Women’s health insurance and institutional delivery in Nepal: Results of the 2022 Nepal demographic Health Survey

Abstract

Background

The Government of Nepal introduced a Health Insurance (HI) program in 2017 to prevent citizens from falling into poverty due to health care costs. This study aimed to assess the association between HI coverage and institutional coverage and institutional delivery among women in Nepal.

Methods

This cross-sectional study used secondary data from the Nepal Demographic and Health Survey (NDHS) 2022, conducted every five years since 1996. A total of 1,056 women who had given birth in the past year were included in the analysis. Statistical significance was assessed at the p < 0.05 level. Data analysis was performed using IBM SPSS Statistics version 25.

Results

Women covered by HI were more likely to deliver at Health Facilities (HF) within the past year (cOR = 5.775; 95% confidence interval (CI): 2.347 to 14.212; P < 0.001). Women with secondary or higher education were more likely to give birth at an HF (aOR = 3.130; 95% CI: 1.744 to 5.618; P < 0.001). Women residing in Koshi and Madesh provinces were less likely to give birth at HF (aOR = 0.399; 95% CI: 0.177 to 0.901; P < 0.05) and (aOR = 0.198; 95% CI: 0.083 to 0.476; P < 0.001), respectively, compared to those in Bagmati province. Women in the poorer and middle wealth index categories were more likely to give birth at HF (aOR = 1.701; 95% CI: 1.021 to 2.833; P < 0.05) and (aOR = 2.597; 95% CI: 1.461 to 4.615; P < 0.01), respectively.

Conclusions

HI coverage, educational level, province, and wealth index were significantly associated with the delivery practice for their most recent child at HF. To increase the use of HF during delivery, targeted interventions addressing educational and economic disparities, particularly in provinces with lower utilization, are needed.

1. Introduction

In 2017, the Government of Nepal (GoN) established the Health Insurance Board (HIB) under the Health Insurance (HI) Act 2017 as an autonomous federal-level body responsible for managing the HI scheme in Nepal. ¹ As part of Nepal’s social security initiative, the National Health Insurance Program (NHIP) is a pivotal pillar in Nepal’s journey to shield individuals from unexpected healthcare expenses, thereby managing risks upfront within the framework of government-provided social health protection. Through this initiative, NHIP aimed to move closer to achieving Universal Health Coverage (UHC) by 2030, a vital aspect of its Sustainable Development Goals (SDGs). Nepal’s Constitution (2015) guarantees free basic health services and recognizes healthcare as a fundamental right; additionally, Article 51 ensures citizens’ access to HI, safeguarding the provision of quality healthcare. The HI program focuses on reducing out-of-pocket health care expenditures. ² The HI program aims to prevent citizens from falling into poverty due to healthcare costs, such as catastrophic expenditures resulting from accidents or diseases, by combining prepayments with risk pooling and mutual support. ³ The HI program was piloted in a single district, Kailali, in 2015. It was extended to all 77 districts across seven provinces and 746 of the total 753 local government levels in Nepal by the end of 2022. ⁴ The introduction of HI and its expansion throughout all local bodies have led to significant efforts towards social health protection, ensuring equitable and universal access to healthcare services for all citizens of the nation. As of 2022, the population coverage stood at 5,967,408 (22.5%) of the national population. ⁴

In Nepal, households’ out-of-pocket health expenditure alone contributes to 56.3% of current healthcare expenditure. ⁴ As in many other developing countries, Nepalese patients’ or their families’ out-of-pocket health expenditure constitutes a large proportion of the amount spent on healthcare; the proportion was estimated to be 30.8% in the Southeast Asia Region of the World Health Organization (WHO). ⁵ In countries where out-of-pocket expenditure is the primary source of healthcare financing, households can experience financial catastrophe and often impoverishment due to out-of-pocket healthcare costs. ⁶

