Study design

This study deployed quantitative research design and utilized secondary data obtained from the Nepal Demographic Health Survey [NDHS] 2022. The NDHS is a nationally representative survey that provides periodically updates about demographic and health characteristics of the people of Nepal [23]. The survey was carried out by New ERA under the supervision of the Ministry of Health and Population (MoHP), was technically assisted by UCF International [13]. Details of the methodology can be obtained from the NDHS 2022 report [13].

Study setting

The study setting of NDHS 2022 was carried out with the territory of Nepal. Since 2015, the administrative structure of Nepal has been organized into three levels of government: the federal government, seven provincial governments, and 753 local governments. There are 6 metropolitan cities, 11 sub-metropolitan cities, 276 urban municipalities, and 460 rural municipalities under local administration [24].

Sample and sampling

The sampling frame for the 2022 NDHS is an updated version of the 2011 Nepal Population and Housing Census (NPHC) provided by the National Statistical Office [13]. The NDHS 2022 survey included samples from all seven provinces and three ecological regions of Nepal, representing national representativeness. Before sample selection, NDHS 2022 involved stratifying all seven provinces into urban and rural areas, resulting in 14 sampling strata from each province. The NDHS 2022 utilized two stages of a stratified sampling approach. Using the probability-proportional-to-size method, 476 primary sampling units (PSUs) were chosen in the first stage, with 248 from urban areas and 228 from rural areas. 30 households were chosen from each PSU in the second stage, resulting in a total sample size of 14,280 households, with 7,440 from urban areas and 6,840 from rural areas. Out of the surveyed households, 15,238 women aged 15–49 were eligible women for the NDHS survey. Of these 14,845 eligible women were successfully interviewed, yielding a response rate of 97.4 percent. In order to address the objectives of the study, we analyzed the 1977 eligible women aged 15–49 with a live birth in the two years preceding the survey.

Data collection and processing

The data collection for the survey was conducted from January 5, 2022 to June 22, 2022. After the finalization of all questionnaires in English, they were translated in the language: Nepali, Maithili, and Bhojpuri. Four types of survey tools: the Household Questionnaire, the Woman’s Questionnaire, the Man’s Questionnaire, and the Biomarker Questionnaire were utilized in the NDHS 2022. To facilitate computer-assisted personal interviewing (CAPI), the system was programmed into tablet computers with the capability to choose from any of three languages. Tablet computers were used to gather data by the enumerators. Nepal Health Research Council (NHRC) and the ICF Institutional Review Board evaluated the survey protocol. Nineteen teams were employed for data collection. One supervisor, one male interviewer, one biomarker specialist, and three female interviewers were included in each team. After the completion of data collection in each cluster, the data files were registered and examined for incompleteness, inconsistencies, and outliers. Inconsistencies and errors were promptly reported to the field team for review in order to minimize issues moving forward.

Outcome variable

The outcome variable of this study was the institutional delivery from the category of place of delivery. The variable “place of delivery” was recoded into a binary variable (yes/no) to represent institutional place of delivery for the calculation of binary logistic regression. Institutional delivery involved government hospital, PHC/primary hospitals, health post, basic health care center, urban health center, public sector, community health unit, private hospital, private clinic, other private and medical sector, FPAN, Merie Stopes, and NGO other, which were merged into a category labeled “yes” for institutional place of delivery. Respondent’s home, other home, and other were merged into “no” for institutional place of delivery.

Independent variables

In this study, there were 14 independent variables which were categorized into 4 groups. The first group, residential factors, included place of residence, ecological zone, and province. The second group, socioeconomic factors, encompassed ethnic group, education level, education level of husband/partner, wealth index. The third group, comprised health service-related factors, included times to reach the nearest healthcare facility, visited health facility last 12 months, and frequency of ANC visits and lastly, bio-demographic factors included mother’s age in 5-year groups, duration of pregnancy, birth order, and pregnancy outcome. Among them, place of residence, ecological zone, province, sex of the household head, wealth index, visited health facility last 12 months, and pregnancy outcome independent variables were used as received, while the rest of independent variables were recoded into adjusted categories. These variables were adjusted to make the data more reader-friendly (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. List of independent variables with adjusted category.

