Assessing the associated factors of anxiety and depression among the adult population to harness the demographic dividend in Nepal using the socio-ecological model

Abstract

Introduction

Anxiety and depression are major global public health concerns and a growing mental health burden among working-age adults, potentially hindering low- and middle-income countries from realizing the demographic dividend. This study aims to examine associated factors with anxiety and depression among adults using a socio-ecological model.

Methods

This study used secondary data from the Demographic and Health Survey (DHS) 2022 of Nepal among adults aged 15–49 years. Anxiety and depression were measured using GAD-7 and PHQ-9 and categorized as binary outcomes. Multivariable binary logistic regression was performed using complex sample analysis, accounting for stratification, clustering, and sampling weights to identify associated factors.

Results

The prevalence of anxiety and depression was higher among women (21.9% and 21.1%, respectively) than among men (11.3% and 11.1%, respectively). Among women, higher odds of anxiety and depression were observed among widowed/divorced/separated (AOR = 1.813 for anxiety;1.850 for depression), smokers (AOR = 1.615;1.528), those in the poorer wealth group (AOR = 2.033;1.848), and residents of Karnali province (AOR = 1.831;1.667) compared to their respective reference groups. Similarly, among men, higher odds were found among widowed/divorced/separated (AOR = 3.980;2.844), smokers (AOR = 1.583;1.354), those in the poorer wealth group (AOR = 2.094;2.280), and residents of Karnali province (AOR = 4.190;6.079), compared to their respective reference groups, respectively.

Conclusions

Harnessing demographic dividend requires prioritizing mental health through a comprehensive multisectoral approach that addresses not only existing individual-level anxiety and depression but also underlying interpersonal, community and societal factors such as structural disparities, economic inequalities, and uneven development across regions through coordinated actions across health, education, employment, social protection, culture-sensitive, and region-specific responsive strategies.

Introduction

The increasing prevalence of anxiety and depression is having an adverse effect on individual wellbeing and economic outcomes [1–5]. Anxiety and depression are the leading contributors to the global burden of mental health, significantly affecting emotional well-being [6], social functioning, productivity, and quality of life among adults [7]. These are common mental health disorders characterized by persistent feelings of worry, fear, sadness, and loss of interest that can affect daily function and overall wellbeing [6]. Globally and in South Asia, anxiety and depression among adults have become major public health concerns, and they create adverse effects on wellbeing, productivity, and socio-economic development [8]. In low- and middle-income countries such as Nepal, rapid social transition, socio-economic inequality, migration, and limited access to mental health services further increase vulnerability to mental health issues [9, 10]. The findings from the 2022 Demographic and Health Survey of Nepal indicate that mental health outcomes vary across key demographic and socio-economic factors such as age, education, and economic status of family [11], reflecting underlying structural inequities in social determinants of mental health [12].

Studies by the WHO show that women worldwide of all ages experience depression and anxiety more frequently than men [13]. This higher burden is closely linked to persistent gender inequalities, which adversely affect women’s well-being, work productivity, and broader human capital development [14, 15]. Such gender disparities in depression and anxiety have been documented across many countries and are shaped by complex interaction among biological, psychological and social determinants [16, 17]. These mental health conditions are also among the leading causes of disability worldwide, contributing to reduced productivity, increased healthcare costs, and reduced quality of life [18]. Collectively, this global evidence emphasizes the urgent need to prioritize mental health through national public health programs and the Sustainable Development Goals (SDGs) [5].

According to National Population and Housing Census 2021 reports, 53.4 percent of Nepal’s population were adults aged 15–49 years, reflecting strong potential to harness the demographic dividend [19–21]. The demographic dividend refers to the opportunity for accelerated economic growth arising from a larger proportion of the working-age population, which can only be achieved through investment in health, education and employment opportunities [21, 22]. To achieve the demographic dividend, it is essential to create policies that lead to a healthy, skilled, and employed workforce [23, 24]. Mental health is a key factor in this context, as poor mental health can reduce productivity, labour force participation and engagement, and overall human capital development [7]. Anxiety and depression may also increase dependency and limit an individual’s ability to fully utilize opportunities [6], resulting in limiting the effective harnessing of the demographic dividend.

According to the Nepal Living Standards Survey 2022/23, 34.4 percent of the population was employed, while 4.7 percent was unemployed and 62.9 percent remained outside the labour force, resulting in a labour force participation rate of 37.1 percent, with an unemployment rate of 12.6 percent [25]. In this context, unemployment is not only an economic issue but also a developmental challenge [26] directly linked to Nepal’s capacity to harness its demographic dividend. Although the Government of Nepal has acknowledged the potential of the demographic dividend and the importance of education, skills, and youth engagement in its National Population Policy 2014 and 2025 [27, 28], limited implementation of mental health and workforce-related policies may hinder the country’s ability to fully harness the demographic window of opportunity.

Although there are several policies aimed at addressing mental health as part of the discourse regarding demographic dividend, significant gaps still exist regarding the provision of services for mental health. Nepal’s National Mental Health Policy and the National Mental Health Strategy and Action Plan have identified mental health as a priority, but neither has received sufficient funding, nor has there been sufficient integration into primary healthcare service delivery systems, and implementation in these two frameworks is not being done well, especially with regard to delivery of services to rural and underserved communities [29, 30]. Access to mental health services in Nepal remains limited due to inadequate mental health professionals, unequal service distribution, limited financial investment, and insufficient integration of mental health care within primary health care services [9]. Furthermore, the existing policies do not adequately take into account gendered differences in the experience of mental illness and therefore are not capable of adequately meeting the needs of women who typically have a greater burden of mental illness than men [31]. Addressing gender differences in mental health will realize both humanistic and economic advantages. Poor mental health, particularly conditions such as anxiety and depression, is likely to reduce the ability of people of working age to participate in the labour market, limit the productivity of those who are employed, and increase the dependency ratio for income earners [32], thereby constraining the effective utilization of the demographic dividend.

By using the socio-ecological model as a framework, this study aims to assess the factors associated with anxiety and depression within the adult population of Nepal and to examine their implications for harnessing the demographic dividend. Additionally, this study also examines the gender differentials in the prevalence of anxiety and depression along with the association of the socio-ecological model’s (individual, interpersonal, community, and societal) factors with the mental health outcomes of adults in Nepal. The findings of this study contribute to policy discourse by positioning mental health as a critical enabler of human capital formation and sustainable socio-economic development in Nepal.

Methods

Study design and data sources

This study utilized a cross-sectional analytical design using data from the 2022 Demographic Health Survey of Nepal, and ethical approval for the DHS 2022 was obtained from the ICF Institutional Review Board and the Nepal Health Research Council (NHRC). The survey was conducted by the Ministry of Health and Population (MoHP) in collaboration with ICF International and is recognized as a national representative survey of Nepal [11]. The sampling frame utilized the most recently updated version of the 2022 Nepal Population and Housing Census (NPHC) provided by the National Statistics Office [11]. This study utilized secondary data from the nationally representative survey Nepal Demographic and Health Survey (NDHS) 2022. Details of the survey’s sampling design and procedures are available in the NDHS 2022 report.