Many Low and Middle-Income Countries (LMICs) have faced severe challenges in sustaining sufficient healthcare financing and providing adequate financial protection against the impoverishing effects of catastrophic illness. ⁷ Evidence from other LMICs, including the Philippines, Ghana, and Burkina Faso, suggests that HI can be a crucial tool in increasing the utilization of maternal health services, such as facility-based delivery.^8–10 Considering these issues, switching from out-of-pocket healthcare payments to prepayment social mechanisms has been widely advocated as an essential step toward reducing the risk of financial hardship. In this context, the WHO introduced prepayment mechanisms in the health sector to share risk and avoid catastrophic healthcare expenditure and the impoverishment of individuals in 2005. ¹¹ In 2013, HI was highlighted as one of the most promising approaches for subsidizing the entire population and achieving universal health coverage. ¹² Various countries around the world have adopted different health financing mechanisms, including voluntary community-based and social HI schemes. ²

In Nepal, few studies have focused on specific aspects of HI, and these mainly examine household structures at a small scale. Furthermore, these studies are limited to districts and even local bodies, which cannot represent the entire society. To address this gap in HI-related academic discourse, this study was conducted with a large, nationwide sample. It utilized a scientific and authentic data source, which may provide robust evidence for policy development nationwide. The overarching objective of the study was to assess people’s engagement in the HI program and the delivery practices for the most recent child in HF among women in Nepal.

In Nepal, despite 81% of births occurring in HF, effective coverage of care quality remains critically low: just 18.5% of facilities are adequately equipped for routine delivery, and roughly 12% can provide emergency obstetric care, while out-of-pocket expenditure continues to drive financial hardship for families. We therefore focus on facility-based delivery as a primary outcome because it serves as a tangible proxy for both financial protection and quality-adjusted access, allowing us to assess whether health insurance engagement meaningfully reduces cost barriers and improves the actual receipt of safe, skilled childbirth care that Nepal’s maternal health policies aim to guarantee. Additionally, while the HI program is designed to facilitate this access, its success is also contingent upon community uptake, which can be hindered by limited health literacy and a general lack of awareness among the population regarding their entitlements to these services. By examining these dual aspects of engagement and delivery practices, this study aims to identify critical barriers and enablers, providing evidence to refine policy implementation and bridge the gap between policy intent and practical outcomes on the ground.

2. Materials and methods

The study applied a cross-sectional survey design and used secondary data from the Nepal Demographic and Health Survey (NDHS) 2022. The NDHS 2022 was Nepal’s nationally representative survey, conducted every five years since 1996 by the Ministry of Health and Population (MoHP). The survey provides current data on basic demographic and health-related information. Although HI information has been implemented since 2016 by the HI Board, NDHS 2022 included HI-related information for the first time, making it the nationally representative source of HI data. ³

2.1. Sample and sampling procedure

All households in the country were considered the study’s population. The survey used an updated sampling frame from the 2011 census, creating 36,020 sub-wards based on rural and urban residential settings. A total of 14,280 households were selected from 476 primary sampling units (PSUs), with 30 chosen households per PSU. Of the selected households, 15,238 women aged 15–49 years were eligible for the survey; however, 14,845 were successfully interviewed from January 5 to June 22, 2022. Among them, only 1,056 women who had given birth within the past year were included as the sample for this study. Details of the sampling procedure can be found in the NDHS 2022 report, which is publicly available on the DHS Program website.

2.2. Inclusion and exclusion citeria

The study included all women aged 15-49 years who were successfully interviewed in the NDHS 2022. Women were included in the final sample if they had given birth in the year preceding the survey. Women who had not given birth in the last year were excluded from the analysis.

2.3 Covariates

Based on the NDHS 2022 data, the study used socio-demographic characteristics, including women covered by HI, age group, ethnicity, educational status, current residence with husband or partner, religion, province, place of residence, current employment status, combined wealth index, and sex of the household head as independent variables. The study further considered the delivery status of women at HF in the past year as the dependent or outcome variable.

