Determinants of Intention To Use Maternity Waiting Homes in Tanzania: A Study Based on Women Using MWH in Dodoma, Iringa and Morogoro


 Background

Maternal mortality rate (MMR) in Tanzania is still persistently high. The data from Tanzania has revealed that maternal mortality ratio (MMR) has increased from 454 per 100,000 live births in 2010 to 556 per 100,000 live births by the year 2015. Establishment of Maternity Waiting Homes (MWH’s) in a few regions has been one of the efforts for trying to serve lives of pregnant women from pregnancy related complication.
Methods

The study adopted a cross-sectional study design and a convenience sampling technique for recruiting the respondents.
Results

The women aged 20–34 are 10 times more likely to use the MWHs in case of future pregnancy (OR = 10.906, p = 0.019) whereas those aged 35 to 49 are 54 times more likely to use MWHs in case of future pregnancy (OR = 54.629, p = 0.006) as compared to those aged 18 to 24 years old. Women who lived 3 to 6 hours of travel to MWH are 11 times more likely to use MWH in case of future pregnancy as compared to those living at the distance of 1 to 3 hours to the MWH (OR = 11.451, p = 0.032).
Conclusion

The results of this study have revealed some interesting findings in the issues of distance by traveling time and age of the respondent to be highly significant factors in predicting intention to use MWHs in case of future pregnancy.


Conclusion
The results of this study have revealed some interesting ndings in the issues of distance by traveling time and age of the respondent to be highly signi cant factors in predicting intention to use MWHs in case of future pregnancy.

Background
The global agenda on reproductive and maternal health has been well stated in the sustainable development goals (SDG3) which targets to reduce maternal deaths to 70 per 100,000 live births by 2030 [1]. Despite of the agreed targets, the recently trends shows that global maternal deaths have increased worldwide from 287,000 in 2010 to 295,000 in 2017, however, the maternal mortality rate (MMR) has signi cantly declined by 35% for the past 17 years from 342 per 100,000 live births to 211 per 100,000 live births [2], [3]. The statistics indicates that women are continuing dying at a steady rate of 810 per day [3]. The global adult lifetime risk of maternal death by 2017 was estimated at 1 in 190 in the world which is nearly half of the lifetime risk during the year 2000 which was 1 in 100 [3]. However, 94% of these deaths occur in low and lower income middle-income countries whereas sub-Saharan Africa accounted for 66% of the global maternal deaths [3]. Based on the existing statistics on maternal mortality and the SDG3 target expected to be achieved by 2030, signi cant efforts must be made to improve maternal health and reduce maternal mortality.
Maternal mortality rate (MMR) in Tanzania is still persistently high. The data from Tanzania has revealed that maternal mortality ratio (MMR) has increased from 454 per 100,000 live births in 2010 to 556 per 100,000 live births by the year 2015 [4]- [7]. Several strategies such as increasing coverage and quality primary health care services for communities in rural and remote areas, timely antenatal care visit, delivery in health facilities and timely postnatal check-up have been adopted by the government in collaboration with the UN agencies and other development partners as a necessary attempt to improve maternal health in the country [5], [8]. In spite of this increase with regard to the last two TDHS, still 37 % of women in the country take risk of delivering their babies out of the health facilities. This has resulted the life time risk of maternal death to stand at 1 in 30 [5]. Some other studies show that 3/4 of maternal deaths result from the direct obstetrical complications of haemorrhage, sepsis, obstructed labour, hypertensive disorders of pregnancy and septic abortion [9], [10].
Major barriers mentioned as limiting these women from accessing better maternal health services are physical distance from the health facilities (lack of transportation), affordability of services, gender and cultural barriers (e.g., women may face unique di culties in accessing delivery services), autonomy in making decision and economic power [11]- [13].
Many consider the experience of adaptation of maternity waiting homes (MWH's) to be a key strategy in bridging the geographical gap in obstetric care in rural areas with poor access to equipped health facilities, and urban areas where the services are available [9]. Maternity waiting homes are buildings which are located near to or within the health facilities. These MWHs offers accommodation to pregnant women with indicators of complications or who live far from health facilities access to obstetric care and opportunity to be visited regularly by a nurse, midwife or doctor [14]. Van Lonkhuijzen, et al, (2009) found that the use of MWHs has been increasing the likelihood of hospital delivery. Tanzania is among the countries which adopted MWHs as the strategy to increase timely access to maternal care for women with risky pregnancy and those living far from the health facilities [16]. MWHs are found in a few regions of Tanzania and were established few decades under the support of the United Nations Fund for Population Activities (UNFPA) as one of the pilot country [17]. Indeed, some studies have revealed that MWH's have been successful in reducing risks and complications of delivery [18], [19].
Nevertheless, the spread of these MWH's over the country has been slow as currently only a few regions are covered. Furthermore, there exists no currently known documented information on a study that has been in place to assess the intention of pregnant women towards using the MWH's services in case of future pregnancy. Therefore, this study aims to identify the factors associated with intention to use MWH's by the women in the three regions of Morogoro, Iringa and Dodoma in Tanzania.