Study setting

Nepal is located on the South Asian subcontinent between China and India. It is a landlocked country with a surface area of 147,181 km². Nepal has one federal government, seven provincial governments and 753 local governments in its administrative structure. Nepal includes seven provinces (Koshi, Madhesh, Bagmati, Gandaki, Lumbini, Karnali, and Sudurpaschim) and three ecological regions, viz., mountain, hill, and Terai (plain area) [33]. As of the 2021 National Population and Housing Census, there are 291,164,578 people living in Nepal, of which 51.1 percent are female and 48.9 percent male [19]. There are 142 classified castes in Nepal, which fit into four broad ethnic classifications such as Brahmin/Chhetri, Janajati, Dalit, and Madhesi. According to the United Nations Development Program (UNDP), Nepal’s Human Development Index (HDI) has an HDI ranking of 145th out of 193 nations with an HDI value of 0.622 [34].

Study population and sample size

The study population consisted of adult women and men aged 15–49 years who were participants in DHS 2022 of Nepal. A total number of 7,410 eligible women and 4,913 eligible men were included for analysis. The survey used a geographically representative sample, ensuring adequate representation across all seven provinces, three ecological regions and two areas of residence of Nepal.

Conceptual framework

This study was guided by the Socio-Ecological Model (SEM) to examine the associated factors of anxiety and depression among adults aged 15–49 years for realization of the demographic dividend in Nepal. The SEM explains mental health outcomes as the result of interactions among multiple associated factors. Firstly, the individual factors refer to personal demographic characteristics, socio-economic attributes, health-related behavior that is directly associated with individual mental health status and decision-making. Secondly, the interpersonal factors refer to relationships with family members, socio-economic status of family, partners and peers, and social networks that shape mental health through support, norms, and role expectations of behavior. Thirdly, the community factors encompass socio-cultural norms and social identities within the community that are collectively associated with the mental health behaviors of individuals. Finally, the societal factors represent structural factors, including policies, geographical location, and social inequalities, that create systemic opportunities or constraints affecting the mental health outcomes of individuals [35]. This multi-level framework informed the selection of variables and the multilevel analytical approach to be adopted in the study (Fig. 1).

Fig. 1.

Framework for associated factors with anxiety and depression among adults in Nepal based on socio-ecological model

Variables

Symptoms of anxiety were assessed using the Generalized Anxiety Disorder 7 (GAD-7) scale, included in the Mental Health Module. The GAD-7 consists of seven items designed to measure the main features of anxiety: persistent and impairing worry [36]. GAD-7 scores range from a minimum of 0 to a maximum of 21, with scores of 0–5 indicating mild anxiety, 6–14 indicating moderate anxiety, and 15–21 severe anxiety [36, 37]. Similarly, symptoms of depression were measured using the Patient Health Questionnaire-9 (PHQ-9), which included nine items derived from Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria for depression [38]. The PHQ-9 produces scores ranging from 0 to 27 [36], where 0–4 indicate minimal or no depression symptoms, 5–9 indicate mild depression, 10–14 indicate moderate depression, 15–19 indicate moderately severe depression, and 20–27 indicate severe depression [38].

Outcome variable

In this study, the outcome variables were anxiety and depression, measured as binary variables (0 indicating absence of symptoms, 1 indicating presence of symptoms). Anxiety was assessed using the GAD-7 scale, where respondents with mild symptoms were classified as “No anxiety” (0), and those with moderate and severe anxiety were classified as “Yes” (1). Similarly, depression was measured using PHQ-9 scale, where individuals with minimal symptoms were categorized as “No depression” (0), while those with mild, moderate, moderately severe, or severe symptoms were categorized as “Yes”. All analyses were further stratified by gender (women and men) to examine sex-specific associations.

Independent variables

In this study, 22 independent variables were used in 4 groups of the Socio-Ecological Model (SEM). For the first group, individual (intrapersonal) factors were based on socio-demographic characteristics, health behaviors, psychosocial factors which included age group, marital status, education level, occupation, currently working, migration status, internet use, media exposure, health program exposure, self-reported status, recent sexual activity, consumption of alcohol, and smoking status. The second group is comprised of interpersonal factors, including wealth index, sex of the household head, decision-making of health care, and covered by health insurance in accordance with household economic status and decision-making autonomy. The third group, community factors, included religion and caste/ethnic group, reflecting social identity and norms, and in the fourth group, societal factors consisted of place of residence, ecological region, and province, capturing structural and geographical context. Currently working status, consumption of alcohol, wealth index, sex of the household head, covered by health insurance, place of residence, ecological region, and province were retained as defined in the original DHS dataset, whereas other variables such as age groups, marital status, educational level, occupation, migration status, self-reported health status, recent sexual activity, smoking status, decision-making for health care, religion, and caste/ethnic group were adjusted as presented in Tables 1 and 2.

Table 1.

Percentage distribution of women and men by anxiety and depression in Nepal by study variables, DHS 2022