2.4. Statistical analysis

Three types of analysis were used in the study: univariate, bivariate, and multivariate. In the initial phase, the study conducted univariate/descriptive (frequencies and percentages) analyses to describe the general characteristics of the respondents: women aged 15 to 49 years. In the second phase, the association between the independent and dependent/outcome variables was assessed using a bivariate (χ2 test) analysis. ¹³ In the bivariate analysis, p < 0.05 was considered statistically significant. ¹⁴ Given the exploratory nature of the analysis, neither forward nor backward stepwise selection was utilized; rather, the enter (forced entry) method was applied. Candidate variables were selected based on statistical significance from the bivariate analysis (p < 0.05) combined with conceptual relevance from prior literature. Finally, multivariate logistic regression was performed to predict the delivery status at HF in the last year. Before conducting the multivariate analysis, the covariates were examined to determine whether they were significantly associated. The study further ensured that there was no multicollinearity issue. Variables that were not multicollinear were then included in the multivariate analysis. ¹⁵ Data analysis was performed using IBM SPSS Statistics version 25.

2.5. Ethical consideration

The proposal was reviewed and approved by the ICF Institutional Review Board (IRB). The Nepal Health Research Council (NHRC) also reviewed and approved the survey proposal (Approval Number: NHRC 535/2021) and the ICF Institutional Review Board (IRB, Protocol No. 178679). Written informed consent was obtained from all participants. For participants under 18 years of age, written informed consent was obtained from their parents or legally authorized representatives. For participants with no formal education, the consent form was read aloud to them in their native language in the presence of an impractical witness. The participants then provided their mark as their consent, the witness signed to confirm this information was fully understood. Moreover, no personal identity was disclosed in the datasets. ³ This study is a secondary analysis of de-identified, publicly available data from the NDHS 2022. The data were accessed via the NDHS program website (https://dhsprogram.com) after completing the standard registration and approval process. As this analysis used only de-identified secondary data, no additional ethical approval was required. All analyses were conducted in accordance with the NDHS program data use policies.

3. Results

The section depicts a flow chart illustrating the selection of the study sample from the NDHS 2022. It further illustrates the delivery status of the recent child in Health Facilities (HF) by socio-economic characteristics. In addition, the section shows associated factors: covered by Health Insurance (HI), age group, ethnicity, education, currently residing with husband or partner, religion, province, place of residence, currently working status, wealth index (a composite measure of household living standards calculated using principal component analysis of household assets and housing characteristics, divided into five population quintiles from poorest to richest), and sex of household head with women’s delivery status at HF within the past year. It further predicts the delivery status at HF covered by HI after adjusting for all socio-demographic characteristics. The delivery location was categorized as either “Health Facilities (HF)” or “Home”, as per the NDHS classification. Nationally, 79.4% of women chose institutional delivery, while 18.4% chose home delivery. Home delivery includes births occurring at the woman’s own home, her parents’ home, or the home of a traditional birth attendant.

Figure 1 presents a flow chart illustrating the selection of the study sample from the NDHS 2022. From the total of 14,845 women interviewed in the NDHS 2022, the sample selection proceeded in two sequential stages. In the first stage, the sample was restricted to those who had delivered within the past year, excluding 13,789 women and resulting in 1,056 women. In the second stage, this subgroup was further restricted to individuals with complete data on place of delivery. No women were excluded due to missing data on place of delivery or lack of consent (n = 0). The final analytic sample, therefore, consisted of 1,056 women with complete data on all variables.

Figure 1.

Nepal demographic and health survey 2022, https://dhsprogram.com.

3.1. Background characteristics of the respondents

Table 1 shows that 1056 women who had given birth within the past year participated in the study. The majority (87.3%) had engaged in the HI program, and a few (12.7%) had not. More than half (52.2%) were under 25 years old, and the largest ethnic group was Janajati (31.9%), followed by Brahmin/Chhetri (26.5%). About half (49.4%) of women had a secondary or higher level of education, and the majority (61.2%) were residing with their husband or partner.

Table 1.

Background characteristics of women who have given birth within the past one year preceding the survey.

