Background
Reducing maternal mortality is a critical global priority. Efforts have resulted in a 47 % decline in global maternal deaths since 1990. Yet, globally an estimated 287,000 maternal deaths continue to occur every year [
1]. In addition, 300 million women in the world suffer from long-term or short-term illness brought on by pregnancy or childbirth [
2]. One of two indicators used to track progress towards achieving MDG 5 is the proportion of births attended by skilled health personnel [
3]. Global policies also aim at shifting the place of delivery from home to hospital as one of the strategies to improve maternal and neonatal survival [
4,
5]. This has led to unprecedented increases in facility births in several countries [
6,
7].
India is one of the countries to have adopted a conditional cash transfer scheme to encourage facility deliveries, being implemented since 2005. Facility deliveries in India have since expanded from 40.7 % in 2005–06 to 72.9 % in 2009–10 [
8]. There is some research evidence associating facility deliveries with a decline in perinatal mortality and still births in India [
9]. It has also led to an increase in institutional delivery by at-risk mothers, which has the potential to reduce maternal morbidity and mortality [
10]. Expanding health system effectiveness and socio-economic development, along with rising institutional deliveries have brought down maternal mortality ratio (MMR) over the years from 390 in 2000 to 179 in 2012 [
2,
11]. However, India still accounts for 19 % of global maternal deaths and is still far from achieving the fifth MDG target of reducing maternal mortality to 100 [
2].
One of the reasons for persistently low utilization of facility deliveries in developing countries is women’s preference for at- home deliveries. This preference is attributed to the satisfaction derived from home delivery in comparison with facility delivery based on previous experience, or perceptions surrounding these services [
12‐
14]. Services of traditional birth attendants continue to be widely used as community members believe health services are necessary only when there are complications [
15]. On the other hand, there have been reports of unkind treatment by health facility staff [
12,
16] poor attitude and absenteeism [
14,
17] and overall perceived poor quality of health services, [
14,
17‐
20] as deterrents to seeking facility delivery. A study showed that women with complications would delay or avoid seeking care in a facility if they had a previous experience of disrespectful treatment by staff [
14]. Other factors related to the level of satisfaction with, or preference for home over facility deliveries in developing countries include the distance of health facilities, costs, and quicker access to traditional birth attendants [
13‐
17,
21,
22]. Good infrastructure is also an important determinant, especially in facilities. Proper waiting areas, beds, housekeeping services, electricity, water and clean toilets are basic structural requirements which affect women’s satisfaction with services [
23‐
25]. Availability and quality of human resources, medicines, and equipment are also important [
26‐
28]. Prompt care, constant attention, perceived ‘good’ care, sharing of information on her condition with the woman and comforting her are other themes that determine maternal satisfaction [
23,
27,
29]. Outcome ultimately determines overall maternal perception. Safety and health of mother and child after delivery significantly influence a positive assessment of care by the mother [
24,
28,
30].
Few studies in India have explored women’s preferences for home, public or private health facility for delivery [
20]. Some studies have explored determinants of utilization of facility delivery [
19,
31,
32]. In India, traditionally medical attention is deemed unnecessary as delivery is believed to be a natural process [
12,
13,
19] and an unassisted birth is considered a sign of courage [
13]. Sometimes, economic status may influence decisions regarding place of delivery, even more than access, especially when choosing between private and public healthcare. Utilization of private health services is seen as an index of wealth and status [
19]. Public health facilities are the main source of affordable facility deliveries for India’s poor, and a good patient experience would be a critical factor for enhancing facility deliveries [
17]. There are, however, instances where women expressed a desire for facility delivery but still ended up delivering at home. The most common reason for this has been noted as the rapid progression of labor with no time to reach a facility, thereby leading to home deliveries [
16,
17]. Other factors inhibiting facility delivery include resistance by older women relatives, and sometimes women being sent back home from the facility as delivery was not due that day [
16]. In spite of the conditional cash incentive scheme introduced in India, there are states like Jharkhand where 60 % of deliveries continue to take place at home [
8].
In this context, it is important to understand the experience of care of women who have delivered at home vis-a-vis those who’ve delivered at a health facility, analyzing any differences so that services can be improved, with the goal to promote facility births, particularly in areas where women still prefer to home births. This study aims to understand women’s experiences of delivery care for home and facility births and the factors that influence women’s decisions regarding their next place of delivery.