Variables	Women				Men
	Anxiety	Depression	Total N	Anxiety		Depression	Total N
	N (%) [95%CI]	N (%) [95%CI]		N (%) [95%CI]		N (%) [95%CI]
Individual factor
Age group
< 20	242 (18.3) [15.8–21.1]	242 (18.3) [15.8–21.1]	1322	98 (10.0) [7.8–12.6]		109 (11.0) [9.0–13.5]	985
20–34	792 (22.1) [20.1–24.3]	758 (21.2) [19.3–23.2]	3581	264 (12.1) [10.3–14.1]		260 (11.9) [10.2–13.8]	2190
35–49	591 (23.6) [21.5–25.8]	564 (22.5) [20.5–24.7]	2507	192 (11.1) [9.2–13.2]		178 (10.2) [8.4–12.4]	1739
Marital status
Never married	281 (17.2) [15.0–19.6]	286 (17.5) [15.4–19.9]	1632	192 (10.9) [9.0–13.1]		217 (12.3) [10.1–14.8]	1768
Married/Living with partner	1248 (22.6) [20.8–24.4]	1185 (21.4) [19.8–23.2]	5533	346 (11.2) [9.7–12.9]		318 (10.2) [8.8–11.9]	3101
Widowed/Divorced/Separated	97 (39.5) [32.7–46.8]	93 (37.7) [31.0–45.0]	245	16 (37.1) [23.1–53.6]		12 (27.4) [15.6–43.5]	44
Education level
No education	489 (25.2) [22.3–28.3]	461 (23.7) [21.1–26.6]	1944	29 (7.5) [5.0–11.1]		29 (7.4) [4.9–11.0]	393
Basic	537 (23.8) [21.5–26.2]	490 (21.7) [19.7–23.9]	2256	220 (11.6) [9.7–13.8]		222 (11.7) [9.8–13.9]	1898
Secondary and higher	599 (18.7) [17.0–20.4]	612 (19.1) [17.3–21.0]	3211	305 (11.6) [10.0–13.5]		296 (11.3) [9.6–13.2]	2621
Respondents’ occupation
Non agriculture	379 (21.3) [18.8–24.1]	358 (20.2) [17.9–22.6]	1778	360 (11.7) [10.1–13.6]		340 (11.0) [9.4–13.0]	3080
Agriculture	839 (23.4) [21.4–25.5]	818 (22.8) [20.9–24.8]	3591	126 (10.9) [9.2–12.9]		135 (11.7) [9.7–14.1]	1156
Not working	405 (19.9) [17.5–22.5]	385 (18.9) [16.5–21.6]	2033	68 (10.2) [7.6–13.5]		71 (10.6) [8.1–13.8]	672
Total #	1623 (21.9) [20.4–23.6]	1561 (21.1) [19.6–22.6]	7403	555 (10.3) [10.0–12.8]		546 (11.1) [9.8–12.7]	4907
Currently working
No	620 (20.9) [18.7–23.2]	602 (20.2) [18.1–22.5]	2975	129 (11.5) [9.3–14.3]		974 (13.0) [10.7–15.6]	1119
Yes	1005 (22.7) [20.9–24.5]	962 (21.7) [20.0–23.4]	4436	425 (11.2) [9.8–12.8]		3392 (10.6) [9.1–12.3]	3794
Migration status
Non-migrant	550 (19.2) [17.5–20.9]	562 (19.6) [17.9–21.5]	2865	393 (10.7) [9.4–12.3]		3264 (10.8) [9.4–12.3]	3658
Migrant	1076 (23.7) [21.6–25.8]	1002 (22.0) [20.3–23.9]	4546	162 (12.9) [10.4–15.8]		1103 (12.2) [9.7–15.2]	1255
Internet use
No	629 (24.8) [22.6–27.1]	607 (23.9) [21.9–26.1]	2537	137 (12.6) [10.4–15.3]		956 (11.8) [9.6–14.5]	1084
Yes	997 (20.5) [18.7–22.3]	956 (19.6) [18.0–21.4]	4873	418 (10.9) [9.5–12.5]		3411 (10.9) [9.5–12.6]	3829
Media exposure
No	451 (29.1) [25.4–33.0]	432 (27.9) [24.4–31.7]	1550	90 (12.0) [9.5–15.2]		680 (9.1) [6.8–12.1]	747
Yes	1175 (20) [18.6–21.5]	1131 (19.3) [17.9–20.7]	5860	464 (11.2) [9.8–12.7]		3687 (11.5) [10.0–13.1]	4166
Health program exposure
No	1241 (21.8) [20.0–23.7]	1187 (20.8) [19.2–22.6]	5694	387 (11.1) [9.5–13.0]		3116 (10.5) [8.9–12.3]	3481
Yes	385 (22.4) [20.1–24.9]	377 (22.0) [19.7–24.5]	1717	168 (11.7) [9.9–13.8]		1250 (12.7) [10.8–14.8]	1432
Self-reported health status
Very good/Good	431 (16.4) [14.3–18.7]	396 (15.0) [13.1–17.2]	2632	186 (8.0) [6.7–9.5]		2156 (7.7) [6.5–9.2]	2336
Moderate	877 (21.3) [19.7–23.1]	843 (20.5) [18.9–22.2]	4110	295 (12.6) [10.6–14.8]		2047 (12.8) [10.8–15.1]	2348
Very bad/Bad	317 (47.4) [42.8–52.1]	325 (48.6) [44.2–52.9]	669	74 (32.1) [25.7–39.2]		164 (28.6) [22.3–35.8]	230
Recent sexual activity
Never had sex	270 (16.9) [14.8–19.3]	276 (17.3) [15.2–19.6]	1597	121 (9.8) [7.7–12.5]		1092 (11.2) [9.0–13.9]	1230
Active in last 4 weeks	727 (21.1) [19.2–23.1]	697 (20.2) [18.4–22.1]	3454	284 (10.7) [9.2–12.3]		2387 (10.1) [8.7–11.7]	2655
Not active in last 4 weeks	622 (26.7) [24.2–29.4]	586 (25.2) [22.8–27.7]	2327	150 (14.6) [11.9–17.7]		888 (13.7) [11.3–16.5]	1029
Total #	1619 (21.9) [20.4–23.6]	1559 (21.1) [19.7–22.7]	7378	555 (10.3) [10.0–12.8]		546 (11.1) [9.8–12.7]	4913
Consumption of alcohol
No	1428 (21.6) [19.9–23.4]	1385 (20.9) [19.4–22.6]	6612	285 (10.0) [8.6–11.7]		290 (10.2) [8.7–12.0]	2836
Yes	198 (24.8) [21.6–28.3]	179 (22.4) [19.2–25.8]	799	270 (13.0) [11.0–15.3]		256 (12.3) [10.5–14.4]	2077
Smoking status
Non-smoker	1522 (21.4) [19.8–23.0]	1468 (20.6) [19.1–22.2]	7116	342 (9.7) [8.4–11.1]		353 (10.0) [8.7–11.4]	3539
