Factors associated with home births in Benin and Mali: evidence from the recent demographic and health surveys : Global Reproductive Health

Journal Logo

Original Article

Factors associated with home births in Benin and Mali: evidence from the recent demographic and health surveys

Bado, Aristide R. PhDa,b,; Badolo, Hermann MScc,d; Johnson, Ermel MDb; Komboigo, Béwendin E. MDe; Padonou, Sètondji G.R. MDf; Diawara, Fatou MDg

Author Information
Global Reproductive Health: Autumn 2022 - Volume 7 - Issue 3 - p e57
doi: 10.1097/GRH.0000000000000057
  • Open


The reduction of maternal mortality and morbidity is an important goal on the health agenda of international, regional, and national organizations/programs. Since the Nairobi International Conference on Safe Motherhood in 1987, the international community has become aware of the seriousness of maternal mortality in the world, especially in developing countries1. The earnest pursuit of lowering the incidence of maternal mortality has been further affirmed at international summits. Indeed, at the Millennium Summit in September 2000, the world’s heads of state set themselves the goal of reducing maternal mortality worldwide by 75% by 20152. Similarly, the Sustainable Development Goals (SDGs) have placed a strong emphasis on decreasing maternal mortality by aiming to reduce the global maternal mortality ratio (MMR) to below 70 per 100,000 live births by 2030 and ensuring universal access to sexual and reproductive health care services, including family planning, information and education, and the integration of reproductive health into national strategies and programs3. Maternal health issues are one of the world’s major health challenges; however, over the past 3 decades, significant progress has been made resulting in a 40% decline in the global MMR between 2010 and 20144. Despite this progress, worldwide, 295,000 women died because of pregnancy and childbirth in 20172. Notably, sub-Saharan Africa has the highest MMR among the 7 regions of the world, with 534 deaths per 100,000 live births2.

Homebirth is one of the main reasons for the high MMR in sub-Saharan Africa5. Sub-Saharan Africa and South Asia together contribute to >85% of maternal deaths. Of these, only half of the deliveries are institutional6. There is growing evidence that the high MMR in sub-Saharan Africa is strongly linked to homebirth as most of the births in this region take place at home7. Homebirths, more so in the absence of trained professional attendants, have been associated with adverse infant and maternal outcomes8,9. Institutional delivery is considered the most critical intervention to reduce maternal mortality and ensure safe motherhood10. The World Health Organisation (WHO) recommends that every delivery be supervised by a skilled attendant, a health professional who can identify and manage normal labor and delivery, identify and treat complications, and/or provide basic care and referral11. However, the proportion of births attended by skilled health personnel is still below the recommended levels12 despite national efforts to increase the availability of maternal health services13 and implement programs to subsidise normal deliveries and/or caesarean sections in some countries14,15.

In West Africa, many women give birth without the assistance of a skilled health professional; according to the results of the Demographic and Health Surveys (DHSs), homebirth rates were 59% in Nigeria (2018)16, 47.3% in Guinea (2018)17, 33% in Mali (2018)18, 16% in Sierra Leone (2019)19, and 14.5% in Benin (2018)20. Such high proportions of homebirth are associated with high maternal mortality and morbidity and under-5 mortality in these countries21. Identifying and understanding the factors related to home births can, therefore, contribute to improving maternal and child health and achieving the related SDG. Benin and Mali are 2 French-speaking countries in the Economic Community of West African States (ECOWAS) that have implemented Emergency Obstetric and Neonatal Care (EmONC) policies to alleviate the economic burden on women during childbirth. These 2 countries also conducted a DHS in 2018 that collected recent data to study the determinants associated with homebirth. The objective of this study was to carry out a comparative analysis of the factors associated with homebirths in Benin and Mali.


Study area

The 2 countries examined in this study (Republic of Benin and Republic of Mali) are in West Africa. The population of Benin is estimated by the United Nations to be 12.4 million in 2021 (https://www.unfpa.org/fr/data/world-population-dashboard). The average number of children born per woman of childbearing age was 5.7 in 2017–201820. Benin’s Human Development Index (HDI) for 2019 was 0.545, which places the country in the 158th position out of 189 countries and territories. According to the results of the 2018 Integrated Regional Survey on Employment and the Informal Sector, nearly 7 out of 10 school-age children were enrolled in school, and the literacy rate among those aged 15 years or more was 41.7%22. Mali’s population is estimated by the United Nations to be 20.9 million in 20211, 51% of which are women. The total fertility rate remains high in Mali, with an average of 6.3 children per woman18. Three-quarters of the population lives in rural areas (74.5%) [CPS/SSDSPF, INSTAT, INFO-STAT, Centre d’Études et d’Information Statistiques (INFO-STAT), 2014, Enquête Démographique et de Santé (EDMS-V) 2012-2013 Mali, Document Ronéotypé, USAID/ INSTAT/Coopération Canadienne, mai, p 4. http://www.worldbank.org/en/country/mali/overview, consulté le 07 décembre 2015.], and the population is unevenly dispersed over a vast territory of 1240,192 km2. Mali’s HDI for 2019 was 0.434, placing it in the “low human development” category. In Mali, in 2016, nearly 6 of 10 children (60.2%) aged 7–12 did not attend the first cycle of basic education, and only 31.0% of the adults were literate23.

