The Dimensions and Determinants of Health-promoting Behavior among Postpartum Women in Bahir Dar City, Ethiopia

Background The postpartum period, known as Aras bét in Amharic, has been given little research, practice, and policy attention in Ethiopia.Objective This study examined the dimensions and determinants of postpartum women health-promoting behaviors in Bahir Dar city, Ethiopia. Method A facility-based cross-sectional study design was employed. Using the Health Promoting Life Prole (HPLP) Questionnaire, quantitative data were drawn from 178 randomly selected mothers who had attended at least one postnatal care visit. Data was analyzed using descriptive and inferential statistics. Results Postpartum women reported moderate levels of overall health-promoting behavior (M=141.62; SD=22.44). Across the subscales, spiritual growth (M= 28.81; SD, 5.35) and physical activity (M=18.16; SD=4.49) showed the highest and the lowest average HPLP score respectively. The t-test and ANOVA results showed that health-promoting behaviors varied across various sociodemographic variables such as religion groups, educational level, number of pregnancies, family type, employment status, monthly income, delivery place, delivery type, ANC and PNC attendance (p<0.05). In the binary logistic regression model, postpartum women who had no schooling were 16% less likely to have “better” overall HPLP-II score as compared to their college/university graduated counterparts ( p <0.05). Implications and Conclusion This study differentiated the dimensions and determinants of postpartum health-promoting behaviors. The study offers implications for social and health service providers and future researchers to consider the holistic dimensions of health promotion during the continuum of care for pregnancy. postpartum women’s relationship with family members and signicant others. Spiritual growth (9 items): referred to postpartum women’s optimistic views to nd meaning in their life after delivery and their practices of religious and traditional rituals. Stress management (8 items): referred to postpartum women’s strategies to respond or cope with stressful life events as a result of the demands of self-care needs, child-care, and parenting roles. HPLP-II has four Likert-type scales with responses ranging from Never (1) to Routinely (4).

The study setting was Bahir Dar city--the capital of Amhara Region, Ethiopia. The city was purposively selected as a study site hence Amhara region is still beset by high maternal mortality and morbidity rates even higher than the national average [1]. Study participants were purposively selected from the Bahir Dar City Administration Health Centre and the Family Planning and Guidance Centre in the urban area and from Tis Abay Zuriya Health Centre and Zenezelema Health Centre were located in the rural-urban periphery of the city.

Study population and Sample Size Determination
Mothers who gave birth within four months prior to the data collection and visited health institutions for postnatal care at least once were designated as the sampling frame. The sample size for this study was determined using a single population proportion formula. Taking the national postnatal care service coverage (12%) (p=0.12) [1], q (1-0.12)=0.88 with a level of signi cance of5% (a=0.05), Z(α/2)=1.96 and margin of error to be 5% (e= 0.05), and adding a nonresponse rate of 10%, a total sample size of 178 postnatal service attending mothers were recruited to complete the questionnaire. The sample size was allocated proportionally among four selected health centers.

Instruments for Data Collection
The health-promoting life pro le (HPLP-II) questionnaire was developed by Walker and his colleagues [24] based on Pender's health promotion model to measure health-promoting behaviors and the improvement of health and quality of life. Health-promoting behavior was operationalized as postpartum women practices of self-care to maintain healthy life style after delivery. This questionnaire consists of 52 items on the six sub-scales of health-promoting behaviors. Health responsibility (9 items): referred to postpartum women's activities related to adopting health-seeking behavior. Physical activity (8 items): referred to postpartum women's engagement in daily routines and formal exercises to maintain the balance of their body's weight. Nutritional management (9 items): referred to postpartum women eating pattern for promoting healing and recovery, and for the su ciency of breastfeeding. Interpersonal relationship (9 items): referred to postpartum women's relationship with family members and signi cant others. Spiritual growth (9 items): referred to postpartum women's optimistic views to nd meaning in their life after delivery and their practices of religious and traditional rituals. Stress management (8 items): referred to postpartum women's strategies to respond or cope with stressful life events as a result of the demands of self-care needs, child-care, and parenting roles. HPLP-II has four Likert-type scales with responses ranging from Never (1) to Routinely (4).
Moreover, a personal information form that contained 12 variables was used to explore socio-demographic characteristics of postpartum women. Data were collected about the mothers' age, religion, marital status, educational background, income source, family income level, employment status, family type, number of pregnancies, place of delivery, nature of delivery, antenatal care, and postnatal care attendance. Variables were measured categorically and coded accordingly.

Data Collection Procedure
Four health extension workers were recruited and oriented about the data collection procedure. Most of the questionnaires were lled out in the waiting rooms in the health centers while mothers came for postnatal visits or babies' vaccination. From the randomly selected respondents, those who did not come to the health center were reached through home visits.

Reliability and Validity Testing
In this study, the Cronbach Alpha coe cient for the HPLP II total scale was 0.94 and had high reliability. Alpha coe cients for the subscales ranged from 0.72 to 0.90. So far, there has been no published data on the validity of the instrument in an Ethiopian postpartum mother's context. Hence, to control for the threats to validity of the data by the proposed instruments, content validity of the instrument was assessed by a health extension worker and a social worker who had experience of working with postpartum women. Based on the practitioners' suggestion, the translation of postpartum support questionnaire from English to Amharic was carefully constructed.

