnancy, delivery and early motherhood experiences and practices for both groups were comparable. Bennett’s formula was used to calculate the sample size. The formula also calculates the number of clusters needed to obtain the required sample size [19]. Using Bennett’s formula for cluster sampling the overall total sample size was 616 to be drawn from 88 clusters (villages) with each cluster contributing 7 respondents. However, 146 respondents were added to take care of non response. Using the 2000/01 Uganda population and housing census, we obtained a ratio of adolescents to mothers as 6:4 and we used it to get the number of adolescent and adult mothers we needed in the study. In total, 762 mothers (442 adolescents and 320 adults) were interviewed. A list of villages and their population sizes was obtained from Uganda Bureau of Statistics (UBOS). The selected villages were visited to draw a sampling frame. During this visit, contact with the village leaders and identifying guides for the survey was made. Before data collection, letters were sent to chairpersons of the local councils with details of date of visit and a request for cooperation. Local area council chairpersons were the main point of entry in the field. A mix of multi-stage and cluster sampling techniques was applied (Figure (Figure11). This chart illustrates the sampling strategy. Using multi-stage sampling method [19] 50 percent of sub-counties were randomly selected from the two counties and municipality. Thereafter, 50 percent of the parishes were randomly selected from each of the selected sub-counties. A number of villages which were proportionate to size of the parish were selected using simple random sampling from each parish. The size of the parish was determined by the number of villages. The more the number of villages in a parish, the more villages selected. The total number of villages selected was 88. In each cluster (village) we targeted to get at least 9 respondents. In each selected village we constructed two sampling frames, one comprising female adolescents 13–19 and the other adult women 20–29. Those who qualified to be on the sampling frames were first time mothers with a child not more than one year of age. About nine mothers representing nine eligible households (5 adolescent and 4 adult mothers) were randomly selected from each of the 88 villages using simple random sampling technique by using a table of random numbers. In case there was more than one eligible woman in a household, the one with the lowest last digit of their birth day (day of the month) was selected. One follow-up appointment was made in case an interview could not be carried out. We used a structured questionnaire with closed and a few open ended questions. Ten female graduate social scientists with experience in field data collection conducted the interviews. They all spoke the local language (Luganda) and received training for five days beforehand. The content of the training included the description of the study objectives, methods of data collection and sampling techniques. The focus of the questionnaire was on health seeking behaviour for curative and preventive services during pregnancy and early motherhood. The questionnaire had five sections, namely: general information, socio-demographic characteristics, general health status, reproductive health and child health. The training comprised a one-to-one question and answer simulation exercise including possible responses in the local language. The questionnaire was translated into the local language (Luganda) and back translated into English to ensure consistency of meaning. It was pretested on the fourth day of training in Kasangati (Nangabo sub-county) 15 km from Kampala city. After a debrief session the questions were refined. All 30 pretested questionnaires were entered in a data entry software [20] and analyzed. These questionnaires are not included in our results. Field procedures, recording and editing answers were reviewed. Written consent was obtained from respondents before interviews begun. Interviews, which lasted for an average of 45 minutes, took place in the compound often under tree shed and sometimes inside the house. Only the respondent was present during the interview. Field supervisors checked for completeness, consistency and validity of every filled questionnaire before data entry. The first author (LA) conducted debrief meetings every evening to discuss daily field progress and make adjustments where necessary. Extensive validity, consistency and range checks were embedded in the data entry software by the third author (NMT). The filled questionnaires were entered in the computer using EPIDATA software [20]. These data were then exported to STATA V.8 for data analysis [21]. Frequency distribution of all variables was run to check for any unfamiliar pattern in the process of data entry. Odds ratios (OR) with their corresponding 95% confidence interval (CI) were calculated. All data analysis was performed using STATA V.8 [21]. Statistical significance was based on p value < 0.05. Ethical approval was obtained from the higher degrees, research and ethical committees of the Faculty of Medicine Makerere University in Uganda and the regional research and ethics committee in Stockholm, Sweden. In addition, permission to conduct the study was obtained from Wakiso district local government. Written informed consent from adult mothers and emancipated minors [22] according to Ugandan ethical guidelines was sought from respondents before interviews began. Participants were informed that there were minimal or no risk to their participation in the study, that participation was voluntary, confidential and that they could withdraw their participation anytime during the interview. They were also informed that refusing to participate would not affect the usual services they normally access at health units.
N/A