At the health level, Benin’s health system is pyramidal with 3 levels: the central or national level constitutes the Ministry of Health, its programs, and the national hospitals; the intermediate or departmental level includes the departmental health directorates, their services, and the departmental hospitals; and the operational or peripheral level is represented by the health zones24. Each health zone comprises 1–4 communes. The health zone is further divided into health areas that group together villages or neighborhoods25.

The Malian health system is structured on the following 3 levels: (i) the operational level of technical support to the community health centers, consisting of the district management team; (ii) the intermediate regional level of technical support, made up of the Regional Health Directorates; and (iii) the strategic national level, comprising the Minister’s office, the General Secretariat, and the central services of the Ministry of Health and Public Hygiene of Mali26. The health services include all the public (state and local authorities), private, community (associations, mutual societies, foundations, etc.), and denominational structures and bodies, as well as the professional health orders, whose action contributes to the implementation of the national health policy.

Data sources

The study employed the most recent data from the DHSs conducted in Mali and Benin in 2018. The sample for this study included women aged 15–49 years who gave birth in the 2 years preceding a survey in each of the 2 countries conducted for this study. A total of 15,362 and 10,519 women were interviewed in the survey in Benin and Mali, respectively, while only 8994 in Benin and 6368 in Mali were eligible for the current study.

The DHS was conducted using a 2-stage cluster sampling design, based on enumeration areas and household samples. The detailed methodology of the surveys was described in the final reports18,20. The DHSs are nationally representative, cross-sectional surveys that collect information on a wide range of public health topics, such as anthropometric, demographic, socio-economic, family planning, and domestic violence data, to name a few. The survey covered men and women aged 15–49 years and children younger than 5 years of age residing in noninstitutional settings. The DHS data sets are available to researchers on the DHS website at http://dhsprogram.com/data/available-datasets.cfm.


Variable outcome

The outcome variable was homebirth, which was generated from the variable “place of birth” recoded as 1 if the birth took place at home and 0 if the birth took place in another place (hospital/clinic/health facility/other). This variable was collected from women based on their self-reports.

Independent variables

The explanatory variables were age, level of education, distance to a health center, place of residence, the household’s standard of living, occupation of the woman, number of antenatal visits made, number of children of the woman, experience of an interrupted pregnancy, sex of the head of the household, the spouse’s level of education and occupation, frequency of media exposure, and internet use. These variables were considered because of their statistically significant relationships with the place of delivery observed in previous studies and their availability in the data set. The primary survey unit (PSU) was considered in the analyses to measure the effect of context on the choice of delivery location.

Statistical analyses

We analyzed the collected data using Stata version 16.0. The analyses included both descriptive statistics (frequencies and percentages) to describe the respondents’ characteristics and multilevel logistic regression analyses to identify the determinants. The results were presented as adjusted odds ratios (aOR), with corresponding 95% confidence intervals (CI) indicating their level of precision. Statistical significance was reported at P<0.05. Intraclass correlations were used to measure intraclass variables27.

Ethical considerations

This study is based on the analysis of secondary data without using any identified information about the participant. All DHSs were approved by ICF International and a national ethics committee in each of the host countries. All participants gave informed written consent before taking part in the survey. In this study, additional ethics approval was not required, but we did obtain written permission from the DHS Program to use the data.


Descriptive analyses and χ2 test results

Table 1 presents the results on homebirths according to the explanatory variables considered in this study for Benin and Mali. The prevalence of homebirths is 13.7% in Benin and 32.9% in Mali. The results of the χ2 test show that the variables with a statistically significant association with homebirth are almost the same in both countries. In Benin, the women’s level of education, marital status, problem of distance to a health center, place of residence, household’ standard of living, the woman’s occupation, number of prenatal visits made, number of children the woman has, the experience of an interrupted pregnancy, sex of the head of household, relationship with the head of household, education and occupation of the spouse, frequency of media exposure and internet use are the variables that were found to be linked to the place of delivery. The same variables (except the experience of an aborted pregnancy) are associated with the dependent variable in Mali. In both countries, no education (19% in Benin and 40% in Mali), having a problem of distance to a health center (24% in Benin and 52.3% in Mali), not having a prenatal consultation (68.2% in Benin and 74.1% in Mali), women having 7 or more children (22.5% in Benin and 37.6% in Mali), having an uneducated spouse (22.5% in Benin and 40.4% in Mali), the frequency of paying attention to the media (21.1% in Benin and 50.8% in Mali), and not using the internet (14.4% in Benin and 36.1% in Mali) are the factors that predispose a woman to homebirth.