Smoker	104 (35.2) [29.3–41.6]	96 (32.6) [27.2–38.5]	294	213 (15.5) [12.9–18.4]		194 (14.1) [11.7–16.9]	1374
Interpersonal factors
Wealth index combined
Poorest	293 (21.8) [19.5–24.3]	308 (22.9) [20.5–25.5]	1344	104 (13.9) [11.4–16.9]		104 (13.8) [11.3–16.7]	751
Poorer	350 (25.5) [22.5–28.9]	353 (25.7) [22.6–29.1]	1372	125 (13.4) [10.7–16.6]		119 (12.8) [10.2–15.8]	933
Middle	373 (24.6) [21.6–27.9]	329 (21.8) [19.1–24.7]	1512	84 (8.8) [6.8–11.3]		93 (9.7) [7.6–12.4]	957
Richer	363 (21.3) [18.8–24.0]	327 (19.2) [16.7–22.0]	1704	124 (11.0) [8.8–13.7]		123 (10.8) [8.4–13.9]	1135
Richest	247 (16.7) [14.0–19.7]	247 (16.7) [14.4–19.3]	1479	117 (10.3) [7.9–13.3]		107 (9.5) [7.1–12.4]	1137
Sex of household head
Male	1025 (21.2) [19.5–22.9]	999 (20.6) [19.1–22.3]	4843	438 (11) [9.6–12.4]		439 (11.0) [9.5–12.6]	4003
Female	600 (23.4) [21.0–25.9]	565 (22.0) [19.9–24.3]	2568	116 (12.8) [10.0–16.1]		108 (11.8) [9.2–15.0]	910
Decision making for health care
Respondent alone	323 (26.3) [23.3–29.6]	299 (24.4) [21.5–27.6]	1227	169 (11.2) [9.3–13.4]		138 (9.1) [7.5–11.1]	1512
Respondent, husband/partner, and others	561 (20.6) [18.7–22.8]	536 (19.7) [18.0–21.5]	2718	138 (10.9) [9.0–13.1]		141 (11.1) [9.1–13.4]	1270
Husband/partner alone	258 (22.9) [20.0–26.0]	253 (22.4) [19.4–25.8]	1128	29 (17.1) [10.0–27.7]		25 (15.2) [9.5–23.4]	168
Someone else and others	105 (22.9) [18.2–28.4]	97 (21.1) [16.3–26.8]	460	10 (6.3) [3.2–12.0]		14 (9.1) [5.2–15.5]	152
Total #	1248 (22.6) [20.8–24.4]	1185 (21.4) [19.8–23.2]	5533	346 (11.2) [9.7–12.9]		318 (10.2) [8.8–11.9]	3101
Covered by health insurance
No	1429 (22) [20.3–23.7]	1381 (21.2) [19.7–22.9]	6506	464 (10.9) [9.5–12.5]		466 (10.9) [9.5–12.6]	4261
Yes	197 (21.7) [18.1–25.9]	183 (20.2) [17.0–23.9]	904	91 (13.9) [11.0–17.4]		80 (12.3) [9.3–16.2]	652
Community factors
Religion
Hindu	1358 (22.1) [20.4–23.8]	1281 (20.8) [19.3–22.4]	6151	474 (11.8) [10.3–13.4]		448 (11.1) [9.7–12.8]	4025
Buddhist	84 (17.8) [13.9–22.4]	87 (18.4) [14.2–23.5]	471	31 (8.0) [5.2–12.3]		46 (11.9) [8.4–16.6]	389
Muslim/Kirat/Christian/Others	184 (23.4) [19.3–28.0]	196 (24.8) [20.6–29.7]	788	50 (10.0) [6.7–14.5]		53 (10.5) [7.7–14.2]	499
Caste/Ethnic group
Brahmin/Chhetri	423 (20.7) [18.5–23.0]	429 (21.0) [18.9–23.2]	2049	172 (13.9) [11.7–16.5]		148 (12.0) [9.9–14.6]	1232
Dalit	313 (28.1) [24.1–32.5]	278 (24.9) [21.8–28.3]	1115	104 (15.7) [12.5–19.6]		87 (13.2) [10.2–17.1]	658
Janajati	535 (19.6) [17.9–21.4]	546 (20.0) [18.0–22.1]	2735	199 (10.6) [8.9–12.7]		241 (12.9) [10.8–15.3]	1869
Madhesi/Muslim/Others	354 (23.4) [19.4–28.0]	310 (20.5) [17.0–24.6]	1511	81 (7.0) [5.2–9.3]		70 (6.1) [4.5–8.1]	1154
Societal factors
Place of residence
Urban	1099 (21.7) [19.7–23.9]	1038 (20.5) [18.7–22.5]	5064	399 (11.5) [9.8–13.5]		384 (11.1) [9.3–13.1]	3462
Rural	527 (22.5) [20.2–24.8]	526 (22.4) [20.1–24.9]	2347	156 (10.7) [9.1–12.6]		163 (11.2) [9.5–13.2]	1451
Ecological region
Mountain	89 (21.9) [16.6–28.3]	97 (23.8) [18.3–30.5]	408	45 (17.7) [12.2–24.9]		52 (20.3) [13.1–30.1]	255
Hill	571 (19.7) [18.1–21.4]	599 (20.7) [18.9–22.6]	2896	218 (11.1) [9.1–13.4]		226 (11.5) [9.2–14.2]	1973
Terai	965 (23.5) [21.0–26.2]	868 (21.1) [18.9–23.6]	4106	291 (10.9) [9.1–12.9]		269 (10.0) [8.4–11.8]	2685
Province
Koshi	302 (24.4) [21.0–28.1]	320 (25.8) [22.3–29.6]	1241	121 (13.7) [10.0–18.6]		111 (12.6) [9.3–16.9]	882
Madhesh	333 (22.1) [17.5–27.4]	302 (20.0) [15.9–24.8]	1512	65 (6.5) [4.6–9.0]		46 (4.7) [3.0–7.1]	997
Bagmati	284 (19.0) [16.0–22.4]	277 (18.5) [15.7–21.7]	1493	164 (13.5) [10.5–17.3]		176 (14.5) [10.7–19.3]	1214
Gandaki	125 (17.8) [14.2–21.9]	121 (17.2) [14.0–21.0]	704	32 (8.2) [5.0–13.1]		34 (8.8) [5.8–13.1]	387
Lumbini	298 (21.9) [18.3–26.0]	268 (19.7) [16.3–23.5]	1360	95 (11.7) [9.1–15.0]		97 (12) [9.4–15.2]	812
Karnali	127 (27.8) [23.9–32.1]	133 (29.0) [25.7–32.5]	458	47 (17.8) [14.1–22.2]		50 (18.9) [15.3–23.1]	266
Sudurpaschim	155 (24.2) [20.9–27.9]	143 (22.3) [18.9–26.1]	641	30 (8.5) [5.8–12.2]		31 (8.8) [6.1–12.5]	355
Total	1626 (21.9) [20.4–23.6]	1564 (21.1) [19.6–22.6]	7410	555 (11.3) [10.0–12.8]		546 (11.1) [9.8–12.7]	4913