Method
Study site
This study was conducted in one district in the state of Jharkhand in eastern India. Jharkhand is among the least developed states of India. More than half of households in Jharkhand (52 %) fall in the lowest wealth quintile. Age at marriage is low in the state, as about 63 % of women aged 20–24 years got married before the age of 18 [
33]. Jharkhand has an MMR of 219, higher than the national average of 178 [
34]. In terms of maternal health indicators, the state shows a dismal picture as only 11 % of women in the state who delivered during the 12 months preceding the survey received full antenatal care and only 40 % deliveries in the state are conducted in health facilities, which is also below the national average of 73 % [
8].
Study design and survey instrument
A cross-sectional community-based study design was selected. The data was collected in a six-week period in April and May 2012. The women were selected according to the following criteria: (i) women who had normal deliveries and (ii) women who delivered in a public health facility or at home with the support of a Traditional Birth Attendant (TBA). Women who had complicated deliveries, whether assisted or C-section, were excluded to obtain a homogeneous sample as women with complicated deliveries would have had a different experience of care. In order to reduce recall errors as far as possible, women who had delivered within 42 days before data collection were interviewed retrospectively at the community.
A structured questionnaire with close-ended questions was developed to capture the experience of care and satisfaction with maternal health services (Addition file
1). It was based on the determinants of maternal care identified from a literature review and qualitative study conducted before the survey [
35,
36]. The process helped in identifying determinants of maternal care across infrastructure, the process of care and outcomes [items listed in Table
2]. Among structural elements, good physical environment, cleanliness, availability of adequate human resources, medicines, and supplies emerged as key determinants. Determinants of the process of care included respectful behavior, privacy, promptness, cognitive care, perceived provider competency, and emotional support. Outcome related determinants included health status of the mother and newborn. Access and cost were also included. Covariates captured contextual factors like a woman’s socio-economic status (age, religion, caste, family type, and household income), reproductive history (parity, place of last delivery) and her experience of antenatal and postnatal care. Evidence from the review showed that maternal characteristics like age and parity affected women’s perceived experience of care [
37‐
39]. Ethnicity and literacy were also significant predictors of maternal satisfaction in developing countries [
27,
39,
40].
Along with capturing women’s experiences with care, they were further asked whether they were willing to deliver at the same place (facility or home) in the future (for their next delivery). The questionnaire was translated into the local language (Hindi and Bengali) and back-translated to check the linguistic accuracy. The translated questionnaire was reviewed by a panel of national and regional experts in maternal health to validate the content and contextual relevance. The questionnaire was also pretested in a similar population setting to confirm the content, ease of administration and comprehension by local women.
Sampling and sample size
A list of all women who had given birth in the study district between 15th December 2011 and 31st March 2012 was generated, which included 2377 women. The list comprising the place of birth (home or facility birth), date of delivery, respondent’s address was obtained from the immunization tracking register maintained by the Auxiliary Nurse Midwife (ANM). This register forms part of the routine report for the Health Management Information System (HMIS) under the National Rural Health Mission (NRHM), Government of India. To cross check the completeness of the register, a random check of 5 % of the data was done in the community by contacting the ASHA (community-level health worker) who collects the information for the immunization register. The sample size calculated for the study was 535 with 80 % power and 95 % confidence interval. The sample was further stratified by place of delivery, based on the proportion of women who delivered at a health facility [
8]. Women in the final sample were selected randomly with 40 % of women having delivered at a health facility and the rest having home births. The randomization was conducted using Microsoft Excel software. Selected women were approached for interviews at their homes. Community health workers of the respective villages helped trace the selected women. There were 20 refusals and 15 women could not be traced during the survey so a final sample of 500 women (210 women who had delivered at a health facility and 290 who had delivered at home) gave their informed consent to participate in the study.
Data collection
The instrument was translated into Hindi and Bengali, the local languages spoken in the study area and was pretested among a similar population. An individual who had good knowledge of the local language translated the English version into the local language for better comprehension. It was also back-translated to check for its original meaning. Ten women data collectors with a social science background and three supervisors were trained for 5 days on the study instrument and data collection procedure. At the time of the interview, an informed consent form was read and explained to the respondents. As more than half of the respondents in the study area were illiterate and the literates had limited writing skills with just four mean years of schooling, verbal consent was taken from the participants. Response to informed consent was recorded in the participant information sheet appended to the questionnaire. Confidentiality was assured and maintained during the data analysis.
The study received ethical approval from Public Health Foundation of India’s Institutional Ethics committee (TRC-IEC 112/11).