Table 1 - Descriptive and χ2 test results.
Benin Mali
Home Delivery Home Delivery
Variables n (%) No (%) Yes (%) P n (%) No (%) Yes (%) P
Age (y)
 15–24 2407 (26.8) 87.2 12.8 0.101 2021 (31.7) 68.6 31.4 0.146
 25–34 4372 (48.6) 86.4 13.6 2854 (44.8) 66.8 33.2
 35–49 2215 (24.6) 85.1 14.9 1493 (23.4) 65.6 34.4
Mother’s education
 No education 5786 (64.3) 81 19 4547 (71.4) 59.4 40.6 0.001
 Primary 1622 (18.0) 93.3 6.7 797 (12.5) 78 22
 Secondary + 1586 (17.6) 98.2 1.8 1024 (16.1) 92.5 7.5
Current marital status
 Never in union 302 (3.4) 92.4 7.6 0.000 169 (2.7) 79.9 20.1 0.001
 Married 6607 (73.5) 85.5 14.5 6034 (94.8) 66.9 33.1
 Living with partner 1724 (19.2) 87.8 12.2 32 (0.5) 68.8 31.2
 Widowed/divorced/separated 361 (4.0) 88.4 11.6 133 (2.1) 60.2 39.8
Problem of distance to a health center
 Big problem 3024 (33.6) 76 24 0.001 2048 (32.2) 47.7 52.3 0.001
 Not a big problem 5970 (66.4) 91.5 8.5 4320 (67.8) 76.3 23.7
 Urban 3636 (40.4) 91.8 8.2 0.001 1726 (27.1) 83.1 16.9 0.001
 Rural 5358 (59.6) 82.5 17.5 4642 (72.9) 61.1 38.9
Household wealth index
 Poorest 1889 (21.0) 65.5 34.5 0.001 1190 (18.7) 43.0 57.0 0.001
 Poorer 1810 (20.1) 82.8 17.2 1260 (19.8) 52.4 47.6
 Middle 1773 (19.7) 88.8 11.2 1334 (20.9) 62.8 37.2
 Richer 1785 (19.9) 96.3 3.7 1361 (21.4) 81.6 18.4
 Richest 1737 (19.3) 99.6 0.4 1223 (19.2) 94.2 5.8
Women’s occupation
 Not working 1538 (17.1) 83 17 0.001 2901 (45.6) 62.8 37.2 0.001
 Professional/technical/managerial/clerical 440 (4.9) 94.1 5.9 166 (2.6) 94 6
 Agricultural and self-employed 3144 (35.0) 79.2 20.8 1442 (22.6) 63 37
 Sales 3872 (43.1) 92.5 7.5 1859 (29.2) 74.6 25.4
No. antenatal care visits during pregnancy
 0 visits 1030 (11.5) 31.8 68.2 0.001 1432 (22.5) 25.9 74.1 0.001
 1–2 visits 1646 (18.3) 84.2 15.8 927 (14.6) 62 38
 3–4 visits 1476 (16.4) 93.2 6.8 1195 (18.8) 74.6 25.4
 5 and more 4842 (53.8) 96.5 3.5 2814 (44.2) 86.5 13.5
No. children
 1–2 3366 (37.4) 89.8 10.2 0.001 2210 (34.7) 72.4 27.6 0.001
 3–4 2793 (31.1) 87.1 12.9 1819 (28.6) 65.5 34.5
 5–6 1679 (18.7) 83.9 16.1 1268 (19.9) 64.2 35.8
 7 &+ 1156 (12.9) 77.5 22.5 1071 (16.8) 62.4 37.6
Ever had a terminated pregnancy
 Yes 1097 (12.2) 90.7 9.3 0.001 747 (11.7) 68 32 0.569
 No 7897 (87.8) 85.7 14.3 5621 (88.3) 67 33
Sex of household head
 Male 7471 (83.1) 85 15 0.001 5512 (86.6) 67.8 32.2 0.002
 Female 1523 (16.9) 92.5 7.5 856 (13.4) 62.5 37.5
Relationship to household head
 Wife 5325 (59.2) 87.5 12.5 0.001 4746 (74.5) 66.8 33.2
 Head 893 (9.9) 91 9 592 (9.3) 61.1 38.9
 Daughter 630 (7.0) 91.4 8.6 299 (4.7) 69.9 30.1
 Daughter-in-law 780 (8.7) 81.2 18.8 383 (6.0) 70.2 29.8
 Other relative 1366 (15.2) 79 21 348 (5.5) 75 25
Husband/partner’s educational attainment
 No education 4373 (48.6) 77.5 22.5 0.001 4332 (68.0) 59.6 40.4 0.001
 Primary 1585 (17.6) 93.3 6.7 552 (8.7) 76.1 23.9
 Secondary & + 1977 (22.0) 97.1 2.9 967 (15.2) 90.6 9.4
 Not precised 396 (4.4) 94.4 5.6 215 (3.4) 83.7 16.3
 Missing information 663 (7.4) 90.2 9.8 302 (4.7) 71.2 28.8
Husband occupation
 Not working 1666 (18.5) 83.1 16.9 0.001 2946 (46.3) 63.2 36.8 0.001
 Professional/technical/managerial 295 (3.3) 98.3 1.7 121 (1.9) 94.2 5.8
 Sales 2806 (31.2) 91.6 8.4 1489 (23.4) 78.4 21.6
 Agricultural—self-employed 2168 (24.1) 72.8 27.2 1441 (22.6) 63 37
 Services/skilled manual 2059 (22.9) 94 6 371 (5.8) 59.3 40.7
Sex of household head
 Male 7471 (83.1) 85 15 0.001 5512 (86.6) 67.8 32.2 0.002
 Female 1523 (16.9) 92.5 7.5 856 (13.4) 62.5 37.5
Frequency of media exposure (radio, TV, magazines)
 Not at all 3518 (39.1) 78.9 21.1 0.001 1370 (21.5) 49.2 50.8 0.001
 Often 4211 (46.8) 89.1 10.9 3230 (50.7) 66.1 33.9
 Regularly 1265 (14.1) 97.6 2.4 1768 (27.8) 82.8 17.2
Use of internet
 Never 8528 (94.8) 85.6 14.4 0.001 5603 (88.0) 63.9 36.1 0.001
 Yes 466 (5.2) 99.5 0.5 765 (12.0) 92.1 7.9
N 8994 (100) 86.3 13.7 6368 (100) 67.1 32.9

