We conducted a descriptive, cross-sectional analysis of baseline data collected from participants enrolled in the parent study, an ongoing population-based, longitudinal cohort study investigating risk factors for cardiovascular disease in Dhulikhel, Nepal.⁹ Baseline data for participants was recently collected for a random sample of 1,073 adults age 18 years or older.

All adults age 18 years and older living in Dhulikhel, Nepal for at least six months were eligible for the parent study. A third of the households residing in each of the nine administrative divisions of Dhulikhel were selected for collection of baseline data. Exclusion criteria included: being pregnant at the time of data collection, temporary residents living in Dhulikhel less than six months or staying in hostels/motels, individuals who are mentally challenged and unable to respond, and individuals who refuse to participate. All participants provided written informed consent or a thumbprint for illiterate participants.

Study protocol for the parent study was reviewed and approved by the Institutional Review Committees of both the hospital in Nepal and the study’s originating university. The study protocol for this secondary data analysis was also reviewed and deemed not to be human subjects research by the originating university, as the authors did not participate in collecting individual data from enrolled patients, data were anonymized upon collection, and authors did not have access to identifying information.

The baseline visit included a standardized questionnaire administered through electronic tablet, physical exam, and laboratory testing. All data enumerators were trained in interviewing and measuring variables. Demographic data, including information regarding age, sex, ethnicity, marital status, education, economic status, diet, tobacco use, alcohol consumption, and physical activity were collected. Height, weight, and waist and hip circumference were measured using standardized instruments. Body Mass Index (BMI) was calculated as kg/m² and categorized as underweight (<18.5 kg/m²), normal (18.5–22.9 kg/m²), overweight (23–27.4 kg/m²), and obese (≥27.5 kg/m²), according to the World Health Organization (WHO) expert consultation guidelines for South Asia.¹⁰ Using a standard digital blood pressure machine (Microlife, Switzerland), blood pressure was measured three times in a seated position on the right arm over loose clothing. The mean of the three measurements was calculated and used for analysis. Hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg or receiving medication for hypertension.

The Global Physical Activity Questionnaire (GPAQ) developed by the WHO was used to assess several components of physical activity, including intensity, duration and frequency in three domains of occupational physical activity, transport-related physical activity, and physical activity during leisure time^.11 The WHO recommends that during one week, adults should do at least 600 MET-minutes of physical activity, including activity for work, during transport, and leisure time.¹¹ Thus, the physical activity variable was dichotomized with a cutoff of 600 MET-minutes of physical activity per week for analysis.

The Center for Epidemiologic Studies Depression Scale (CES-D) was used to assess depressive symptoms. The CES-D is a 20-item measure that asks participants to rate how often over the past week they have experienced symptoms associated with depression such as restless sleep, poor appetite, sadness, and hopelessness. There are four response options – rarely (<1 day), some (1–2 days), occasionally (3–4 days), and most (5–7 days). Items are scored 0–3 with an instrument range of 0–60. The CES-D represents an accurate and valid measurement of depression with high internal consistency and is frequently used in clinical and research settings to determine depression symptomatology in the general population.^12,13 Prior studies in Nepal have used the CES-D questionnaire to examine depressive symptoms among leprosy-affected patients and female sex workers.^14,15 Higher scores indicate greater frequency of depressive symptoms and correlate with an increased risk of clinical depression. Depression scores were dichotomized with a cutoff score of 16 or greater on the CES-D for analysis.

STATA 13 (Statacorp LP, College Station, TX) was used for data analysis. Descriptive statistics were presented in frequency and percentages to identify the distribution of risk factors for depressive symptoms among socio-demographic characteristics. Baseline characteristics were compared between high and low depressive symptoms groups using Pearson’s chi-squared tests for categorical variables and t-tests with unequal variance for continuous variables. All tests were two-tailed and p≤0.05 was considered statistically significant. Multiple logistic regression was used to examine associations between the binary CES-D depressive symptom score variable by gender, age, education level, marital status, body mass index, and hypertensive status. Potential confounders including gender, age, ethnicity, religion, education, and marital status were adjusted for in the regression analysis.

Table 1 shows the socio-demographic characteristics of participants by gender. Among 1,073 participants, the mean age was 40.3 years (SD: 16.4), and 58.4% were female. 49.2% of the participants were of Newar ethnicity, and 84.8% self-reported their religion as Hindu. Women were much more likely to have no formal education (42%) and to be separated or widowed (9.4%) compared to men (17.3% and 1.1%, respectively). Women were also more likely to be obese (20.1%) compared to men (13.7%). More males were current smokers (35.7%) compared to females (14.2%). Men were more likely to be heavier drinkers, with 29.6% reporting having >3 drinks/week compared to 7.8% for women. Men were also more likely to be hypertensive (39.5%) compared to women (23.1%).

