Prevalence of cultural malpractice during the perinatal period and its determinants among reproductive age women in southwest Ethiopia: A community-based cross-sectional study

Background Cultural malpractices are accepted cultural norms and socially shared practices that have a negative impact on health. Cultural malpractices vary in type and number in different communities. This study aimed to determine the prevalence of cultural malpractice during the perinatal period and identify its predictors among reproductive-age women in rural communities of southwestern Ethiopia. Methods A community-based cross-sectional study was conducted from May 5 to 31, 2019 in Semen Bench district, southwestern Ethiopia; among reproductive-age women who had experienced at least one prior delivery. A systematic random sampling technique was employed to select 422 women for the interview. After collection, the data were entered into EpiData and exported to STATA-14 for further analysis. Descriptive analyses were performed and presented in texts and tables. Besides, binary and multivariable logistic regressions were computed to identify determinants of cultural malpractice. Result A total of 414 women completed the survey, resulting in a response rate of 98%. We found that 26.33% (95% CI: 22.15, 30.85%) had food taboos during pregnancy, 31.88% (95% CI: 27.42, 36.61%) delivered their last child at home, and 33.82% (95% CI: 29.27, 38.6%) practiced pre-lacteal feeding. Lack of formal education (AOR: 11.22, 95% CI: 6.24, 20.15), lack of ANC follow-up (AOR: 10.82, 95% CI: 5.46, 21.42), rural residence (AOR: 6.23, 95% CI: 2.18, 17.78), and avoiding colostrum (AOR: 21.94, 95% CI: 9.73, 49.48) were significantly associated with cultural malpractice during the perinatal period. Conclusion The prevalence of cultural malpractice is notably high in the study area. Hence, community-based measures including expansion of education and promotion of maternal health services are important to reduce cultural malpractice during the perinatal period.

Background: Cultural malpractices are accepted cultural norms and socially shared practices that have a negative impact on health. Cultural malpractices vary in type and number in di erent communities. This study aimed to determine the prevalence of cultural malpractice during the perinatal period and identify its predictors among reproductive-age women in rural communities of southwestern Ethiopia.
Methods: A community-based cross-sectional study was conducted from May to , in Semen Bench district, southwestern Ethiopia; among reproductive-age women who had experienced at least one prior delivery. A systematic random sampling technique was employed to select women for the interview. After collection, the data were entered into EpiData and exported to STATA-for further analysis. Descriptive analyses were performed and presented in texts and tables. Besides, binary and multivariable logistic regressions were computed to identify determinants of cultural malpractice.

Introduction
Different communities have socially shared practices that are deeply rooted in their culture and have an important association with their health (1). Cultural malpractices are accepted cultural norms and socially shared practices that have a negative impact on health (2,3). The perinatal period, which includes pregnancy, delivery, and the post-natal period, is accompanied by various cultural malpractices that substantially affect the health of the mother and her children (4, 5). Globally, several types of cultural malpractices have been reported (2,6,7). Ethiopia, as a multi-ethnic nation, is one of the countries where assorted cultural malpractices particularly during pregnancy, delivery, and post-natal period are practiced; among these, home delivery, pre-lacteal feeding, food prohibition, abdominal massage, and avoiding of colostrum were commonly documented (5,(8)(9)(10)(11). Bench Maji Zone is also a home for such malpractices where home delivery, food prohibition, pre-lacteal feeding, and abdominal massage are commonly exercised (8,12). These malpractices were found to have a significant association with residence, educational status, antenatal care (ANC) follow-up, gravidity, and distance from the health facility (5,10,(13)(14)(15).
Cultural malpractices not only hinder women from utilizing maternal, reproductive, and child health services, but they also have a direct impact on the wellbeing and survival of both the mother and her offspring (2,7,16). Previously, studies have concentrated on the biomedical causes and management of maternal death; yet, they ignored cultural and traditional practices that can deter maternal morbidity and mortality (6,17).
Maternal mortality is extremely high worldwide, accounting for 295,000 pregnancy-related deaths in 2017; almost all (94%) of these deaths occurred in resource-constrained settings (18,19). Ethiopia is one of the nations where a higher number of maternal deaths are reported; a recent demographic and health survey estimated that there were around 412 maternal deaths per 100,000 live births (20). Various strategies, like improving access to reproductive health care and maternal death surveillance and response systems, have been implemented to reduce maternal mortality in Ethiopia; however, it persists as a major public health concern (21,22). Cultural malpractices are among the important contributors to the high maternal mortality rate (23). It is estimated that about 5-15% of maternal deaths are due to cultural malpractice (6,10,24).
Cultural malpractices vary in type and number across different communities (8,9,25,26). Thus, comprehensive evidence about the magnitude and determinant factors of cultural malpractices, especially during the perinatal period across different societies is crucial to reduce maternal mortality. Hence, this study was conducted to determine the prevalence of cultural malpractice during the perinatal period and identify associated factors among reproductive-age women in rural communities of Semen Bench district, southwestern Ethiopia.