Results of the multivariate multilevel logistic regression analysis

Measures of variation (random effects)

As shown in Table 2, model 0 (empty model), there is significant variation in the probability of giving birth at home across PSUs in both Benin and Mali (P<0.001). According to the intraclass correlation, the act of giving birth at home could be attributed respectively to factors linked to the PSUs, which in several cases correspond to villages in the rural environment and individual factors. The variations between the PSUs remained statistically significant, even after controlling for all the factors in the model 1 (with all the independent variables) in Benin and Mali.

Table 2 - Multilevel logistic regression analysis on the predictors of home deliveries in Benin and Mali.
Benin Mali
Model 1 Model 1
Variables Model 0 Odds Ratio (95% CI) Model 0 Odds Ratio (95% CI)
Fixed effects
 Age (y)
  15–24 1 1
  25–34 1.05 (0.80–1.36) 0.74 (0.58–0.94)**
  35–49 0.97 (0.68–1.39) 0.66 (0.48–0.91)**
Mother’s education
  No education 2.52 (1.59–4.00)*** 2.21 (1.58–3.10)***
  Primary 2.08 (1.27–3.43)** 1.42 (0.96–2.10)
  Secondary + 1 1
Current marital status
 Never in union 1 1
  Married 0.54 (0.27–1.04) 0.93 (0.48–1.81)
  Living with partner 0.98 (0.49–1.94) 3.85 (1.09–13.51)**
  Widowed/divorced/separated 0.54 (0.25–1.17) 1.25 (0.56–2.80)
Problem of distance to a health center
 Big problem 1.27 (1.05–1.54)** 1.32 (1.10–1.59)**
 Not a big problem 1 1
 Urban 1 1
 Rural 1.62 (1.14–2.30)** 1.76 (1.05–2.98)***
Household wealth index
 Poorest 14.5 (6.1–34.40)*** 3.63 (2.16–6.12)***
 Poorer 11.8 (5.0–27.9)*** 3.1 (1.85–5.14)***
 Middle 9.6 (4.1–22.6)*** 2.74 (1.67–4.51)***
 Richer 5.2 (2.2–1.5)*** 1.78 (1.17–2.72)**
 Richest 1 1
Women’s occupation
 Not working 1 1
 Professional/technical/managerial/clerical 1.13 (0.54–2.33) 0.34 (0.07–1.48)
 Agricultural and self-employed 0.59 (0.08–3.93) 0.006 (0.001–0.009)
 Sales 0.45 (0.06–3.16) 1.10 (0.74–1.64)
No. antenatal care visits during pregnancy
 0 visits 31.32 (24.10–40.70)*** 12.91 (10.21–16.33)***
 1–2 visits 3.07 (2.41–3.91)*** 2.37 (1.87–3.00)***
 3–4 visits 1.40 (1.05–1.86)** 1.42 (1.14–1.78)**
 5 and more 1 1
No. children
 1–2 0.59 (0.41–0.84)** 0.58 (0.42–0.82)**
 3–4 0.71 (0.52–0.97)** 0.94 (0.71–1.26)
 5–6 0.84 (0.62–1.12) 0.96 (0.73–1.26)
 7 &+ 1 1
Ever had a terminated pregnancy
 Yes 1 1
 No 1.00 (0.74–1.35) 1.13 (0.88–1.45)
Sex of household head
 Male 1 1
 Female 0.67 (0.41–1.09) 1.47 (0.92–2.37)
Relationship to household head
 Wife 1 1
 Head 1.62 (0.91–2.88) 0.76 (0.44–1.30)
 Daughter 0.84 (0.52–1.34) 0.60 (0.36–1.00)
 Daughter-in-law 1.30 (0.95–1.78) 1.14 (0.77–1.66)
 Other relative 1.31 (1.04–1.66)** 0.68 (0.43–1.07)
Husband occupation
 No education 1 1
 Professional/technical/managerial 0.94 (0.26–3.36) 1.46 (0.24–8.89)
 Sales 2.12 (029–15.22) 0.72 (0.47–1.11)
 Agricultural—self-employed 2.48 (0.37–16.57) 71629.58
 Services/skilled manual 1.70 (0.24–11.79) 1
Frequency of media exposure (radio, TV, magazines)
 Not at all 1 1
 Often 0.77 (0.64–0.93)** 0.91 (0.74–1.11)
 Regularly 0.42 (0.26–0.67)*** 0.65 (0.50–0.85)**
Use of internet
 Never 1 1
 Yes 0.38 (0.11–1.34) 0.87 (0.60–1.26)
Constant −3.60 (−3.89 to −3.31) 0.017 (0.006–0.044)*** −1.35 (−1.63 to −1.08)*** 0.045 (0.017–0.117)***
Random effects
 Primary sampling unit 5.00 (4.01–6.23) 1.65 (1.23–2.20) 5.52 (4.47–6.80) 2.93 (2.33–3.69)
 Intracluster correlation coefficient (95% CI) 0.603 (0.549–0.654) 0.333 (0.272–0.401) 0.626 (0.576–0.674) 0.472 (0.415–0.529)
Log likelihood −2677.16*** −2114.39*** −2798.55*** −2408.92***
CI indicates confidence interval.