# refers to different total from all other variables total due to missing value.

N refers to Number .

CI refers to Confidence Interval.

Table 2.

Factors associated with anxiety and depression by gender in Nepal with study variables, DHS 2022

Variables	Women		Men
	Anxiety	Depression	Anxiety	Depression
	AOR (95%CI)	AOR (95%CI)	AOR (95%CI)	AOR (95%CI)
Individual factor
Age group
< 20	1.377 (1.028–1.845) *	1.348 (1.005–1.808) *	1.245 (0.762–2.032)	1.109 (0.670–1.836)
20–34	1.294 (1.094–1.529) **	1.261 (1.060–1.501) **	1.231 (0.946–1.602)	1.227 (0.934–1.611)
35–49	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Marital status
Never married	1.341 (0.495–3.635)	1.557 (0.610–3.973)	0.892 (0.575–1.383)	1.326 (0.801–2.197)
Married/Living with partner	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Widowed/Divorced/Separated	1.813 (1.341–2.452) ***	1.850 (1.327–2.578) ***	3.980 (1.942–8.158) ***	2.844 (1.298–6.232) **
Education level
No education	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Basic	1.066 (0.871–1.306)	1.053 (0.866–1.279)	1.738 (1.053–2.870) *	1.474 (0.896–2.424)
Secondary and higher	0.901 (0.701–1.158)	1.066 (0.836–1.360)	2.280 (1.361–3.822) **	1.555 (0.902–2.680)
Respondents’ occupation
Non agriculture	1.120 (0.863–1.454)	1.167 (0.908–1.500)	1.227 (0.761–1.980)	1.505 (0.912–2.482)
Agriculture	1.132 (0.903–1.418)	1.271 (1.004–1.609) *	1.141 (0.704–1.850)	1.555 (0.971–2.489)
Not working	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Currently working
No	1.055 (0.866–1.286)	1.125 (0.926–1.368)	1.149 (0.805–1.640)	1.518 (1.046–2.202) *
Yes	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Migration status
Non-migrant	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Migrant	1.264 (1.086–1.472) **	1.142 (0.990–1.317)	1.226 (0.942–1.595)	1.162 (0.893–1.512)
Internet use
No	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Yes	0.921 (0.780–1.087)	0.872 (0.744–1.021)	0.733 (0.567–0.948) *	0.783 (0.597–1.026)
Media exposure
No	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Yes	0.652 (0.541–0.786) ***	0.653 (0.544–0.784) ***	0.826 (0.609–1.120)	1.223 (0.859–1.741)
Health program exposure
No	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Yes	1.194 (1.008–1.414) *	1.197 (1.020–1.405) *	1.118 (0.855–1.462)	1.258 (0.981–1.614)
Self-reported health status
Very good/Good	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Moderate	1.372 (1.154–1.632) ***	1.436 (1.200–1.719) ***	1.645 (1.296–2.087) ***	1.798 (1.423–2.274) ***
Very bad/Bad	4.331 (3.467–5.410) ***	5.125 (4.137–6.349) ***	5.619 (3.862–8.176) ***	5.057 (3.426–7.464) ***
Recent sexual activity
Never had sex	0.703 (0.255–1.934)	0.606 (0.234–1.570)	1.179 (0.649–2.143)	1.049 (0.622–1.769)
Active in last 4 weeks	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Not active in last 4 weeks	1.232 (1.062–1.431) **	1.193 (1.019–1.397) *	1.279 (0.906–1.807)	1.142 (0.839–1.555)
Consumption of alcohol
No	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Yes	1.111 (0.907–1.360)	1.015 (0.814–1.265)	1.169 (0.912–1.499)	1.162 (0.923–1.463)
Smoking status
Non-smoker	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Smoker	1.615 (1.179–2.210) **	1.528 (1.132–2.063) **	1.583 (1.245–2.013) ***	1.354 (1.077–1.701) *
Interpersonal factors
Wealth index combined
Poorest	1.692 (1.247–2.296) **	1.702 (1.279–2.265) ***	2.140 (1.347–3.398) **	2.342 (1.451–3.781) **
Poorer	2.033 (1.515–2.728) ***	1.848 (1.387–2.462) ***	2.094 (1.321–3.320) **	2.280 (1.395–3.726) **
Middle	1.950 (1.451–2.620) ***	1.572 (1.164–2.124) **	1.038 (0.616–1.749)	1.386 (0.838–2.291)
Richer	1.502 (1.119–2.016) **	1.299 (0.987–1.710)	1.356 (0.803–2.290)	1.474 (0.846–2.571)
Richest	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Sex of household head
Male	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Female	1.094 (0.931–1.284)	1.052 (0.894–1.238)	1.273 (0.850–1.906)	1.032 (0.658–1.618)
Decision making for health care
Respondent alone	1.288 (0.963–1.722)	1.277 (0.936–1.744)	1.808 (0.900–3.633)	0.978 (0.514–1.864)
Respondent, husband/partner, and others	0.931 (0.705–1.229)	0.953 (0.697–1.302)	1.718 (0.832–3.546)	1.184 (0.621–2.257)
Husband/partner alone	1.054 (0.780–1.425)	1.111 (0.805–1.535)	3.117 (1.258–7.721) *	1.782 (0.793–4.006)
Someone else and others	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Covered by health insurance
No	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Yes	1.059 (0.790–1.420)	0.990 (0.756–1.296)	1.518 (1.066–2.162) *	1.338 (0.894–2.002)
Community factors
Religion
Hindu	1.140 (0.821–1.584)	1.080 (0.778–1.500)	1.460 (0.885–2.409)	1.102 (0.707–1.720)
Buddhist	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Muslim/Kirat/Christian/Others	1.247 (0.848–1.833)	1.440 (0.957–2.167)	1.529 (0.804–2.909)	1.214 (0.723–2.038)
Caste/Ethnic group
Brahmin/Chhetri	0.868 (0.655–1.150)	1.099 (0.834–1.448)	2.180 (1.488–3.193) ***	2.153 (1.463–3.169) ***
Dalit	1.299 (0.972–1.736)	1.362 (1.048–1.770) *	2.512 (1.654–3.814) ***	2.394 (1.594–3.595) ***
Janajati	0.819 (0.631–1.062)	1.010 (0.772–1.320)	1.703 (1.190–2.437) **	2.343 (1.618–3.393) ***
Madhesi/Muslim/Others	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Societal factors
Place of residence
Urban	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Rural	1.054 (0.881–1.260)	1.090 (0.908–1.310)	0.925 (0.714–1.199)	1.015 (0.768–1.342)
Ecological region
Mountain	0.981 (0.718–1.342)	1.006 (0.742–1.365)	1.685 (1.017–2.791) *	1.928 (1.053–3.529) *
Hill	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Terai	1.349 (1.056–1.724) *	1.037 (0.826–1.303)	1.530 (1.111–2.107) **	1.534 (1.083–2.174) *
Province
Koshi	1.259 (0.876–1.809)	1.398 (0.983–1.987)	2.548 (1.488–4.362) **	3.247 (1.804–5.846) ***
Madhesh	Ref. (1)	Ref. (1)	Ref. (1)	Ref. (1)
Bagmati	1.071 (0.691–1.659)	0.943 (0.629–1.412)	3.061 (1.868–5.015) ***	4.727 (2.648–8.441) ***
Gandaki	0.971 (0.632–1.493)	0.854 (0.560–1.304)	1.815 (0.926–3.559)	2.788 (1.410–5.514) **
Lumbini	1.041 (0.721–1.503)	0.974 (0.676–1.402)	2.074 (1.314–3.272) **	2.986 (1.772–5.031) ***
Karnali	1.831 (1.201–2.792) **	1.667 (1.126–2.469) *	4.190 (2.421–7.252) ***	6.079 (3.289–11.236) ***
Sudurpaschim	1.276 (0.898–1.813)	1.156 (0.806–1.658)	1.449 (0.828–2.536)	2.082 (1.124–3.855) *

* = p < 0.050, ** = p < 0.010, *** = p < 0.001

Bold text values: Significant values.

Ref.: Reference category.

Furthermore, media exposure was considered for respondents (women and men) who reported reading newspaper or magazine, frequency of watching television, and frequency of listening to radio, almost every day, at least once a week, and less than once a week. The codes were assigned as “Yesˮ for the activity in which there was participation and “No” for all others. Furthermore, respondents (women and men) were classified as internet user if they reported using it yes, either in last 12 months and yes or before last 12 months; otherwise, they were classified as non-users. Additionally, respondents (women and men) were classified as having health program exposure if they had heard or seen at least one of the eight specific health-related programs on the radio or television; if not, they were classified as having no health program exposure.