Data analysis
The software Epi Info Version 7 was used for data entry. Double data entry was conducted. The data was analyzed using PASW Statistics 17 (formerly known as SPSS). Women’s experience of care during delivery was explored by using cross-tabulation and chi-square tests. To understand women’s preference for future delivery place multiple logistic regression analysis was done taking into factors like ethnicity, socio-economic status and reproductive history of women.
Discussion
The study findings show that there is no major difference in the experience of care between home and facility births, the only difference in care being with regard to pain relief through massage/medicine and the low cost of delivery for those having home births. Findings show a growing preference for facility delivery among women in the study area. The major reasons for this preference were the availability of good supplies i.e. medicine and the perceived health benefits for mother and baby.
In South and Southeast Asia, even recently, more than 70 % of all births in the lowest two wealth quintiles occurred at home [
13]. Women consider the Traditional Birth Attendant (TBA) as a culturally acceptable and competent health worker [
41]. Residing in the same community, she offers respectful, prompt care and personalized attention to the woman during delivery [
36]. Poor accessibility of the institution and factors like lower maternal education, being a rural resident and multi-parity increase the likelihood of home delivery [
13,
14,
17,
21,
42]. On the other hand, evidence from developed countries shows that convenience, privacy, and respect are also very important determinants of a woman’s preference for home deliveries [
12,
16,
41]. However, the findings of this study show that there is no major difference in the experience of care between facility and home births, in terms of promptness and respectful care. For process of care indicators like time taken by provider to attend after onset of labor, women not being left alone in delivery room, sharing of information about progression of labor, delivery conducted in a secluded place and absence of male members, the experience of care was same for women who have delivered at home and at the facility.
This study’s findings show that a significant proportion of women, who delivered either at a facility (74.4 %) or at home (63 %), were inclined to deliver their next child at the health facility. One of the main reasons for their preference is that they perceive better availability of medicines and supplies, and improved health outcomes of both mother and newborn at facilities. There are other studies from India which have shown that safety of mother and child was the prime concern for households opting for institutional delivery, along with monetary incentives under the conditional cash transfer scheme, and motivation by community health workers [
43]. Utilization of public facility is also related to socio-cultural factors like religion, caste, age, education and economic condition [
43,
44].
Findings for unadjusted odds ratios in this study show that women with higher education level, belonging to higher age group, who have previously had facility delivery, and those with health problems post-delivery were more likely to give birth at a health facility in the future. Unadjusted figures show those who have received antenatal checkup, belonging to higher income level, and living in joint family, are also more likely to deliver their next child at a health facility. Other studies in similar settings have also shown that the number of ANC visits is an important factor leading towards delivery at an institution, as during ANC visits it is most likely that the women receive counseling from providers about the importance of safe delivery and are encouraged to deliver at a health facility [
45‐
47]. Controlling for other variables, this study found that the mother’s education, age, and current place of delivery are the factors that strongly predict their decision on the place of future delivery. This finding is consistent with earlier studies, which have shown educational status of mothers to be one of the most significant associated factors for utilization of institutional delivery [
41‐
43,
46,
48‐
52]. Our study also showed that mothers whose last delivery was at a facility and who had a good experience of care were more likely to go again for a facility delivery. This highlights the need for sustaining the quality of technical as well as interpersonal care in institutions [
43]. Particularly, women’s perspective needs to be taken into account in facility quality improvement initiatives to make services more responsive to their needs. Women’s preference towards facility delivery is a significant positive aspect and measures need to be taken to address their preferences so that there is improved utilization of services.
Limitations
The multivariate analysis includes selected socio-economic and demographic correlates, while other programmatic, institutional; accessibility and cultural factors could also affect delivery place preference. [
49]. Although special attention was paid to retrieve the lists of recently delivered women from health workers, there is a possibility some women might have been missed out, especially those who delivered at home. The responses could suffer from recall bias. The study did not consider women who had a complication during pregnancy and pertains to only normal deliveries. This was done to have a homogeneous sample as women in the above-mentioned category would have had a different experience of care. The study did not explore all dimensions of respectful care and limited its inquiry to the experience of any abuse in the form of shouting, using abusive language, delivery in a secluded place and not being left alone in the delivery room.
Competing interests
The authors declare that they have no competing interests.
Authors’ contribution
SB, BIA and AS conceptualized the study and were involved in data collection. RR assisted in data analysis. BIA provided technical guidance in development of tools and analysis of data. SB prepared the initial draft and AS, RR and BIA revised and finalized the manuscript. All authors read and approved the final manuscript.