Measures of association (fixed effects)

Table 2 presents the results regarding the factors associated with homebirth among women in Benin and Mali. The probability of giving birth at home is higher among women with no education (aOR=2.52, 95% CI=1.59–4.00) and among women with primary education (aOR=2.08, 95% CI=1.27–3.43) compared with women having secondary education or above in Benin. The same trend is observed in Mali, and the aORs are 2.21 (95% CI=1.58–3.10) and 1.42 (95% CI=0.96–2.10), respectively, for women with no education and those with primary education.

In both countries, women experiencing distance problems to a health facility are more likely to give birth at home than those who do not face this problem. In Benin, women experiencing the problem of distance to a health center are 1.27 times (95% CI=1.05–1.54) more likely to give birth at home. This risk is 1.32 times (95% CI=1.10–1.59) for women facing the same problem in Mali.

Women residing in rural areas in Benin [aOR=1.62 (95% CI=1.14–2.30)] and Mali [aOR=3.20 (95% CI=1.97–5.22)] were more likely to give birth at home compared with those residing in urban areas. In both countries, the standard of living of the household is significant—women from poor households are more likely to give birth at home than women from better-off households.

In Benin and Mali, the odds of giving birth at home is strongly related to the number of antenatal visits a woman makes during her pregnancy. Not having made any ANC visits increases the odds of having a homebirth by 31.3 (aOR=31.3; 95% CI=24.10–40.70) in Benin and 12.91 (95% CI=10.21–16.33) in Mali. Similarly, women who went on 1–2 antenatal visits were more likely [aOR=3.07 (95% CI=2.41–3.91) in Benin and aOR=2.37 (95% CI=1.87–3.00) in Mali] to give birth at home than women who made 5 or more antenatal visits. Furthermore, the risk of giving birth at home was 1.4 times (95% CI=1.05–1.86) higher in Benin and 1.42 times (95% CI=1.14–1.78) higher in Mali among women who had made 3–4 antenatal visits than among those who had made at least 5 ANC visits.

In terms of the number of children, the results show that compared with women with 7 or more children, those with 1 or 2 children [aOR=0.59 (95% CI=0.41–0.84) in Benin and 0.58 (CI=0.42–0.82) in Mali] are less likely to give birth at home.

In addition, women who often or regularly listened to the media were less likely to have homebirth compared with those who did not pay attention to the media [aOR=0.42 (95% CI=0.26–0.67) in Benin and aOR=0.65 (95% CI=0.50–0.85) in Mali for women who regularly followed the media].


This study aimed to determine the prevalence of home deliveries in Benin and Mali, 2 West African countries, and the factors associated with them. Data from the 2018 DHS conducted in Benin and Mali were used. We found a difference in the prevalence of home deliveries in both the countries with 13.7% and 32.9% in Benin and Mali, respectively. This difference between the 2 countries could be explained by the availability and geographical coverage of health facilities and by the differences in health policies in place for maternal and child health. Benin in 200928 and Mali in 200529 developed and implemented a free cesarean section policy to increase access to emergency obstetric care for women to reduce maternal morbidity and mortality. Although this policy has improved access to care for pregnant women, it has not succeeded in eliminating homebirths in the countries, and the problems with women’s access to obstetric care still exist, especially in Mali30.

The results of our study showed that distance to the health center is an important factor influencing the place of delivery; the risk of giving birth at home was higher among women in both countries who considered the distance to the health facility to be a problem. This finding is consistent with the results of several previous studies in Africa5,9,10,12,31–34, Peru35, and India36. The distance to health services certainly has a dual influence on their use: it is used as a reason for not seeking care in the first place and is a real barrier to accessing care37. Many pregnant women do not even try to reach a facility for delivery because it is difficult to walk several kilometres during labor and impossible if labor begins at night when transport is often not available. The barrier effect of the distance is the strongest when combined with the lack of transport and poor road conditions38.