Data and statistical analysis

Stata 15.1 was used for data processing and statistical analyses [39]. Analysis of complex samples was done in Stata using the commands ‘svyset’ and ‘svy’ by adjusting for sampling weights, clustering, and stratification in sampling, producing accurate estimates as well as valid significance testing. The multivariable binary logistic regression analyses were conducted separately for anxiety and depression as outcome variables among women and men aged 15–49 years. The selection of explanatory variables was guided by socio-ecological model. Variables were categorized into four domains: individual, interpersonal, community, and societal factors. Domain-specific multivariable logistic regression models were fitted to examine the association of factors within each socio-ecological level with anxiety and depression. This modelling strategy reduced model complexity and minimized the risk of overfitting by limiting the number of variables included in each model. Before undertaking binary logistic regression, the variance inflation factors (VIF) were examined for the detection of multicollinearity, and no multicollinearity issues were detected among independent variables [40]. The adjusted odds ratio (AOR) with 95% confidence intervals (CI) reports results and p < 0.05 as the level of statistical significance.

Result

Among 7,410 women, 21.9 percent had moderate-to-severe anxiety and 21.1 percent had mild-to-severe depression, whereas among 4,913 men, the corresponding proportions were 11.3 percent and 11.1 percent respectively. Regarding anxiety (GAD-7), mild anxiety symptoms were most common among both women (78.1%, 95% CI 76.4%–79.6%) and men (88.7%, 95% CI 87.2%–90.0%). However, the combined prevalence of moderate and severe anxiety was notably higher among women (21.9%, 95% CI 20.4%–23.6%) compared to men (11.3%, 95% CI 10.0%–12.8%). Women also consistently reported higher proportions of moderate and severe anxiety symptoms than men across all severity categories, as shown in Fig. 2.

Fig. 2.

Gender-wise prevalence of anxiety among 15–49 adults in Nepal

Similarly, for depression (PHQ-9), minimal depression symptoms were the most common among both women (78.9%, 95% CI 77.4%–80.4%) and men (88.9%, 95% CI 87.3%–90.0%). In contrast, the combined prevalence of mild-to-severe depression symptoms was higher among women (21.1%, 95% CI 19.6%–22.6%) than men (11.1%, 95% CI 9.8%–12.7%). Across severity categories, women consistently reported higher proportions of mild, moderate, moderately severe, and severe depressive symptoms compared to men, as shown in Fig. 3.

Fig. 3.

Gender-wise prevalence of depression among 15–49 adults in Nepal

Table 1 showed the percentage distribution of anxiety and depression among women and men in Nepal by selected individual (intrapersonal), family (interpersonal) and community and societal factors using DHS 2022 data of Nepal. Overall, women revealed a higher prevalence of both anxiety (21.9%; 95% CI 20.4%–23.6%) and depression (21.1%; 95%CI: 19.6%–22.6%) compared to men, among whom the prevalence of anxiety (11.3%; 95% CI 10.0%–12.8%) and depression (11.1%; 95%CI: 9.8%–12.7%) for men, presented in Table 1.

Among women, the highest prevalence of anxiety (23.6%) and depression (22.5%) was observed among those aged 35–49 years, whereas among men, the highest prevalence of anxiety (12.1%) and depression (11.9%) was reported among aged 20–34 years. Across the marital status categories, women consistently showed a higher prevalence of both outcomes than men. Widowed/Divorced/Separated women had the highest prevalence of anxiety (39.5%) and depression (37.7%), while men in the same category also showed noticeably elevated levels. The clear gradient was observed between self-reported health status and mental health outcomes, with anxiety and depression increasing markedly among both women and men reporting very bad/bad health; nearly half of women and about one-third of men in this group experienced these conditions. Smoking was also associated with a higher prevalence of both outcomes in both sexes, with comparatively higher levels among women smokers. Additionally, anxiety and depression were more prevalent among individuals from poorer and poorest households, particularly among women compared to men. Clear disparities in anxiety and depression were observed across caste/ethnic groups, with the highest prevalence reported among Dalit women and men. In contrast, comparatively lower prevalence was observed among Madhesi, Muslim and other caste/ethnic groups, particularly among men. Ecological variation was also evident, with high prevalence noted among individuals residing in the Terai and mountain regions. At the provincial level, Karnali province showed the highest prevalence of both outcomes among women and men, whereas Madhesh province demonstrated comparatively lower prevalence, especially among men.

Table 2 presents the adjusted odds ratios (AORs) with 95% confidence intervals for anxiety and depression among women and men, analyzed separately using the socio-ecological model. The findings demonstrate that factors operating at the individual, interpersonal, community, and societal levels were significantly associated with both outcomes, with distinct gender-specific patterns observed across the different levels of influence.

Among women, age was significantly associated with anxiety and depression, with those aged < 20 years showing higher odds of anxiety (AOR = 1.377, 95% CI 1.028–1.845) and depression (AOR = 1.348, 95% CI 1.005–1.808) compared to women aged 35–49 years.

Marital status was a strongly associated factor among both sexes, as widowed/divorced/separated women had higher odds of anxiety (AOR = 1.813, 95% CI 1.314–2.452) and depression (AOR = 1.850, 95% CI 1.327–2.578), while men in the same category showed even greater odds for anxiety (AOR = 3.980, 95% CI 1.942–8.158) and depression (AOR = 2.844, 95% CI 1.298–6.232), compared to those married/living with a partner. Poor self-reported health status is another strong associated factor, with women reporting very bad/bad health exhibiting higher odds of anxiety (AOR = 4.331, 95% CI 3.467–5.410) and depression (AOR = 5.125, 95% CI 4.137–6.349), while men also demonstrated markedly elevated odds for anxiety (AOR = 5.619, 95% CI 3.862–8.176) and depression (AOR = 5.057, 95% CI 3.426–7.464), compared to those reporting very good/good health. Similarly, smoking status was significantly associated with both outcomes, as women smokers had higher odds of anxiety (AOR = 1.615, 95% CI 1.179–2.210) and depression (AOR = 1.528, 95% CI 1.132–2.063), while among men, smoking was associated with increased odds of anxiety (AOR = 1.583, 95% CI 1.245–2.013) and depression (AOR = 1.354, 95% CI 1.077–1.701), compared to non-smokers.

Among interpersonal factors, wealth index was a significant associated factor of both anxiety and depression among women and men. Women from the poorer wealth quintile had higher odds of anxiety (AOR = 2.033, 95% CI 1.515–2.728) and depression (AOR = 1.848, 95% CI 1.387–2.462) compared to those from the richest wealth quintile. Similarly, men from the poorest wealth quintile showed increased odds of anxiety (AOR = 2.140, 95% CI 1.347–3.398) and depression (AOR = 2.342, 95% CI 1.451–3.781) compared to their counterparts in the richest wealth quintile.

Among societal factors, women residing in the Terai region had a higher odds of anxiety (AOR = 1.349, 95% CI 1.056–1.724) compared to those living in the Hill region, while among men, residence in the Terai was associated with increased odds of anxiety (AOR = 1.530, 95% CI 1.111–2.107) and depression (AOR = 1.534, 95% CI 1.083–2.174). Provincial differences were also evident, as women from Karnali Province had significantly higher odds of anxiety (AOR = 1.831, 95% CI 1.201–2.792) and depression (AOR = 1.667, 95% CI 1.126–2.469), while men from Karnali Province showed markedly greater odds of anxiety (AOR = 4.190, 95% CI 2.421–7.252) and depression (AOR = 6.079, 95% CI 3.289–11.236), compared to those from Madhesh province.