https://doi.org/10.1371/journal.pone.0322309.t001

Statistical analysis

We utilized Stata 13.0 for data analysis [25]. Variables were presented in terms of frequency and percentage with respect to both total and institutional place of delivery separately. Complex sample analysis has been performed using Stata 13.0 with “svyset” and “svy” commands. This method accounts for the survey’s stratification, sampling weights, and clustering, ensuring accurate estimates and valid significance tests. Both univariate and multivariate binary logistic regression analyses were performed to find the relationship between institutional delivery and various factors (residential factors, socio-economic factors, health service-related factors, and bio-demographic factors). Descriptive analysis was performed to summarize the characteristics of the sample population. Before the binary logistic regression test, we assessed the variance inflation factors (VIF) for the multicollinearity test and found that there were no multicollinearity issues among the variables within each group of independent variables [26]. The results of binary logistic regression analysis were presented in terms of crude odds ratio (COR) and adjusted odds ratio (AOR) at a 95% of confidence interval (CI). The COR is calculated independently for each variable, assessing the individual association with the outcome variable. On the contrary, AOR is calculated by considering multiple variables simultaneously among the group of four factors mentioned factors separately, accounting for potential confounding effects.

Ethical approval

The DHS Program provided dataset access for this study. The survey protocol was reviewed and approved by the institutional review board of ICF and Nepal Health Research Council [13].

Residential factors

Women living in rural areas revealed less tendency towards institutional delivery compared to those residing in urban areas aligned with the study [10,27,28]. Women in rural areas of Nepal are less likely to opt for institutional delivery due to lack of adequate health facilities, lack of easy access of transportation, lack of educational awareness, perception of poor quality of care, inadequate infrastructure in health facilities, and lack of skilled health personnel [22,29,30]. The odds of likelihood of institutional delivery were found to be higher in the hill and terai region when compared to the mountain region. The likelihood of institutional delivery was found to be lower in Madhesh Province [10] than in Koshi Province and twice as high in Gandaki and Bagmati Provinces.

The uneven utilization of institutional delivery services across provinces may be associated with disparities in both the availability and usage of maternal health services within those provinces [10,31]. Women in Madhesh Province are less likely to choose institutional delivery, even though there is largely plain and flat area, providing physical access to health services. Despite this, women in Madhesh Province are attributed by poor socio-economic conditions and socio-cultural beliefs, considering institutional delivery unnecessary unless complications occur [22]. Women from areas with less developed areas are less likely to utilize institutional delivery services compared to those from relatively more developed areas [29,32]. Addressing the issues, implementation of targeted interventions is essential for institutional delivery services among women residing in mountain regions, with a specific focus on Madhesh Province. Investing in infrastructure reduces travel time to health facilities, making it easier for women to access delivery services, and can encourage institutional delivery among women residing in regions with less developed area compared to those in more developed areas [29,32].

Socio-economic factors

The findings of the study shed light on the involved interplay of socio-economic factors in influencing the choice of institutional delivery in Nepal. Firstly, analysis revealed notable disparities across ethnic groups. Muslim ethnic group exhibited the lowest prevalence of institutional delivery among other ethnic groups, probable causes such as lower levels of education and religious and cultural beliefs. While Brahmin/Chhetri ethnic groups demonstrated the highest rates compared to Janajati, Dalit and Madhesi ethnic groups. A similar study also reveals women from Brahmin/Chhetri ethnic group were observed to have a greater tendency to deliver their babies in health facilities [33,34]. Another study also yielded a similar result, indicating that women from less advantaged castes were less likely to deliver their babies in healthcare facilities [22]. Women from less advantaged ethnic groups are less tendency to utilize institutional delivery services compared to the relatively more advantaged ethnic groups. Ethnic differences were evident, with less likely among Muslim ethnic groups for institutional delivery, indicating the need for targeted interventions for Muslim ethnic groups.