The results of the study showed that ANC visits are associated with homebirths in a statistically significant manner. Women who did not make ANC visits were more likely to give birth at home; the more ANC visits a woman makes during her pregnancy, the less likely she is to give birth outside a health facility. This result confirms those found by Gebremichael and Fenta32 in 9 sub-Saharan countries, Kimario et al39 in Tanzania, Sangho et al40 in Mali, and Paraiso et al41 in Benin. The significant effect of ANC underscores the role that pregnancy care plays in informing women of the benefits of institutional delivery and linking them to appropriate services42. ANC visits has been shown to have a positive impact on the quality of life of women, and it is indeed the most favorable point of contact for mothers to obtain more information about the risks and problems they may encounter during childbirth43.

Regarding individual characteristics, the results indicated that a woman’s education level, marital status, and the number of children were statistically significant in their association with homebirth. Women with no education, those who were married, and those with 7 or more children were more likely to give birth at home. This finding is also consistent with previous studies32,39,42–44. The communities demonstrating high fertility may be more conservative in their attitudes toward using the service and the expected roles for women and may have lower levels of economic development, which influence a woman’s ability to seek care during labor. High fertility may also reflect a lack of reproductive health services or a lack of awareness of these services if available, both of which have clear implications for the use of maternal health services45.

Concerning the place of residence, living in a rural area was associated with a higher proportion of homebirths in both Mali and Benin. This can be attributed mainly to poorer geographical access and transport difficulties in reaching health facilities, which generally also lack adequate infrastructure for care35.

The standard of living of households was also found to be a factor influencing the place of delivery; women from poor households are more likely to give birth at home. Affordability mainly influences whether the woman goes to a facility or not46. Therefore, policies that waive the delivery fees and provide free delivery are likely to increase women’s access to health services. In Burkina Faso, the findings of Ben Ameur et al47 show that eliminating fees for facility-based deliveries benefited the poorest households and reduced inequalities of access between the poor and the rich.

The present study found that nonexposure to the media was associated with a greater likelihood of having a homebirth in both the countries studied. This result is consistent with other studies conducted elsewhere in the world48–51. The effect could be explained by the fact that most media programs repeatedly promote institutional delivery, which may influence mothers to develop a positive behavior toward institutional delivery43. Also, in India, Sinha and Chattopadhyay48 have shown that any exposure to mass media significantly increased the likelihood of receiving full ANC visits and institutional delivery in both regions. However, as found in Ethiopia, limited electricity distribution in rural areas and low levels of education may combine to reduce media exposure, denying women the benefits of media-facilitated health promotions43.

The results of the multilevel logistic regression analysis showed that the contextual effect was significant, demonstrating that community factors influence home birth. Previous research has shown the effect of contextual and community factors on the choice of birthplace. Thus, community factors, geographical factors, the availability and physical access to health services, and road conditions are the mechanisms through which community effects can manifest themselves31. The community characteristics that had a strong direct positive influence on women’s decision to seek maternity care included the percentage of women in a community who had given birth in a health facility, high standard of living in urban neighborhoods, and the presence of a health worker in the community providing antenatal care (ANC)42,45. The constraints on the use of maternal health services have been associated with poor road conditions, a high average number of children per woman in the community, and living far from medical care facilities31. Furthermore, community beliefs and norms are reflected in an individual’s healthcare decisions, as individual behavior is influenced by how they believe the community judges their actions12. Community-level characteristics represent a unique social context that not only affects how individuals perceive and respond to health or other problems in the social environment but also exerts independent effects on the health outcomes of individuals in the community34.


Delivery outside of a health facility and not assisted by skilled health personnel may expose women to a higher risk of maternal and infant morbidity and mortality. However, more than 30% of the women in Mali and over 15% of the women in Benin give birth at home, and the factors associated with homebirth are the low levels of education, residence in rural areas, lack of antenatal visits during pregnancy, distance to the health center, and individual characteristics such as marital status and the number of children. Nonexposure to the media also increases the risk of giving birth at home. Further, contextual factors related to the village residence (or area of residence otherwise) are also important in understanding why some women give birth at home.

Increasing the demand for and access to women’s health services (prenatal consultations, assisted deliveries, etc.) by improving the availability and quality of services and establishing community health centers could help significantly reduce the risk of homebirths and, thus, help the fight against maternal and infant mortality. This could be achieved by bringing care closer to women through the development of community health care systems that duly on-board local actors, cultural values, endogenous resources, and initiatives developed and implemented by the community.

Conflict of interest disclosures

The authors declare that they have no financial conflict of interest with regard to the content of this report.


The authors thank the MEASURE DHS project for their support and for free access to the original data.