Data Analysis
Data were analyzed using both descriptive and inferential statistics using the Statistical Package for Social Science (SPSS-PC version 20.0) program. Frequencies, percentages, means, and standard deviations were used to describe the sociodemographic characteristics and to determine the levels of healthpromoting behavior.
The statistical differences between the groups of independent variables with that of HPLP pro le scores were analyzed using independent samples t-tests and one-way analysis of variance (ANOVA). Binary logistic regression was used to predict the effect of each of the independent variables on the dependent variable (HPLP and its subscales). The overall PSQ and its subscales score were dichotomized into better and lower using the median scores as a reference. A p-value of 0.05 was considered to be statistically signi cant.

Results And Discussion
The socio-demographic characteristics of 178 postpartum women respondents are presented in Table 1. The mean age of the study sample was 28.7 (SD 4.95). Religiously, 145 (81.5%) respondents were Christians and 33 (18.5%) were Muslims. More than half of postpartum women (58.4%) lived with their nuclear family; whereas 41.6% lived in extended families. Over one-quarter (27%) of respondents were rst time mothers and the remaining 73% were experienced mothers. The mean family income of the respondents was $71 per month. The overall obtained score for HPLP-II for the study sample ranged from 71 to 185 (M=141.62; SD=22.44). The majority of respondents (53.4%) had an overall HPLP-II score equal to or greater than the average. With regard to the subscales, postpartum women in this study showed the highest mean score in spiritual growth (SG) (28.81+5.35) followed by nutritional management (NU) (27.59+4.68), interpersonal relationships (IR) (25.26+4.49), health responsibility (HR) (21.91+ 6.71), and stress management (SM) (19.89+4.91). Physical activity (PA) showed the least average score (18.16+4.49).
Similar result of moderate level of HPLP score was obtained among Turkish worker women [25; 26]; Taiwan pregnant women [13]. This means that healthpromoting life-style behavior can reach up to the moderate level of awareness despite differences in women's cultural and economic status. A study in Turkey reported highest mean score was found for interpersonal support (74.3±14.
The binary logistic regression model across the subscales showed that marital status, marital status, educational background, number of pregnancies, and PNC attendance showed a statistically signi cant relationship (p<0.05) with postpartum women's spiritual growth and nutritional management.
A study by Ahmadi and Roosta [9] determined the correlates of health-promoting behavior among reproductive age women in Iran. The results indicated a signi cant association between women's education, husbands' education, socio-economic class, and health-promoting lifestyle.
Mirghafourvand et al. [10] evaluated health-promoting behaviors among women of reproductive age in Iran. A statistically signi cant relationship was found between health responsibility and age, primary support source, marital status, education, spouse's education and a crowding index (p< 0.05). A statistically signi cant relationship was found between stress management and social support, education, occupation, marital status, spouse's education and body-mass index (p<0.05). The study also found a statistically signi cant relationship between physical activity and social support, age, marital status, education, occupation, and spouse's occupation. Table 4 about her

Limitations of the Study
Ethiopia is a home for more than 80 culturally diverse ethnic groups. This study explored health-promoting behavior in Amhara ethnic group residing in Bahir Dar city and might not re ect the beliefs and practices of postpartum women in the entire country. Study participants were drawn from postpartum women who had attended PNC at least once. Hence, the ndings cannot be generalized to postpartum women who had no visit to a health center. Moreover, the ndings are relevant to the situation of postpartum women who were in their rst four months after delivery. Conceptually, the proposed study employed an instrument which was developed in the western context. The researchers took caution to consider culturally relevant concepts during translating the instruments into Amharic based on pilot tests and feedback from experts. However, the modi ed instrument might not convey the same message as the original scales.

Implications
Understanding postpartum women's health-promoting behaviors and its determinants could help interdisciplinary approach between health and social service providers to design comprehensive and culturally-sensitive strategies to promote postpartum women's health. Taking into account the scarcity of research in the area of maternal health promotion in Ethiopia, possible future qualitative research could be generated explore mothers' unique experiences, to identify opportunities and challenges, and roles of the informal and formal social networks in promoting maternal health during postpartum period.

Conclusion
This study is one of the few to bring an insight to those sociodemographic factors that carry signi cant weight in promoting mothers' health during postpartum period. The dimensions of health-promoting behaviors found to vary across different sociodemographic context in which women are immersed. Acknowledgements: We would like to acknowledge the study participants and the health sector staff who gave their time to the study. We are also grateful to the data collection team. The authors declare that this paper embodies only their own work and all the sources used are indicated or acknowledged by means of a complete reference list.
Availability of data and materials: All data generated or analysed during this study are included in this published article.
Authors' contributions: KS has developed the idea for the paper. AM was the supervisor and SZ drafted the initial version and critically reviewed the draft and provided additional content. All authors approved the nal version.
Ethics approval and consent to participate: Study participants were recruited voluntarily. Maximum care was taken to safeguard the con dentiality of participants' personal information. Verbal consent was sought and documented by the respondents. Ethical clearance was obtained from the Institutional Review Boards of Amhara Regional Health Bureau and research ethics review committee at Bahir Dar University.