Higher levels of education of women and her husband/partner also displayed a greater tendency towards institutional delivery compared to lowers level of education and people with no education. Some of studies also observed that there is a reduced likelihood of institutional delivery practice with a lower level of education among women [10,28,35] and husbands [33]. Similarly, educated women are more likely to deliver their babies in health facility compared to uneducated women [22,36]. So, education plays a crucial role in encouraging institutional delivery practices in Nepal. Educated women are more likely to possess the awareness, knowledge, and decision-making power necessary to prioritize institutional delivery. They understand the advantages of institutional delivery and are better equipped to navigate the healthcare system to access maternal healthcare services and contribute to reducing maternal and neonatal mortality rates in the country.

Women from the poorest wealth category revealed less likely to have institutional deliveries compared to the richest wealth category. A similar situation was also observed in the study, women from the poor wealth category were less likely to select institutional delivery in comparison to women from the richest wealth category [10,22,28], primarily due to the hidden costs associated with such deliveries (e.g., medicines, transportation) [37]. As a result, women from poorer households may notice institutional delivery as economically burdensome, leading them to choose alternative options such as home deliveries assisted by traditional birth attendants. This highlights the need to address financial obstacles and provide comprehensive support to ensure equitable access to institutional delivery services for all women, regardless of their socioeconomic status.

Health service-related factors

Women are more likely to have institutional delivery if it takes less time to reach the nearest health care facility compared to those women who require more time to reach health facilities. Similarly, women who visited health facilities in the last 12 months were more likely to opt for institutional delivery than those who did not visit health facilities in the last 12 months. The distance between the residence and the institutional delivery facility has been identified as a major hindrance for women [30] in accessing and utilizing maternal health services, impacting both antenatal care (ANC) visits and institutional delivery [38]. This barrier affects both antenatal care visits and the choice of institutional delivery. So, initiatives aimed at reducing transportation barriers, improving infrastructure, and expanding healthcare facilities closer to communities can be supported to mitigate this challenge and ensure that all women have equitable access to essential maternal healthcare services.

Antenatal care is recognized as a pathway to institutional delivery [39]. A higher number of ANC visits demonstrated a higher tendency towards institutional delivery compared to those who never attended ANC visits. This result is in line with a study conducted in Ethiopia, a higher number of ANC visits tended to increase the utilization of institutional delivery services [29]. Same finding was also found in another study in Ethiopia, the finding that ANC attendance is associated with institution delivery [40]. Moreover, women who attended ANC for four or more times were observed to have greater tendency to give birth in an institutional delivery compared to those who attended less than four times [41,42]. The findings consistently indicate that a higher number of ANC visits are associated with an increased likelihood of opting for institutional delivery services. ANC as an entry point to maternal healthcare, serving not only to monitor pregnancy but also to educate and counsel women about the benefits of institutional delivery. Regular attendance to antenatal care (ANC), especially aiming for four or more visits, plays a vital role in promoting institutional delivery and enhancing maternal and child health outcomes.

Bio-demographic factors

While examining bio-demographic factors, this study and similar research exhibited that mothers in higher age group were more likely to opt for institutional deliveries compared to mothers of lower age groups category [43,44]. Higher age category of mothers being more inclined towards institutional deliveries [10,28] due to awareness and concern about potential complications associated risks with advanced maternal age, prioritizing the safety and support provided by healthcare facilities during childbirth [45]. This study also assumes the same probable causes regarding the high age category of mothers.

The prevalence of institutional delivery among primiparous was notably higher [46], aligning with the findings of this study. Women with more parity were less likely to choose institutional delivery compared to those with fewer parity, despite the fact that risk for birth complications and maternal mortality is also high for women in higher parities [47]. It is also noted that the parity and institutional delivery are inversely correlated, which is consistent with present finding, i.e., women in first parity were more likely to deliver in a health facility compared to those in second or higher parity [48–50]. The inverse correlation between parity and institutional place of delivery is significantly influenced by previous childbirth experiences, perceptions of quality of care [6,8,9], and behavior of health care providers [7]. The reduced utilization of institutional delivery may be attributed to time and resource constraints faced by those managing larger families [39]. The study highlights a trend where primiparous women exhibit more likely to have institutional delivery, while higher-parity women show lower utilization, probably due to resource constraints.