1. Wondimu MS, Woldesemayat EM. Determinants of home delivery among women in rural pastoralist community of hamar district, southern ethiopia: a case-control study. Risk Manag Healthc Policy 2020;13:2159–67.
2. WHO UNFPA, World Bank Group and the United Nations Population Division. Maternal Mortality: Levels and Trends 2000 to 2017. Geneva: World Health Organization; 2019.
3. United Nations General Assembly. Transforming Our World: The 2030 Agenda for Sustainable Development. New York, NY: 2015. Available at: https://www.un.org/ga/search/view_doc.asp?symbol=A/RES/70/1&Lang=E.
4. Callister LC, Edwards JE. Sustainable development goals and the ongoing process of reducing maternal mortality. J Obstet Gynecol Neonatal Nurs 2017;46:e56–e64.
5. Shimazaki A, Honda S, Dulnuan MM, et al. Factors associated with facility-based delivery in Mayoyao, Ifugao Province, Philippines. Asia Pac Fam Med 2013;12:5.
6. Mitiku AA, Dimore AL, Mogas SB. Determinants of home delivery among Mothers in Abobo District, Gambella Region, Ethiopia: a case control study. Int J Reprod Med 2020;2020:88565760.
7. United Nations Economic Commission for Africa. Assessing Progress in Africa Towards the Millennium Development Goals. Addis Ababa, Ethiopia: Economic Commission for Africa; 2014.
8. Esena RK, Sappor M-M. Factors associated with the utilization of skilled delivery services in the Ga east municipality of Ghana part 2: barriers to skilled delivery. Int J Sci Tech Res 2013;2:195–207.
9. Tuladhar H. Determinants of home delivery in a semi urban setting of Nepal. Nepal J Obstet Gynaecol 2009;4:30–37.
10. Mekonnen ZA, Lerebo WT, Gebrehiwot TG, et al. Multilevel analysis of individual and community level factors associated with institutional delivery in Ethiopia. BMC Res Notes 2015;8:376.
11. WHO. Skilled birth attendants. 2021. Available at: https://www.who.int/reproductivehealth/topics/mdgs/skilled_birth_attendant/en/. Accessed September 14, 2021.
12. Kifle MM, Kesete HF, Gaim HT, et al. Health facility or home delivery? Factors influencing the choice of delivery place among mothers living in rural communities of Eritrea. J Heal Popul Nutr 2018;37:22.
13. Ayamolowo LB, Odetola TD, Ayamolowo SJ. Determinants of choice of birth place among women in rural communities of southwestern Nigeria. Int J Africa Nurs Sci 2020;13:100244.
14. Ridde V, Morestin F. A scoping review of the literature on the abolition of user fees in health care services in Africa. Health Policy Plan 2011;26:1–11.
15. Lange IL, Kanhonou L, Goufodji S, et al. The costs of “free”: experiences of facility-based childbirth after Benin’s caesarean section exemption policy. Soc Sci Med 2016;168:53–62.
16. National Population Commission (NPC) [Nigeria] and ICF. Nigeria Demographic Health Survey 2018. Abuja, Nigeria, and Rockville, MD: 2019.
17. Institut National de la Statistique (INS) et ICF. Enquête Démographique et de Santé En Guinée 2018. Conakry, Guinée, et Rockville, MD: 2018.
18. Institut National de la Statistique (INSTAT) et ICF. Mali Demographic and Health Survey 2018. Bamako, Mali et Rockville, MD: 2019.
19. Statistics Sierra Leone (Stats SL) and ICF. Sierra Leone Demographic and Health Survey 2019. Freetown, Sierra Leone, and Rockville, MD: 2019. Available at: https://www.dhsprogram.com/ pubs/pdf/FR365/FR365.pdf.
20. Institut National de la Statistique et de l’Analyse Économique (INSAE) et ICF. Enquête Démographique et de Santé Au Bénin, 2017-2018. Cotonou, Bénin et Rockville, MD: 2019. Available at: http:// dhsprogram.com/pubs/pdf/FR350/FR350.pdf.
21. Titaley CR, Dibley MJ, Roberts CL. Type of delivery attendant, place of delivery and risk of early neonatal mortality: analyses of the 1994-2007 Indonesia Demographic and Health Surveys. Health Policy Plan 2012;27:405–16.
22. Ministère de la Santé du Bénin. Plan Directeur National de Lutte Integree Contre Les Maladies Tropicales Négligées 2016–2020. 2016. Available at: https://www.ntdenvision.org/sites/default/files/docs/benin_plan_directeur_mtn_2016-2020_version_finale_25sept2017.pdf.
23. Institut National De La Statistique du Mali. Rapport de l’enquete Modulaire Eet Permanente Aupres Des Ménages 2016. 2017. Available at: http://www.instat-mali.org/laravel-filemanager/files/shares/eq/rana16pas1_eq.pdf.
24. Ministère de la Santé du Bénin. Politique Nationale de Sante de La Reproduction. 2008. Available at: http://familyplanning-drc.net/docs/5_PolitiqueNationaleDeSanteDeLaReproduction_Juillet_2008.pdf. Accessed September 14, 2021.
25. Salami L, Ouendo EM, Fayomi B. Qualité du système d’information et de suivi des interventions en santé dans les zones exposées au financement basé sur les résultats en 2014 au Bénin. Pan Afr Med J 2017;28:1–11.