Variables		%	N
Covered by health insurance	No	87.3	922
	Yes	12.7	134
Age group	Less than 25 years	52.2	552
	25-34	43.2	456
	35 or above	4.6	48
Ethnicity	Janajati	31.9	337
	Brahmin/Chhetri	26.5	279
	Dalit	17.4	183
	Madhesi	17.1	180
	Muslim	7.2	76
Education	Secondary or above	49.4	522
	Basic	33.9	358
	No education	16.7	176
Currently residing with husband/partner	Husband staying elsewhere	38.8	408
	Husband living with respondents	61.2	643
Religion	Hindu	83.2	878
	Islam	6.8	72
	Others (Kirat/Christian/other)	5.2	55
	Buddhist	4.8	51
Province	Madhesh	24.4	258
	Koshi	19.5	206
	Lumbini	16.1	170
	Sudurpashchim	10.1	106
	Karnali	7.4	78
	Gandaki	6.6	70
Place of residence	Urban	65.2	688
	Rural	34.8	368
Currently working	No	62.8	663
	Yes	37.2	393
Wealth index combined	Poorest	21.2	224
	Poorer	21.8	230
	Middle	20.3	215
	Richer	19.8	209
	Richest	16.9	179
Sex of household head	Male	70.9	749
	Female	29.1	307
Total		100.0	1056

The overwhelming majority of women (83.2%) identified themselves as Hindu; the highest proportion of women resided in Madhesh province (24.4%), followed by Koshi (19.5%) and Lumbini (16.1%). Majorities (65.2%) resided in urban areas, and 62.8% of women were not currently working. Women were evenly distributed across combined wealth indexes, from poorest to richest. The majority (70.9%) of women revealed that men were the household heads of their families.

3.2. Association between socio-economic characteristics of the women and delivery practice at Health Facilities

Table 2 presents the association between socio-demographic factors such as women covered by HI, age group, ethnicity, education, province, place of residence, and wealth index, combined with delivery practice by women at HF in the last year, by bivariate analysis. Women engaged in the HI program had a significant association with delivery at HF (P<0.001), with 96.2% of women in the HI program and 81.5% of women not in it. The women’s ethnicity was significantly associated with delivery at HF in the last year (P<0.001), with Brahmin/Chhetri (91.6%), Janajati (86.9%), and Dalit (76.9%) being the most common.

Table 2.

Association between socio-economic characteristics of the women and delivery practice at health facilities.

Variables		Place of delivery for the most recent live birth in the last year preceding the survey		Total		Chi-square (P value)
		Home	Health facilities	%	N
		%	%
Covered by health insurance***	No	18.5	81.5	100.0	918	​
	Yes	3.8	96.2	100.0	134	18.4, p=000
Age group	Less than 25 years	17.7	82.3	100.0	548	​
	25-34	15.2	84.8	100.0	456	​
	35 or above	19.1	80.9	100.0	48	1.3, p=0.510
Ethnicity***	Brahmin/Chhetri	8.4	91.6	100.0	278	​
	Janajati	13.1	86.9	100.0	337	​
	Dalit	23.1	76.9	100.0	183	​
	Muslim	27.2	72.8	100.0	76	​
	Madhesi	25.2	74.8	100.0	179	38.4, p=0.000
Education***	No education	35.5	64.5	100.0	175	​
	Basic	22.4	77.6	100.0	358	​
	Secondary or above	6.4	93.6	100.0	520	92.5, p=0.000
Province***	Bagmati	5.4	94.6	100.0	168	​
	Koshi	17.8	82.2	100.0	206	​
	Madhesh	31.9	68.1	100.0	257	​
	Gandaki	8.5	91.5	100.0	69	​
	Lumbini	12.3	87.7	100.0	168	​
	Karnali	15.9	84.1	100.0	78	​
	Sudurpashchim	8.0	92.0	100.0	106	70.6, p=0.000
Place of residence	Urban	15.9	84.1	100.0	687	​
	Rural	18.2	81.8	100.0	366	0.84, p=0.358
Wealth index combined***	Poorest	25.7	74.3	100.0	223	​
	Poorer	23.8	76.2	100.0	230	​
	Middle	17.2	82.8	100.0	212	​
	Richer	11.5	88.5	100.0	209	​
	Richest	1.7	98.3	100.0	179	54.5, p=0.000
Total		16.7	83.3	100.0	1052	​

Note. Statistically significant at ***=p<0.001.