Discussion

This study reveals a complex and clearly gendered landscape of anxiety and depression among adults in Nepal, underscoring the critical importance of analyzing women and men separately to uncover unique socio-demographic and economic determinants. The clear differences in significant association between genders underscore that the drivers, contexts and potential patterns of anxiety and depression are fundamentally different for men and women, findings consistent with global literature on gender and substance use [41]. The WHO reports that gender inequalities, including limited socio-economic opportunities, caregiving responsibility, and exposure to violence and discrimination, contribute to a high risk of anxiety and depression among women globally [13]. Our analysis, therefore, justifies and is grounded in a gender-stratified approach, as a combined model would have obscured these pivotal differences and led to flawed conclusions.

The higher prevalence of mild anxiety and minimal depression among males in Nepal, compared to moderate-to-severe forms among females, may reflect different symptom expression and help-seeking behavior rather than a truly lower burden among men. Due to norms associated with masculinity and stigma, men are less likely than women to seek care or report their mental health issues [42]. As a result, the majority of men will only seek care once they are feeling severe symptoms of anxiety and depression [42], while more severe cases among women are more likely to be reported or diagnosed through health system contact [43–45]. In Nepal, evidence also suggests that gender norms and limited health service utilization among men contribute to under-recognition of progressing severity, resulting in clustering of reported symptoms at the mild level [45].

Individual factors

This study identifies that there are various individual factors associated with anxiety and depression among adults of Nepal. It included age, marital status, educational level, health program exposure, self-reported health status, recent sexual activity, and smoking status.

At the individual level, younger adults emerged as a significant determinant of both anxiety and depression among women. Adults below 35 years had significantly greater odds of anxiety and depression compared to those aged 35–39, indicating that younger cohorts who constitute the backbone of Nepal’s potential demographic dividend are particularly vulnerable. Consistent with these findings, globally, the burden of anxiety and depression has increased among adolescents and young adults, with highest prevalence in the 20–24 age group [46]. Similar findings have also been found in South Asian countries such as India and Bangladesh, where adolescents and young adults experience a high burden of anxiety and depression due to comparable socio-economic challenges, unemployment, academic pressure and rapid social transition [47]. This also aligns with global evidence that shows anxiety and depression due to educational pressure, employment insecurity, and transition to adult roles usually peak in early adulthood [48]. With the early onset of mental health conditions among adults, it impairs productivity and, in the long term, affects economic participation [48, 49], thereby preventing the realization of the full potential of the demographic dividend.

The association between marital status and mental health symptoms, including anxiety and depression, is very strong [50] for widowed, divorced, and separated [5, 16] adults compared to never-married adults. In South Asian cultures, marriage is considered a social protective institution that provides emotional support, social integration, and financial security [51, 52]. Conversely, marital dissolution often leads to social isolation, stigma, and financial insecurity, resulting in an elevated risk of anxiety and depression [53, 54]. Moreover, separation and divorce are increasing in Nepal, indicating changing social and family dynamics, which may contribute to the risk of psychosocial vulnerability among affected individuals [10]. In Nepal, the evidence shows that divorced or separated individuals had significantly higher odds of anxiety and depression among women, potentially due to emotional stress, economic insecurity, social isolation, and reduced social support systems [51, 52]. Therefore, the poor mental health of widowed, divorced, and separated adults during their prime working age groups may subsequently hinder labour force participation, productivity, social participation and stability, and human capital development, thereby limiting Nepal’s ability to effectively harness its demographic dividend.

Level of education showed a significant association with anxiety among men, particularly among adults with basic and secondary or higher education, indicating greater odds of anxiety compared with those with no formal education. While higher education is viewed as a protective factor for mental health, in fact, it seems to be in keeping with the growing trend of evidence from low- and middle-income countries that education paradoxically is associated with psychological distress when economic opportunities fail to match expectations [49, 55]. The phenomenon aligns with the status inconsistency hypothesis, where a mismatch between educational achievement and occupational or economic outcomes contributes to anxiety and emotional strain [56, 57]. The same experiences are reported where educated individuals face competitive labour markets and social pressure to succeed, leading to a greater likelihood of developing anxiety and depression [58, 59], particularly among men. Rapid social and educational transitions without parallel growth in labour market expansion may lead to worsening mental health issues among the educated young adults of LMICs [59, 60]. In the context of Nepal, higher education attainment does not necessarily guarantee stable and secure job opportunities and livelihood and educated young adults may experience heightened stress and feel under strain due to unemployment, job insecurity, and unmet aspirations, which may ultimately hinder their productive management and limit the country’s ability to fully harness the demographic dividend.

The strongest association with mental health outcomes was seen in self-reported health status. Adults who reported moderate or poor health status are more likely to have anxiety and depression than those who reported their health as good, with the odds ratio exceeding four times in those reporting poor health status. These findings reinforce the bidirectional relationship between physical and mental health, whereby chronic illness increases psychological distress, while mental illness worsens physical outcomes [53]. Therefore, integrated strengthening of primary health care services [48] that incorporate both mental and physical health is essential to improve the mental well-being of adults and maximizing the potential benefits of the demographic dividend.

Behavioral factors also played a significant role. Smoking was found to have a consistently significant and increased likelihood of experiencing anxiety and depression among adults. This evidence strongly supports that tobacco use and mental disorder are closely linked, with nicotine dependence both amplifying and masking psychological symptoms [48]. Therefore, smoking cessation programs in Nepal should include mental health screening and counselling to improve outcomes [61, 62]. Smoking and untreated mental health problems among adults can weaken the workforce efficiency of adults and ultimately diminish the economic gains expected from the demographic dividend.

The significant association of age, marital status, education, and smoking with anxiety and depression indicates various individual health and behavioral factors may affect productivity and workforce participation among adults, thereby limiting Nepal’s demographic dividend potential.

Interpersonal factors

The socio-economic status of the family is an important associated factor for mental health [63]. Adults from lower-wealth families, there was a significantly increased risk of anxiety and depression compared to those from the richest/richer wealth families [5, 16, 62, 64]. The pattern of mental health associated with the level of wealth is also seen in many previous studies on poverty and mental health. A similar study also reported that an individual with a higher wealth index is associated with a lower level of anxiety and depression, indicating an inverse relationship between socio-economic status and mental health issues [62]. The finding of inverse association between mental health and wealth indicates that wealth creates resistance against life stressors and provides material stability, whereas poverty magnifies anxiety and depression symptoms [65]. In practical terms, financial insecurity would appear to cumulatively exacerbate concerns about basic needs like food [66], shelter and education, leading to increased levels of distress [67]. Therefore, because of an increase in financial insecurity, there is heightened emotional distress because of the compounding effect of worrying about a person’s living situation. Ultimately, a lower wealth index is closely linked with mental health among adults, which can limit productivity and labour force participation and hinder the realization of the demographic dividend.