Research Design
The study adopted a cross-sectional survey study design that was conducted from February to April 2021.

Study Area
This study was conducted in three regions of Tanzania Mainland which are Morogoro, Iringa and Dodoma. Selection of the survey areas was based on the criteria of existence of MWHs in these regions that provide services to women waiting for delivery, particularly from rural areas.
De ning the population The population of the study involved women in reproductive age with the size for speci ed regions as per 2012 census estimated as: Morogoro had 539,645 persons, Iringa had 230,283 persons and Dodoma had 471,069 persons [20].
The sampling procedure and sample size The study involved 235 respondents from the main sample (i.e. from the women attending maternity waiting houses in the selected hospitals) who were drawn using a convenience sampling technique based on their availability on the MWHs during the survey. The age range of the women was 15 -47 years.

Data Collection Methods
In this study, primary data was collected by using a standard paper questionnaire from the women attending the MWHs. The questions were asked by skilled nurses who were given a training on the questionnaires regarding how to properly conduct the interviews and record the responses.

Variables of the Study
The study variables are of two categories that includes outcome variable and independent variables (covariates).
Outcome variable: willingness/intention to use MWH in the future Covariates: Socio-demographic, geographical and economic variables that includes region, hospital name, age, marital status, education, occupation, economic status, parity of the respondent, household size and distance to the MWHs by travel time.

Data Processing, Analysis and Presentation Methods
The data collected through standard questionnaires were transformed into computer based database using IBM-SPSS 25. Coding of the paper questionnaire was done before entry into IBM-SPSS 25 data template. The data was then cleaned up and managed by using IBM-SPSS 25 and STATA 15. Finally, descriptive and inferential statistics analyses and models were performed by using STATA 15.

Ethical Approval
Ethical approval was obtained from ethical and research clearance committee of the University of Dodoma. The permits to conduct research were obtained from the respective regional authorities and district administrations. Verbal informed consent to the respondents was obtained from each participant before beginning the interviews. Majority of women were aged 18 to 24 years (38.94%), followed by those aged 25 to 34 at 32.74% and most of the women were either married or cohabiting that is 83.12%. Most of the women were selfemployed at 80.85% and mostly with primary level of education. On the other hand, 76% of the women were from the average income families, 52.04% of the women spent about 1 to 3 hours travelling to the MWH and few women had parity of more than 6 (16.31%). In this case, we run the unadjusted and adjusted multivariate logistic regression model for binary outcomes to determine the factors associated with women intention to use MWHs in case of future pregnancy.

Unadjusted Logistic Regression Model
Upon running the unadjusted multivariate logistic regression model, the covariates of age, parity, household size, economic status of the family and distance of MWH by travel time were signi cant at 5% level as clearly shown in Table 3 below. These signi cant covariates were included in the nal adjusted model ( Table 4) that was interpreted for the factors associated with women willingness to use MWH in case of future pregnancy. The reference category selected for both covariates was that with the highest frequency because some of the rst/last categories had very few observations that could affect the results.  Women with parity of 6 or more are the only signi cant category and are less likely to use MWHs in case of future pregnancy (OR=0.012, p=0.006) as compared to women with parity of 1 pregnancy. On the other hand, poor household status is signi cant as the women from these families are less likely to use MWHs in case of future pregnancy (OR=0.127, p=0.002) as compared to average income families.
Women who lived 3 to 6 hours of travel to MWH are 11 times more likely to use MWH in case of future pregnancy as compared to those living at the distance of 1 to 3 hours to the MWH (OR=11.451, p=0.032).

Conclusion
Since facility based studies like this one lack the results of mothers who never stayed at MWHs, we recommend household survey to be conducted in order to make a more informative comparison between MWH users and nonusers. Furthermore, qualitative studies can be conducted to include key informants on whether these MWHs can provide a long term solution to maternal and child morbidity and mortality in the county as it is still a puzzle yet to be solved.
Likewise, there is a need to build more MWHs nearby to the hospitals that usually receives many referral cases of pregnant women from smaller health centers as women with risky pregnancy needs a closer monitoring by health workers during the time they await delivery.