Materials and methods
Study area, design, and period A community-based, cross-sectional study was conducted from May 5 to 31, 2019 in the Semen Bench district. It is one of the ten decentralized districts in the Bench Maji zone, Southern Nation, Nationalities, and Peoples Region (SNNP) of Ethiopia. The district is composed of thirty-one Kebeles (the smallest administrative unit in Ethiopia), of which twenty-eight were rural. The estimated total Abbreviations: ANC, antenatal care; AOR, adjusted odds ratio; CI, confidence interval; COR, crude odds ratio. population of the district was 159,480; of which 87,748 (51.26%) were females; the majority of the population in the district (152,973) are rural residents, with agriculture serving as their main economic source (27). There are four health centers and thirty-one health posts that provide maternal and child health services in the district.

Source population
All reproductive-age women in the Semen Bench district who had experienced at least one prior delivery.

Study population
All reproductive-age women in randomly selected kebeles of the Semen Bench district who had experienced at least one prior delivery.

Inclusion criteria
All women who had experienced at least one delivery and were available during the period of data collection were included in the study.

Exclusion criteria
Women who were seriously ill and/or had difficulties to communicate were excluded.

Sample size determination and sampling procedure
The sample size was determined based on the single population proportion formula considering a 95% confidence interval, 5% margin of error, and cultural malpractice proportion of 50.9% from a previous study (28).
where, n = is the desired sample size. Zα/2 = the value of standard normal distribution corresponding to the significance level at α of 5%, which is 1.96. P = is the proportion of cultural malpractice. Therefore, Adding 10% non-response; n = 384 + (10 * 384/100) = 422. Therefore, the final sample size for the study was 422 women of reproductive age.
Out of the thirty-one kebeles in the district, seven kebeles (one urban and six rural) were randomly selected. The study participants .

FIGURE
Schematic presentation of the sampling procedure to assess cultural malpractice during the perinatal period in southwest Ethiopia.
were recruited using systematic random sampling by considering household lists of randomly selected kebeles as sampling frames. For a household without eligible women, the next household was considered. Conversely, for households with more than one eligible woman, one of the women was selected randomly ( Figure 1).

Data collection tool, procedure, and quality control
A structured questionnaire (see Supplementary material), that was adapted from different literature was used (10,11,28). A face-to-face interview was employed to collect the data through house to house survey. The data were collected by seven midwives and they were supervised by two public health professionals. To assure the quality of the study, 2 days training was given to data collectors and supervisors about the objectives of the study and techniques of data collection. The tool was also pretested on 5% of the calculated sample size in one of the kebeles not included in this study. Furthermore, the collected data were checked every day for quality, completeness, and consistency.

Operational definitions
Perinatal period: denotes the period during pregnancy, delivery, and postnatal period (5).
Cultural malpractice: accepted cultural practices that have a negative impact on health (5, 10).
Food taboo: avoiding certain foods during some occasions, like pregnancy, due to cultural beliefs (3,10).
Pre-lacteal feeding: feeding the neonate something other than breast milk in the first 3 days after birth (29).

Data processing and analysis
The collected data were entered into EpiData and exported to STATA version 14 for cleaning, categorizing, and further statistical analysis. Consequently, descriptive and inferential statistics were computed using the software. The bi-variable and multi-variable logistic regressions were employed to identify factors associated with cultural malpractice after checking for assumptions and assessing multicollinearity. Variables with a P-value < 0.25 during bi-variable analyses were chosen for subsequent multi-variable logistic regression. In the multi-variable analysis, significant statistical association was declared at a P-value < 0.05. Besides, the Hosmer-Lemeshow goodness-of-fit test and classification table were used to verify the fitness of the final models.

