Postnatal care utilization and associated factors among women of reproductive age Group in Halaba Kulito Town, Southern Ethiopia

A community-based cross-sectional study was conducted from May 1 to 30, 2016 in Halaba kulito town. The town is administrative centre of Halaba special district which is found in Southern Nation Nationality and Peoples Region (SNNPR). It is located 245 km south from Addis Ababa through Worabe district and 90 km from Hawassa, the capital city of SNNPR. The town has five kebeles (smallest administrative units in Ethiopia). The estimated total population residing in the town is 39,507 people, with consisting of males 19,358 (49%) and females 20,149 (51%) of the population. Reproductive age group women account 9205 (23.2%) of total female population. Population density of the town is 4 persons per hectare and the average family size is estimated to be 6. In the town there are one health centre and one primary hospital. Basic essential obstetric care is provided in the health centre while comprehensive essential obstetric care is provided in the primary hospital. All reproductive age women who gave live birth in 2015 and resident of the town were included in the study population. Those women who had difficulty in communication due to severe illness were excluded from the study.

The sample size for PNC utilization was computed by using single population proportion formula by assuming that proportion of PNC utilization was 0.2 [8], 95% confidence interval, 4% margin of error and considering 10% of non-response rate. The final sample size was 422. Whereas double population formula was used to compute sample size to identify factors associated with PNC use. The computation was made via StatCalc application of Epi. Info version 7 with impute of 95% confidence interval, 80% power and case to control ratio 1:1. Odds ratio and expected prevalence of exposure among non-users were taken from studies previously conducted in the country [6, 9, 28]. Some of the exposure variables used to compute sample size were new at least one post postpartum danger sign (N = 214), ANC use (N = 236), and place of delivery (N = 206). Maximum sample size obtained was (total N = 367). However, sample size calculated using single population formula was judged sufficient source to identify associated factors with PNC services utilization.

Five kebeles found in the town were included in the study. Then after, total sample size was proportionally allocated to each kebele and systematic random sampling was used to select households where eligible women were living. First household was selected by lottery methods and then the remaining households were selected by skipping K-intervals. The units of analysis for this study were women (aged 15–49 years) who gave live birth in 2015.

Data were collected using structured, closed ended interviewer administered questionnaire. The questionnaire was prepared in English and translated to Amharic. The dependent variable was PNC service utilization; and independent variables included in the questionnaire were socio-demographic characteristics, obstetric history, awareness about postpartum danger signs, ANC utilization, place of delivery and awareness about PNC service. Before actual data collection started two days training was given for data collectors on the objective, sampling procedure, and process of data collection by the principal investigators. The questionnaire was pre-tested on 10% of the calculated sample size. Additional adjustments in the sequence and wording of the questionnaire were made based on the results of the pre-test. Five BSC in public health participated in data collection through home visit. The collected data were checked daily for completeness and consistency by principal investigators.

Data entry, cleaning, and analysis were done using SPSS version 20.0 software. Binary logistic regression was used to identify independent effect of each independent variables on the outcome variables and multivariable logistic regression analysis used to control confounders and identify factor associated with PNC services utilization. The “Backward: conditional” method was used to perform the multivariable analyses. Interactions were explored and interaction was observed. Absence of multicollinearity among independent variables was checked. Variable whose variance inflation factor (VIF) greater than 10 were not entered to multivariable analysis [31] Appendix 1. The fitness of logistic regression models was assessed using the Hosmer-Lemeshow test. In bivariate and multivariable logistics regression analysis, a significant association was declared when the p-value was less than 0.05 and strength of association was determined by computing odds ratio at 95% CI.

Support letter was obtained from Hawassa University College of Medicine and Health Sciences Ethical review board. Permission to proceed the study was attained from the town health office administrative officials. Verbal consent was obtained from the study participants after the brief explanation given on the objectives as well as the benefit of the study. Confidentiality and privacy of every respondent’s information were ensured by not using any identifiers of the study participants.

One hundred sixty-five (41.1%) respondents were pregnant for three and more times. One hundred forty-one (35.5%) respondents had three and more alive children. The majority of respondents 362 (90.3%) attended ANC during their last pregnancy. Among ANC attendants 111 (30.7%) attended for four and more times. Respondents who gave their last birth at the health center and hospital were 277 (69.1%) and 85 (21.2%) (Table 2).

Overall PNC utilization among reproductive age women was 47.9% (95%CI = (0.43, 0.53). The first within 24 h after birth and the last 42 days after birth visits of PNC were more utilized. Husband and wife jointly made a decision in 91% cases to utilize PNC. Majority 371(92.5%) of respondents aware of at least one danger signs of postpartum. Health professional were leading sources of information 268 (72.2%). More than half (56.9%) of mothers knew at least four PNC services (Table 2).

At bivariate analysis socio-demographic factors associated with PNC utilization were maternal education, maternal occupation, paternal education, paternal occupation, and household monthly income (Table 1). Likewise, factors such as frequency of ANC, aware of at least one postpartum danger sign, and place of delivery were associated with PNC utilization (Table 2).

Interactions were checked for variables found at the final model in multivariable analysis Appendix 2. Among interaction terms created, only knew child care by frequency of ANC had significant effect with PNC services utilization. Hence, interaction term, knew child care by frequency of ANC, was included in the final model. Consequently, factors associated with PNC utilization were maternal occupation, frequency of ANC, place of delivery, aware of at least one postpartum danger sign and knew child care by frequency of ANC. The government employed were 3 times (AOR = 3.01, 95% CI (1.36, 6.67) more likely to utilize PNC as compared to traders. Mothers who had three ANC visits were 4 times (AOR = 4.29, 95% CI = 1.59, 11.55) and who had four ANC visits were 9 times (AOR = 9.55, 95% CI = (3.46, 26.36) more likely to utilize PNC as compared to mothers who had single ANC visit. Mothers from a household where husband alone is a responsible to make decision to use PNC were 3 times more likely to use PNC as compared to a mothers from a household where mother makes decision autonomously. However, decision made by husband alone was not significantly associated with PNC utilization (AOR = 3.32, 95% CI = 0.4, 27.33). Mothers who gave last birth at Health centre were 10.76 times (AOR = 10.76, 95% CI = 3.26, 35.57) and at Hospital were 13 times (AOR = 13.15, 95% CI = (3.64, 47.50) more likely to utilize PNC as compared to mothers who gave their last birth at home. Odds of PNC utilization among reproductive age women who didn’t aware for at least one danger signs of postpartum were .06 times (AOR = 0.06, 95% CI = (0.01, 0.37) less likely to utilize PNC as compared to women who aware at least one postpartum danger signs. Mothers who didn’t know child care and had three ANC visits were 0.14 times (AOR = 0.13, 95% CI = 0.02, 0.8) less likely to utilize PNC services as compared to mothers knew child care and had one ANC visit (Table 3).

The conclusion of this study was derived from primary data through rigours descriptive and analytic data analysis. However, the study had some limitations. Hence, interpretation of the results needs certain consideration. Because the study use cross-sectional data, the study unable to conclude definite temporal relationship between independent and dependent variables. So that, associations were looked for. Also, in this study data were collected retrospectively, this might introduce recall bias. To reduce recall bias women who gave a live birth in 2015 only were recruited for the study. Small sample size is another limitation of this study. Single population proportion formula was used to derive optimum sample size for PNC service utilization and double population proportion formula was used to derive optimum sample size to determine associated factors. However, sample size calculated using single population formula was judged sufficient source to identify associated factors with PNC services utilization.