Women with secondary or higher education had a significantly higher association with delivery at the HF in the last year (P<0.001), with 93.6% of women with secondary or higher education giving birth at the HF, compared to 77.6% with basic education and 64.5% with no education. Provinces are significantly associated with delivery practices at HF (P<0.001); women residing in Bagmati (94.6%), Sudurpaschim (92.0%), and Gandaki (91.5%) provinces were delivered at HF at rates like those in other provinces. Additionally, women’s combined wealth index showed a significant association with delivery at HF (P<0.001), with 98.3% of the richest, 88.5% of the richer, and 82.8% of the middle surpassing the poorer (76.2%) and the poorest (74.3%).

3.3. Effect of health insurance on delivery at health facilities in the past year

Table 3 presents the results of the multivariate logistic regression examining factors associated with women’s facility-based delivery status in the last year. Two models are presented: Model I shows crude odds ratios (cOR) from unadjusted bivariate analyses, and Model II shows adjusted odds ratios (aOR) from multivariate analyses controlling for all socio-demographic variables (age, ethnicity, education, province, place of residence, and wealth index). Both Models included 1,056 observations. Only subsets of covariates were predicted to affect delivery at health facilities significantly. In Model I (unadjusted), women who engaged in HI were more likely to give birth at HF within the past year (cOR = 5.775; 95% CI: 2.347 to 14.212; P<0.001) than those who had not engaged in the HI program.

Table 3.

Multivariate logistic regression of delivery at health facilities in the last year by women.

Variables		Model I			Model II
		cOR	95% CI		aOR	95% CI
Covered by health insurance	No (Ref.)	1.00	​	​	1.00	​	​
	Yes	5.775 ***	2.347	14.212	2.027***	1.773	3.994
Age group	Less than 25 years (Ref.)	​	​	​	​	​	​
	25-34	​	​	​	0.832	0.562	1.230
	35 or above	​	​	​	0.830	0.349	1.974
Ethnicity	Brahmin/Chhetri (Ref.)	​	​	​	1.00	​	​
	Janajati	​	​	​	1.052	0.557	1.987
	Dalit	​	​	​	1.130	0.575	2.220
	Muslim	​	​	​	0.965	0.396	2.350
	Madhesi	​	​	​	0.957	0.439	2.086
Education	No education (Ref.)	​	​	​	1.00	​	​
	Basic	​	​	​	1.171	0.733	1.871
	Secondary or above	​	​	​	3.130***	1.744	5.618
Province	Bagmati (Ref.)	​	​	​	1.00	​	​
	Koshi	​	​	​	0.399*	0.177	0.901
	Madhesh	​	​	​	0.198***	0.083	0.476
	Gandaki	​	​	​	0.743	0.236	2.336
	Lumbini	​	​	​	0.614	0.258	1.462
	Karnali	​	​	​	0.635	0.226	1.790
	Sudurpashchim	​	​	​	1.299	0.453	3.727
Place of residence	Urban (Ref.)	​	​	​	1.00	​	​
	Rural	​	​	​	1.092	0.747	1.597
Wealth index combined	Poorest (Ref.)	​	​	​	1.00	​	​
	Poorer	​	​	​	1.701*	1.021	2.833
	Middle	​	​	​	2.597**	1.461	4.615
	Richer	​	​	​	3.335***	1.752	6.350
	Richest	​	​	​	14.688***	4.142	52.082
​	Constant	4.391***			3.27*
	-2 Log likelihood	923.8			778.2
	Cox & Snell R Square	0.023			0.149

Note. Statistically significant at ***=p<0.001, **p<0.01 and *=p<0.05.

In the adjusted model (Model II), the association between HI coverage and facility delivery remained statistically significant (aOR = 2.027; 95% CI: 1.773 to 3.994; p<0.001), though the effect was attenuated compared to the unadjusted model. Women with secondary or above education levels were more likely to give birth at HF (aOR = 3.130; 95% CI: 1.744 to 5.618; P<0.001) than non-educated women. Women residing in Koshi and Madhesh provinces were less likely to give birth to their child at HF (aOR = 0.399; 95% CI: 0.177 to 0.901; P<0.05) and (aOR = 0.198; 95% CI: 0.083 to 0.476; P<0.001), respectively, than in Bagmati province. Bagmati province was selected as the reference category because it had the highest proportion of facility-based delivery (94.6%) among all provinces. Poorer and middle wealth index women were more likely to give birth to a child at HF (aOR = 1.701; 95% CI: 1.021 to 2.833; P<0.05) and (aOR = 2.597; 95% CI: 1.461 to 4,615; P< 0.01), respectively, compared to women with the poorest wealth index.