Socio-demographic characteristics of participants
From the total calculated sample size of 422 reproductive-age women, 414 women completed the survey resulting in a response rate of 98%. Almost two out of every five women had no formal  education. More than two-thirds of the respondents were rural dwellers. Most (84.3%) of the women were married, and the rest were either widowed (11.8%) or divorced (3.9%). More than half of the participants were housewives (Table 1).

Obstetrical characteristics of participants
Among the participants, 178 had an antenatal care (ANC) follow-up. About one-third (35.3%) of the women had an illness during their last pregnancy. Most of the home deliveries were attended by families, and about one-fourth of women avoided feeding colostrum (  their newborn. Other cultural malpractices like abdominal massage, uterine massage, using an unclean blade to cut the umbilical cord, shaking of the abdomen or uterus, applying cow dung or butter on the umbilical stump, late initiation of breastfeeding, and avoiding the colostrum were also reported ( Figure 2).

Factors associated with cultural malpractice
Various factors were found to have a statistically significant association with the most commonly reported cultural malpractices (food taboo, home delivery, and pre-lacteal feeding).
Factors associated with food taboo: residence, educational status, and ANC follow-up were significantly associated with food taboo during pregnancy. Accordingly, rural dwellers were 2.25 (95% CI: 1.15, 4.38) times at higher risk of food taboo than urban residents. Similarly, women who did not attend formal education were having a 2.25 (95% CI: 1.15, 4.38) times higher risk of food taboo as compared to women who attended primary and above education. The odds of food taboo were 4.56 (95% CI: 2.61, 7.97) times higher among women who did not have ANC follow-up than among women who had (Table 3).
Factors associated with home delivery: marital status, educational status, and ANC follow-up significantly determined home delivery practice. Widowed and divorced women had a 5.97 (95% CI: 2.79, 12.78) times higher risk of home delivery than married ones. Women with no formal education were 11.22 (95% CI: 6.24, 20.15) times more likely to give birth at home as compared to women who attended primary and above education. Likewise, women who didn't have ANC follow-up were having a 10.82 (95% CI: 5.46, 21.42) times higher risk of home delivery than women who had ANC follow-up (Table 4).   Factors associated with pre-lacteal feeding: rural dwellers were 6.23 (95% CI: 2.18, 17.78) times more likely to practice pre-lacteal feeding than urban residents. Respondents who were Amhara in their ethnicity were 4.94 (95% CI: 2.09, 11.66) times more likely to practice pre-lacteal feeding than Bench ethnic groups. Similarly, women who didn't attend formal education had a 6.30 (95% CI: 2.67, 14.87) times higher risk of pre-lacteal feeding as compared to women who attended primary and above education. Women who had five or more gravida were 7.17 (95%: 2.95, 17.44) times more likely to practice pre-lacteal feeding than women who had fewer gravida. Likewise, women who didn't have ANC follow-up were having 6.73 (95% CI: 3.1, 14.61) times higher risk of pre-lacteal feeding as compared to women who had ANC follow-up. Women who avoided feeding colostrum to their newborns were 21.94 (95% CI: 9.73, 49.48) times more likely to practice pre-lacteal feeding than their counterparts (Table 5).