26. Juillet A, Touré B, Ouedraogo H, et al. Évaluation Du Système de Santé Au Mali. 2017. Available at: https://www.hfgproject.org/evaluation-du-systeme-de-sante-au-mali/. Accessed September 14, 2021.
27. Merlo J, Chaix B, Ohlsson H, et al. A brief conceptual tutorial of multilevel analysis in social epidemiology: using measures of clustering in multilevel logistic regression to investigate contextual phenomena. J Epidemiol Community Health 2006;60:290–7.
28. Mongbo V, Ouendo EM, De Brouwere V, et al. La césarienne de qualité : étude transversale dans 12 hôpitaux au Bénin. Rev Epidemiol Sante Publique 2016;64:281–93.
29. De Sardan JPO, Ridde V. L’exemption de paiement des soins au Burkina Faso, Mali et Niger: Les contradictions des politiques publiques. Afr Contemp 2012;243:11–32.
30. Fournier P, Dumont A, Tourigny C, et al. Improved access to comprehensive emergency obstetric care and its effect on institutional maternal mortality in rural Mali. Bull World Health Organ 2009;87:30–38.
31. Gage AJ. Barriers to the utilization of maternal health care in rural Mali. Soc Sci Med 2007;65:1666–82.
32. Gebremichael SG, Fenta SM. Determinants of institutional delivery in Sub-Saharan Africa: findings from Demographic and Health Survey (2013–2017) from nine countries. Trop Med Health 2021;49:45.
33. Mwewa D, Michelo C. Factors associated with home deliveries in a low income rural setting-observations from Nchelenge district, Zambia. Med J Zambia 2010;37:234–9.
34. Solanke BL, Rahman SA. Multilevel analysis of factors associated with assistance during delivery in rural Nigeria: implications for reducing rural-urban inequity in skilled care at delivery. BMC Pregnancy Childbirth 2018;18:1–15.
35. Hernández-Vásquez A, Chacón-Torrico H, Vargas-Fernández R, et al. Factors associated with home births in Peru 2015–2017: a cross-sectional population-based study. Heliyon 2021;7:e06344.
36. Kaul S, You W, Boyle KJ. Delivery at home versus delivery at a health care facility–a case study of Bihar, India. 2012. doi:10.22004/ag.econ.124625
37. Thaddeus S, Maine D. Too far to walk: maternal mortality in context. Soc Sci Med 1994;38:1091–110.
38. Okeshola FB, Sadiq IT. Determinants of home delivery among Hausa in Kaduna South Local Government Area of Kaduna State. Nigeria Am Int J Contemp Res 2015;3:78–85.
39. Kimario FF, Festo C, Shabani J, et al. Determinants of home delivery among women aged 15-24 years in Tanzania. Int J Matern Child Heal AIDS 2020;9:191–9.
40. Sabgho O, Traoré SO, Kamate YD, et al. Determinants of home delivery in two neighborhoods in Commune V of Bamako. Mali Med 2020;XXXV:63–69.
41. Paraiso N, Aguet VD, Nkwidjan H, et al. Accouchements à domicile dans la zone sanitaire de Pobè-Adja-Ouèrè-Kétou en 2011, Bénin. Ve Congrès International d’Épidémiologie Adelf-Epiter/Revue d’Épidémiologie et de Santé Publique. 2012; 60:S136.
42. Stephenson R, Baschieri A, Clements S, et al. Contextual influences on the use of health facilities for childbirth in Africa. Am J Public Health 2006;96:84–93.
43. Yebyo H, Alemayehu M, Kahsay A. Why do women deliver at home? Multilevel modeling of Ethiopian national demographic and health survey data. PLoS One 2015;10:1–14.
44. Kaba M, Bulto T, Tafesse Z, et al. Sociocultural determinants of home delivery in Ethiopia: a qualitative study. Int J Womens Health 2016;8:93–102.
45. Sagna ML, Sunil TS. Effects of individual and neighborhood factors on maternal care in Cambodia. Heal Place 2012;18:415–23.
46. Gabrysch S, Campbell OMR. Still too far to walk: literature review of the determinants of delivery service use. BMC Pregnancy Childbirth 2009;9:34.
47. Ben Ameur A, Ridde V, Bado AR, et al. User fee exemptions and excessive household spending for normal delivery in Burkina Faso: the need for careful implementation. BMC Health Serv Res 2012;12:412.
48. Sinha A, Chattopadhyay A. Utilization of maternal and child health care services in North and South India: does spousal violence matter? Int J Popul Stud 2017;2:107–22.
49. Begum HA, Sayem AM, Nili NY. Differentials in place of delivery and delivery assistance in urban slum areas, Bangladesh. J Fam Reprod Heal 2012;6:49–58.
50. Carroli G, Rooney C, Villar J. How effective is antenatal care in preventing maternal mortality and serious morbidity? An overview of the evidence. Paediatr Perinat Epidemiol 2001;15:1–42.
51. Alemayehu M, Mekonnen W. The prevalence of skilled birth attendant utilization and its correlates in North West Ethiopia. Biomed Res Int 2015;2015:436938.

Homebirth; Associated factors; Mali; Benin; Sub-Saharan Africa

Copyright © 2022 The Authors. Published by Wolters Kluwer on behalf of the International Federation of Fertility Societies.