4. Discussion

This study analyzed data from 1,056 women who had given birth within the past year to examine factors associated with facility-based delivery in Nepal. The overall facility delivery rate was 83.3%. In the adjusted multivariate model, engagement in the HI program, education, province, and wealth index remained significantly associated with facility delivery. After adjusting for socio-demographic factors, the strongest predictors of facility delivery were secondary or higher education (aOR = 3.130; 95% CI: 1.744 to 5.618), higher wealth status (richest vs. poorest: aOR = 14.688; 95% CI: 4.142 to 52.082), and province of residence, with women in Koshi (aOR = 0.399) and Madesh (aOR = 0.198) significantly less likely to deliver at facilities compared to those in Bagmati province. HI coverage remained positively associated with facility delivery (aOR = 2.027; 95% CI: 1.773 to 3.994), though the effect was attenuated after adjustment. The wide confidence interval observed for the unadjusted insurance effect (cOR = 5.775; 95% CI: 2.347 to 14.212) is likely due to the small number of uninsured women who delivered at a HF compared to the nearly universal delivery among insured women, resulting in greater estimate instability despite the large overall sample size. Similarly, the side confidence interval for the Richest wealth quintile (95% CI: 4.142 to 52.082) in the adjusted model is likely due to the very small number of home deliveries in that category (just 1.7%), resulting in near-perfect prediction and estimate instability.

This study suggests that among women who gave birth in the previous year, those engaged in the HI program were more likely to use HF than those who were not covered by HI. This outcome was consistent with a Philippine study that found that those women covered by the HI scheme were more likely to use Facility-Based Delivery (FBD). ⁸ Our findings, which show a positive association between insurance status and delivery at HF, align with the broader body of evidence from other LMICs, including the Philippines, Ghana, Nigeria, Kenya, and Burkina Faso.^8,16 A systematic review of 14 studies examining the association between HI and institutional delivery, conducted across countries such as Ghana, Nigeria, Kenya, and Burkina Faso, found that 10 indicated a significant positive association, while four did not. ¹⁷ A more recent systematic review confirmed that health insurance is consistently associated with increased utilization of maternal health services, including facility-based delivery. ¹⁸ A more recent study from Indonesia ¹⁹ and another from Ethiopia ²⁰ have similar reported positive associations between HI enrollment and facility-based childbirth, reinforcing these findings across diverse LMIC contexts. Nevertheless, this association is not driven solely by financial coverage; it is also heavily influenced by geographical access to hospitals, such as poor road conditions, long travel distances, and high transportation costs, as well as personal beliefs, cultural norms, and women’s previous childbirth experiences, all of which can discourage facility use even when insurance is held. Insurance alone cannot address dependency on family members to care for older children while the women is in the hospital or overcome deeply ingrained cultural preferences for home births with traditional birth attendants. ²¹ These logistical and psychological factors often explain why some studies report only a weak or non-significant link between insurance and delivery practice, highlighting that removing financial barriers alone is insufficient without simultaneously addressing physical accessibility and community perceptions of maternal care.

Our study revealed a positive, statistically significant association between socio-demographic factors, including engagement in the HI program, education, province, and the combined wealth index, and child delivery practice at Health Facilities (HF) last year. In a similar vein to the present findings, ¹⁶ revealed that the community where family members are living, and peer group involvement influenced women’s maternal health behavior. The authors further highlighted that educated family members had higher incomes, lived in urban areas, and often referred their childbearing to health facilities rather than home, even if their family members were health workers or doctors. Still, they did not mention people’s engagement in the HI program.