Discussion
This study was conducted to estimate the prevalence of cultural malpractice during the perinatal period and identify factors associated with it among reproductive-age women in rural communities of southwest Ethiopia. Consequently, the prevalence of food taboo during pregnancy, home delivery, and pre-lacteal feeding were 26.33, 31.88, and 33.82% respectively. Besides; residence, marital status, educational status, ANC followup, gravidity, and avoidance of colostrum were found to have a significant association with cultural malpractices.
We found that the prevalence of food taboo during pregnancy was 26.33% (95% CI: 22.15, 30.85%). This finding was higher than previous reports from different parts of Ethiopia (10,11,28,30). The possible explanation could be due to differences in residence and period of study. Previous studies were conducted   among urban dwellers whereas the current study mainly included rural women. Besides, this study estimated food taboos during pregnancy, whereas the previous studies investigated food taboos during delivery and the postnatal period. On the other hand, this finding was lower than previous researches from Ethiopia (5,31), and Nepal (25). This could be partly because of sociocultural variations.
Our study also found that the prevalence of home delivery was 31.88% (95% CI: 27.42, 36.61%). This result was higher than previous reports in Ethiopia (13,15,32). This might be due to the difference in the study setting; the previous studies were institutional based while our study was community-based. However, our finding was lower than studies conducted in the Afar and Benishangul Gumuz Regions of Ethiopia (33)(34)(35)(36). The possible explanation could be due to the poor infrastructure in the underserved regions of the country, thus leading to poor access to health care.
Furthermore, this analysis indicated that the prevalence of prelacteal feeding was 33.82% (95% CI: 29.27, 38.6%). This finding is lower than previous researches conducted in the same country (37)(38)(39), this might be due to the difference in time of study; these studies were conducted in earlier times when access and quality of health care were relatively suboptimal. In contrast, our finding was higher than previous reports from various parts of Ethiopia (40)(41)(42)(43). This might be attributed to the socio-cultural variation across the country.
Moreover, we also identified determinants of cultural malpractice among reproductive-age women in southwest Ethiopia. Accordingly, ANC follow-up and educational status had a consistent association with the identified cultural malpractices.
In line with previous studies, women who didn't have ANC follow-up had a higher risk of food taboo, home delivery, and pre-lacteal feeding (5,10,34). This might be attributed to the effect of counseling and client education during ANC (43,44). Similarly, women with no formal education were more likely to practice food taboos, home delivery, and pre-lacteal feeding than women with primary and above education. This association was documented elsewhere (5,10,13,15,29), and could be related to the lack of awareness about the harmful effect of cultural malpractices on the health of the mother and her offspring among women without formal education. The results of this study indicated that residence is significantly associated with food taboos during pregnancy and pre-lacteal feeding. Rural residents were having a higher risk of food taboo and pre-lacteal feeding than urban women. This finding was also reported by previous studies (5,10,45). One possible explanation is that rural residents have limited access to health care and are less exposed to health information (42, 46).
Our study also identified that marital status is significantly associated with home delivery. Divorced and/or widowed women had a higher risk of home delivery as compared to married women. This association was also supported by previous reports (47, 48), and might be due to the enabling effect of partner support on improving maternal and reproductive health care service utilization (49,50).
This study revealed that respondents who were Amhara in their ethnicity were more likely to practice pre-lacteal feeding than Bench ethnic groups. This could be partly due to the difference in socio-cultural attributes between the two ethnic groups. Consistent with previous studies, women who have had five or more gravidity were more likely to practice pre-lacteal feeding than their counterparts (5,26). This could be due to the higher prevalence of cultural beliefs among older women. Lastly, we found that avoiding the colostrum was strongly associated with pre-lacteal feeding. This association was documented elsewhere (29,43,51) and might be due to the misconception among mothers that colostrum is non-nutritious and causes diarrhea (52).

Limitations
The findings of this study should be interpreted considering the following limitations. Firstly, as we investigated events that occurred in the past, recall bias could not be ruled out. Secondly, the cross-sectional nature of the study makes it difficult to establish a temporal association.

Conclusion
The prevalence of cultural malpractice is notably high in the study area. Educational status, ANC follow-up, residence, marital status, gravidity, and avoiding colostrum had a significant association with cultural malpractices during the perinatal period. Hence, community-based measures, including the expansion of education and promotion of maternal health services are important to reduce cultural malpractice during the perinatal period.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement
Ethical clearance was obtained from the College of Health Science, Mizan-Tepi University Ethical Review Board. Considering the non-invasive nature of data collection procedures, verbal consent was obtained from each woman who participated in the survey after a detailed explanation of the purpose of the research and the right to withdraw from the study at any time. For women who practiced cultural malpractice, detailed descriptions of possible health consequences and appropriate alternative measures were discussed by data collectors. Besides, confidentiality was assured by not recording personal identifiers and using the data only for the purpose of this study.