Table 2 shows the socio-demographic characteristics of participants by depression status. The mean CES-D score in the sample was 11.7 (SD: 5.3), and the estimated prevalence of depressive symptoms was 21.3%. Women were more likely to have depressive symptoms (24.1%) than men (17.3%). Increasing age, especially age above 60 years old, those with no formal education, and those who were separated or widowed also represented significant risk factors for depressive symptoms. Indeed, when examining the age distribution of participants with depressive symptoms, the prevalence of depressive symptoms remains stable at approximately 20% for adults until about age 60, when it begins to steadily increase to almost 50% at age 90 (Figure 1). Thus, increasing age above 60 years old was associated with a higher prevalence of depressive symptoms. Having no formal education was also associated with higher risk of depressive symptoms, as 28.8% of those with no formal education had depressive symptoms, compared with only 17.2% among those who had less than high school and 18.8% who had high school level education or more. Being separated or widowed was also significantly associated with a higher risk of depressive symptoms, as 37.5% of those who were separated or widowed had depressive symptoms, compared to only 19.3% among those who were currently married.

The results of the multiple logistic regression analysis to determine the association between socio-demographic and lifestyle factors and depression symptoms are shown in Table 3. In unadjusted models, a statistically significant increased risk of depressive symptoms was associated with female gender (P=0.007), increasing age above 60 years (P=0.002), having no formal education (P<0.001), being separated/widowed (P=0.002), and physical inactivity (P<0.001).

In the adjusted model, increasing age above 60 years (P=0.01), having no formal education (P= 0.01), and physical inactivity (P<0.001) were associated with a statistically significant increased risk of depressive symptoms. In the adjusted model, for every one-year increase above age 60, odds of having depressive symptoms were higher by 5%. Having no formal education was associated with 2.2 times higher odds of depressive symptoms when compared to those with at least a high school education. Physical activity as a risk factor remained relatively unchanged in the adjusted model, as recommended levels of physical activity were associated with 63% lower odds of depressive symptoms. Risk factors that were found to have no statistically significant association with depressive symptoms after adjustment for demographic variables included gender, ethnicity, religion, tobacco use, alcohol use, BMI, and hypertensive status.

This analysis of adults age 18 and older in Dhulikhel found an overall prevalence of depressive symptoms of 21.3%. Statistically significant risk factors for depressive symptoms in the adjusted model included having no formal education, increasing age above 60 years old, and physical inactivity. Being obese was associated with decreased risk of depressive symptoms. Marital status, gender, ethnicity, religion, tobacco and alcohol use, and hypertension status were not statistically significantly associated with depressive symptoms in the adjusted model.

The prevalence of depressive symptoms of 21.3% found in this sample is relatively consistent with previously reported studies, which range from 11.7% prevalence of depression among 2,100 Nepalese adults age 18–65,¹⁶ to 15% among 321 hypertensive patients in Kathmandu,⁵ to 27.5% among 720 adults in five districts in Nepal.¹⁷ This prevalence of depressive symptoms is also four times greater than the estimated global prevalence of major depressive disorder of 4.7%, indicating significant mental health burden in this population.¹⁸

Low levels of education and illiteracy have been described as risk factors for depression in prior studies among type 2 diabetics and hypertensive patients, respectively, attending tertiary care centers in Kathmandu, Nepal.^5,7 Older age has also been shown in prior studies to be a risk factor for depressive symptoms in Nepal. A study among 321 patients with hypertension attending a tertiary care center in Kathmandu, Nepal found increasing depression scores with each additional decade of age, with 29% of those age above 64 years having depressive symptoms, compared to only 11% of those who were age 45–64 years and 24–44 years.⁵ Moreover, Joshi et al. found that among 379 patients with type 2 diabetes attending three treatment centers in Kathmandu, there was a positive association between depression and those age 61–70 years.⁷ Kohrt et al. also found a similar trend in a post-conflict setting in northwestern Nepal, with increasing depression scores associated with increasing decades of age, with the highest risk group being those age 52–80.¹⁹ A community-based assessment among 165 adults age 60 and above living in the Kathmandu Valley found a prevalence of depression of 29.7%.²⁰

Physical inactivity has also been shown to be a risk factor for depression and is known to mediate the relationship between depression and mortality among community-dwelling adults in a multicenter study in the US among adults age 65 and older.²¹ Prior studies show that gender may modify the relationship between physical activity and depression, as Gautam et al. showed that among 489 community dwelling adults age 60 and above in Nepal, physical activity was associated with decreased depressive symptoms in men but not in women.²²

We also found that being obese (BMI ≥27.5) was associated with lower odds of depressive symptoms by 33% when compared with participants of normal weight. This represents a new finding that has not been reported in previous data from Nepal. Results from prior studies globally have been mixed, as those conducted in the U.S. and Europe have found a positive association between obesity and depression.^23,24 Nevertheless, data from Asian countries such as Japan, Taiwan, and China have shown a negative association between obesity and depression.^25–27 Still others, such as a study in Korea, found a U-shaped distribution, where the highest level of depressive symptoms was found among the underweight (BMI <18.5), followed by the severely obese (BMI ≥30) and then the obese (BMI 25–30).²⁸ Our data showed that compared to participants with normal BMI, the highest risk of depressive symptoms was found among the underweight (BMI <18.5), followed by overweight (BMI 25–29.9) and then obese (BMI ≥27.5).