This study found that a large majority of women sought care in health facilities. Consistent with the present findings, the Institute of Medicine (IoM), USA (2002), stated that health insurance status affects the care received by women giving birth and their newborns, with uninsured women and their newborns receiving, on average, less prenatal care and fewer expensive prenatal services. As a result, insured women report greater difficulty obtaining the care they believe they need. The report further articulated that uninsured women are more likely to have poor outcomes during pregnancy and delivery than are women with insurance. Further, uninsured newborns are more likely to have adverse consequences, including low birth weight and death, than are insured newborns. Although the IoM report originates from a high-income setting (USA), its findings underscore the universal importance of financial protection for maternal health outcomes, a principle that is equally relevant in LMIC contexts. Similar to the present findings, a study in China by. ²² Showed an association between insurance and facility-based deliveries, including skilled health workers. While China is classified as an upper-middle-income country, its experience with different insurance schemes provides valuable insights for Nepal’s expanding HI program. Attendance at birth was also consistent across different types of insurance schemes, such as all (100%) women who came from the National Health Insurance Scheme (NHIS), 91% of women who came from rural Co-operative Medical Systems (CMS), and less than half (46%) of uninsured women. Two studies in Ghana provide conflicting evidence: a study by ⁹ found that women insured under the NHIS were more likely to deliver in a hospital and with professional assistance, whereas another study by ²³ found no such effect.

A study in Brazil by ²⁴ confirmed our findings that HI, which reflects the extent of better-quality health care, may influence the impact on maternal health (MH) services. In Brazil, HI-related quality services, such as increased staffing, expanded bed capacity, universal access to surfactants, improved respirators, and enhanced laboratory techniques, positively affect health outcomes, including hospital delivery and maternal and neonatal outcomes. In contrast, a study in Peru ²⁵ revealed that a quality improvement program that included improvements in physical infrastructure, health care quality, and the expansion of social HI coverage was less likely to be delivered in health facilities. However, the author clarifies that this was due to a potential lack of awareness among the target population about the facility improvements mentioned.

Other literature found that HI is not the only factor affecting the use of health care facilities during childbirth; it is obviously one of many factors that influence health outcomes. These factors are also examined in our study. Similar to the present findings, a couple of other studies stated that socio-economic status, ²⁶ ethnicity, ²⁷ wealth, ²⁸ educational attainment ²⁹ and health-related behaviors such as diet, exercise, drug use, smoking, and consuming alcohol ¹⁷ may affect the visit and delivery of a child at a health facility. A study in the Philippines suggests that, for poor and rural women, access to insurance increased utilization of FBD. Research^8,30 in Burkina Faso has reported similar findings regarding institutional delivery and other health care services in China. ³¹

4.1. Strengths and limitations of the study

The NDHS 2022 was a national representative survey. Therefore, it can be assumed that the results presented here could represent the entire country of Nepal that would benefit from policy intervention. Importantly, no women were excluded from the final analytic sample due to missing data on place of delivery or lack of consent, as all 1,056 women who had given birth in the past year had complete information on place of delivery, and informed consent was obtained from all respondents in the original survey. However, the study has some limitations too. The survey was primarily quantitative; as a result, qualitative information was missing, leading to a lack of in-depth understanding. Because we used secondary data, some variables were not included in the survey. Proxy measures were used to determine the wealth index. Furthermore, we relied on the sample size determined for the main NDHS survey, which may not have been specifically powered for this sub-analysis of women with a recent birth. This is a limitation of our study. All the limitations of the quantitative survey persisted in this study too.

5. Conclusions

The study showed that health insurance coverage, educational level, province, and the combined wealth index were significantly associated with the delivery practice of the recent child in health facilities. Women enrolled in the health insurance scheme were more likely to deliver at health facilities in the past year than those not enrolled. In the same case, while adjusting for all socio-demographic variables, the delivery status at HF was not statistically significant. Women with secondary or higher education were more likely to give birth to a recent child at HF. Women in the poorer and Middle wealth indices were more likely to give birth at HF than those in the poorest wealth index.

In contrast, there was insufficient evidence to claim that age and place of residence were statistically significantly associated with the delivery practice of the recent child in health facilities. In fact, age group, ethnicity, and place of residence were not significant predictors of the delivery practice of the recent child in health facilities. Thus, to increase the use of HF during delivery, targeted interventions addressing educational and economic disparities, particularly in provinces with lower utilization, are needed.

ORCID iD

Mohan Kumar Sharma https://orcid.org/0000-0002-7600-8223