In addition to wealth quantile, decision-making for health care and health insurance coverage were also found significantly associated with anxiety among men. These factors may be associated with mental health through differences in financial security, access to health services, and individual control over health care-related decisions. A previous study has also indicated that limited decision-making autonomy and lack of health insurance coverage are linked with poor mental health outcomes, largely due to limited access to health services, financial insecurity and reduced social support [68, 69]. This highlights the importance of household-level socio-economic factors being essential to effectively convert Nepal’s demographic dividend into improved mental health and overall human development outcomes. The association of wealth index, decision-making for health care, and health insurance coverage with mental health among adults highlights how household economic insecurity creates psychosocial stress which adversely impacts human capital productivity necessary for harnessing the demographic dividend.

Community factors

Caste and ethnic group were strongly associated with anxiety and depression at the community level [5, 64], with structural inequality being evident. Similar patterns have also been reported, where Dalit and some Janajati caste and ethnic groups showed higher odds of anxiety and depression, particularly among men, with the Dalit ethnic group being at greater risk [5, 64]. This reflects deep-rooted social exclusion, as qualitative studies note Dalits in Nepal face severe discrimination, stigma and social isolation, which has a direct impact on the mental well-being of individuals from that caste and ethnic group [70]. These findings suggest that the disadvantage based on the caste-based system in Nepal is a source of community-level stress that should be addressed. Comprehensive anti-discrimination and community outreach services for Dalit and some Janajati groups would help mitigate this mental health inequity [71]. When a large segment of adults from these caste and ethnic groups experience untreated anxiety and depression, their capacity for sustained labour force participation, skill acquisition, and productivity is compromised. The significant effect of caste and ethnic group among men indicates community-level social inequalities; as a result, caste-based mental health disparities may contribute to poor mental health and limit productive social participation and risk turning Nepal’s demographic opportunities into a demographic disaster.

Societal factors

Geography was another major structural determinant of anxiety and depression in Nepal [16]. Significant disparities were observed across ecological regions and provinces [50], indicating that structural inequalities meaningfully shape mental health risk. The individuals living in the Terai regions and those who live in mountain regions experience considerably higher levels of anxiety and depression than the individuals living in the hilly regions. Specifically, individuals living in the mountain region experience considerably greater odds of mental health issues [72], with the greatest level among men. These regions are characterized by environmental and economic stressors, such as climate vulnerability, physical isolation [73, 74], and limited transport and access to health and mental health services. In contrast, the Terai region has unique challenges, including high population density, temperature stress [75], seasonal flooding [18, 75], and economic insecurity [76], all of which contribute to additional psychological stress. Disparities between provinces were also very clear; compared to Madhesh Province, the provinces of Koshi, Bagmati, Lumbini, and most significantly, Karnali Province [50] had significantly higher odds of anxiety and depression, which was approximately two to six times higher in Karnali than in Madhesh Province. These mental health disparities pose a serious constraint from the perspective of harnessing the demographic dividend of Nepal. A healthy, dynamic, and productive working-age population is a prerequisite for demographic dividends in all provinces of Nepal. When a large proportion of adults in these provinces is suffering from untreated mental health issues, it undermines their ability to participate in education, employment, and community development. Persistent geographical disparities in mental health can thus reinforce spatial poverty traps and impact national economic productivity performance, ultimately limiting Nepal’s ability to effectively harness the demographic dividend.

Thus, to harness the demographic dividend, targeted policies addressing provincial and ecological disparities in mental health access and regional development are essential to reduce health inequalities and strengthen the productivity of the working-age population.

Conclusions

This study demonstrates that anxiety and depression in adults of Nepal are affected by individual, family, community, and societal-level factors that have implications for Nepal’s opportunities to fully realize its demographic dividend. Mental health risks were unevenly distributed across population groups but were strongly patterned by age group, marital status, self-reported health status, smoking habit, economic status, health program, caste and ethnic group, and geographical location. These patterns highlight that mental health in Nepal is not just an individual health problem but also a systemic challenge shaped by structural inequalities, community-based stigma, poverty, and uneven development across geography.

These inequalities reflect long-standing patterns of social exclusion, limited-service access, and limited economic opportunities, which collectively undermine psychological well-being and human capital formation, as reflected in the demographic dividend perspective. Mental health issues among working-age adults may hinder Nepal’s demographic dividend by reducing labour force participation and human capital development. To promote adult mental health well-being and maximize demographic dividend in Nepal, it is essential that policymakers be aware of policies allowing for the establishment of equitable mental health services in primary healthcare settings and social protection systems while providing targeted intervention to economically disadvantaged, socially marginalized and geographically underserved adult populations. Aligned with the socio-ecological model, improving mental health in Nepal required multilevel action, including national and provincial policies. Addressing mental health should therefore be recognized as both a public health priority and an economic development strategy. Strengthening community-based mental health service, integrate mental health screening, counselling, and psychosocial support into social protection and poverty-reduction with particular attention to socio-economically disadvantage groups, and widowed/divorced/separated individuals. Addressing mental health as a key component of human capital development is essential for harnessing the demographic dividend. Thus, to fully leverage the country’s demographic dividend, Nepal must invest in culturally sensitive, context-specific, and region-specific mental health vulnerabilities through geographically targeted investments are essential to ensure equitable, acceptable, and effective mental health services among Nepal’s diverse populations.

Author contributions

OCT: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Supervision, Validation, Writing – original draft, Writing – review & editing YBG: Conceptualization, Data curation, Methodology, Supervision, Validation, Writing – review & editing DVS: Conceptualization, Methodology, Supervision, Validation, Writing – review & editing SP: Conceptualization, Formal analysis, Methodology, Supervision, Validation, Writing – review & editing.

Funding

This study did not receive any grants from any funding agencies.

Data availability

https://dhsprogram.com/data/dataset/Nepal_Standard-DHS_2022.cfm?flag=1. To access the datasets, Users must be logged in.

Declarations

Ethics approval and consent to participate

This study used publicly available, fully anonymized secondary data from the Demographic and Health Survey Program. Permission to access and use the dataset was obtained from DHS Program/ICF. An approval letter has been obtained from the DHS Program to use the datasets. The original DHS survey protocols was reviewed and approved by the Nepal Health Research Council (Reference #678:494/2021 P on September 30, 2021) and the ICF Institutional Review Board (ICF Project #180,657.0.001.NP.DHS.01 on August 25, 2021). As this study involved secondary analysis of de-identified data, no additional ethical approval was required. Written informed consent was obtained from all the participants by the original Nepal Demographic and Health Survey before data collection. For participants child and adolescent, informed consent was obtained from a parents or guardian in accordance with the DHS protocol. As this study was based on secondary analysis of fully anonymized data, no additional informed consent was required.

Consent for publication

Not applicable. This study involved secondary analysis of anonymized publicly available data and does not contain any identifiable individual information.

Competing interests

The authors declare no competing interests.