One potential explanation for this finding is that being obese may be a sign of higher socioeconomic status (SES) within the caste system. Data from India has shown that high caste is positively associated with obesity, mainly through increased wealth.^29,30 Moreover, a prior study examining the relationship between BMI and SES in 37 lower- and middle-income countries found that increased BMI was associated with increased wealth and higher SES.³¹ In this way, being obese may signify having higher income and greater access to food, thus less likelihood to have depressive symptoms.

Another explanation could be due to differing perceptions of body weight, as studies of body weight among women in South India have shown that women tend to have discrepancies in self-perceived body weight and actual body weight.³² In this way, women who are overweight or obese perceive themselves as normal weight.³² This finding is an important consideration, as obesity is a risk factor for many NCDs, including cardiovascular disease, hypertension, and diabetes. Our data, however, show that those who are underweight are most at risk for depressive symptoms and being obese was associated with a decreased risk of depressive symptoms.

While female gender has been associated with depressive symptoms in Nepal in previous studies,^5,7 another study conducted in Nepal showed no association between gender and depressive symptoms among a cohort of 385 patients with type 2 diabetes attending tertiary care centers in Kathmandu, Nepal.⁶ This analysis found that although female gender was associated with depressive symptoms in unadjusted models, after adjustment for other demographic variables it was no longer statistically significant (OR:1.22, 95% CI: 0.87–1.71). This was likely due to education confounding the association between female gender and depressive symptoms given that 42% of females had no formal education and having no formal education was strongly associated with depressive symptoms.

Our study did not show a significant association of alcohol consumption and tobacco use with depressive symptoms. Although some studies among type 2 diabetics in Nepal have shown an association between tobacco use and depressive symptoms⁵ other studies conducted among type 2 diabetics and hypertensive patients in Kathmandu, Nepal have not shown an association between either tobacco use or alcohol use and depressive symptoms.^6,7 In this analysis, the lack of a statistically significant association may be reflective of the fact that both tobacco use and alcohol use were self-reported measures and not many individuals with depressive symptoms reported heavy smoking or drinking, as almost 70% of those with depressive symptoms had never smoked, and 81% had less than one alcoholic drink per week.

Hypertension has also been associated with depressive symptoms in previous studies in Nepal.⁵ However, our analysis did not show an association between hypertensive status and depressive symptoms. This could be due to the fact that our definition of hypertension in the study included individuals taking anti-hypertensive medications. Taking anti-hypertensive medications has been associated with lower risk of depression.⁵ Thus, the fact that some hypertensive individuals were receiving treatment for their hypertension could have confounded the relationship between hypertension and depressive symptoms.

Our study also did not show a significant association between ethnicity and depressive symptoms. While some prior studies that have shown an association between low-caste ethnicity groups and higher prevalence of depression,^19,33 others have not found a significant association between ethnicity and depressive symptoms.^5–7 Our finding is likely due to the fact that the majority of our sample belonged to the Brahmins and Newars, which are the higher income ethnic groups.³⁰ Thus, with a higher socio-economic status, they may have been more likely to have access to healthcare facilities to start treatment for chronic conditions and less likely to develop associated depressive symptoms.

This study has several limitations. The cross-sectional design of this study prevents the ability to establish causal associations or directionality between risk factors and development of depressive symptoms. Since subjective self-reported behaviours were reported in the questionnaire data, participants may have not disclosed the full extent of their depressive symptoms due to stigma. Additionally, the questionnaire format may not have captured other unmeasured variables that could be contributing to the development of depressive symptoms, and the estimated prevalence from this analysis likely underestimates the true prevalence of depression in the community. This analysis is also specific to community members in Dhulikhel, a suburban town with a tertiary healthcare centre, and thus these results may not be generalizable to other populations in Nepal due to high levels of ethnic and geographic variation within the country.

To our knowledge, this represents one of the first suburban, community-based assessments of depressive symptoms among a general population of adults age 18 and older in a non-conflict zone of Nepal. The random sampling technique, large sample size, and extensive collection of information on demographic and clinical risk factors represent strengths of this study in determining the prevalence and risk factors for depressive symptoms in this cohort.

Major depressive symptoms are highly prevalent in Nepal and a significant treatment gap exists between the population’s mental health needs and appropriate accessible care. Integrated programs specifically targeting the uneducated, the elderly, and the physically inactive should be explored to address this high burden of depressive symptoms.