Prevalence and determinants of multimorbidity among resettled Bhutanese older adults in Ohio, USA

Abstract

Background

Multimorbidity, the presence of two or more chronic conditions, is a significant global health issue, especially among older adults. Accelerated aging and a myriad of stressor factors related to displacement and socio-economic disadvantage position profoundly exacerbate physical health and well-being problems among refugees, including resettled Bhutanese older adults. Despite these risks, research on multimorbidity burden and its determinants among the resettled Bhutanese older adults is limited. This study aims to assess the prevalence of single and multiple chronic conditions (multimorbidity) and identify associated factors among resettled Bhutanese older adults in Ohio.

Methods

This cross-sectional study involved 276 resettled Bhutanese aged 55 and older from four major cities in Ohio—Akron, Cincinnati, Cleveland, and Columbus, recruited through snowball sampling. Multimorbidity, the dependent variable, was defined as the presence of two or more chronic conditions—specifically hypertension, arthritis, diabetes, high cholesterol, heart disease, chronic obstructive pulmonary disease, kidney disease, or cancer—based on participants’ self-reports of diagnoses or medication use. Binary logistic regression was used to identify associated factors.

Results

Common conditions included hypertension (63.0%), high cholesterol (43.5%), diabetes (42.8%), and arthritis (42.0%). Notable combinations were hypertension with high cholesterol (36.6%), diabetes (33.3%), and arthritis (31.5%). The study found a 62.0% prevalence of multimorbidity. Higher odds of multimorbidity were linked to identifying as ascribed to Hindu religion (aOR:2.50, 95% CI:1.26–4.96) and experiencing depressive symptoms (aOR:3.13, 95% CI:1.46–6.72). Employed individuals had 85% lower odds of multimorbidity (aOR:0.15, 95% CI:0.04–0.58).

Study implications

This study provides valuable insights into the high burden of chronic diseases and multimorbidity among resettled Bhutanese older adults in Ohio. The study emphasizes the need for targeted health research focused on chronic disease prevention and lifestyle management interventions aimed at preventing chronic conditions.

Introduction

Multimorbidity, defined as the co-occurrence of two or more chronic medical conditions in an individual [1, 2], is a significant global concern, especially among older populations. A recent review highlighted a global multimorbidity prevalence of 37.3% (95% CI: 34.9–39.4%) in community settings, with an even higher prevalence of 51.0% (95% CI: 44.1–58.0%) among individuals over 60 years old [3]. This trend highlights the growing challenge of multimorbidity in aging populations. Previous research have documented that the increasing prevalence of multimorbidity among the older population is attributed to the acceleration of the aging process [3–6]. Additionally, life course studies suggest that the development of multimorbidity results from the accumulation of adverse lifestyle and socioeconomic vulnerabilities over one’s lifespan, with accumulated stressors manifesting as adverse health conditions in later life [7–10]. While multimorbidity is a concern across the general population, it is particularly pronounced among older resettled populations [11, 12]. This group is subjected to profound stressors due to displacement and resettlement, which can exacerbate health conditions and increase vulnerability to developing multiple chronic illnesses. However, data on the prevalence of multimorbidity within these groups remain scarce.

A recent systematic review and meta-analysis, including 38 studies, found that the prevalence of multimorbidity among migrants, refugees, asylum seekers, and displaced persons varied widely, from 0.6% to 75.0% [12]. The review noted that multimorbidity generally increased with age. While migrants overall often had a lower prevalence of multimorbidity compared to the native-born population, refugees specifically experienced higher rates. Consistently, a Canadian study on health and life satisfaction found that older refugees face significant health challenges, with a multimorbidity prevalence approximately 8% higher than the average for other immigrant groups [11]. Additionally, older refugees increasingly reported worse health compared to the general older adult population, likely due to a higher prevalence of chronic conditions [11]. Evidence indicates that social, political, or economic marginalization often leads to poverty and homelessness, increasing the likelihood of adverse health outcomes and multimorbidity among refugees [13].

Given the limited literature on multimorbidity among older refugee populations, this study focuses on exploring multimorbidity among resettled Bhutanese older adults. The history of Bhutanese refugees, who are ethnically Nepali, began in the 1990s with the enactment of Bhutan’s “One Nation, One People” policy in 1989 [14]. This policy, which enforced Drukpa Buddhist culture, marginalized the Hindu Lhotshampa community by banning the Nepali language and Hindu practices and imposing severe penalties, including imprisonment, for non-compliance. As a result, approximately one-sixth of Bhutan’s population was displaced and sought refuge in Nepal, where they lived in refugee camps for about two decades. In 2007, the United Nations High Commissioner for Refugees (UNHCR) offered resettlement opportunities in eight countries, including in the USA [15]. Since 2008, over 100,000 Bhutanese refugees have been resettled abroad, with 85% settling in the USA [14, 15]. The largest concentrations of Bhutanese refugees in the USA are in Pennsylvania, Texas, New York, Georgia, Ohio, Arizona, Colorado, Washington, Virginia, and North Carolina [16]. Local Bhutanese communities in Ohio have observed a trend of individuals moving from other states, making Ohio home to the largest Bhutanese population both in the United States and outside Bhutan [17].

Existing research on the resettled Bhutanese population has largely focused on infectious diseases and mental health, often overlooking the broader spectrum of health challenges faced by resettled Bhutanese older adults [16, 18]. Most studies have examined chronic diseases in silos within this group. For example, during their resettlement in the US, the Centers for Disease Control and Prevention (CDC) published a health profile of Bhutanese refugees based on medical evaluations [16]. This report revealed that chronic diseases, such as hypertension, asthma, emphysema, seizure disorders, learning impairments, mental disorders, diabetes, thyroid issues, and visible disabilities, were prevalent [16]. Another clinical study conducted in Atlanta found that Bhutanese adults aged 65 and older had an 11% higher prevalence of hypertension compared to younger Bhutanese adults [19]. While both studies offered valuable insights into these conditions, they did not explore multimorbidity within this population. Consequently, there is a significant gap in the literature, highlighting the need to update chronic disease estimates and provide insights into multimorbidity among older Bhutanese in the USA. Thus, this research aimed to assess the prevalence of single and multiple chronic conditions (multimorbidity) and the factors associated with multimorbidity among resettled Bhutanese older adults in Ohio.

Materials and methods

Study design, participants, and sampling

This study used a cross-sectional design and was conducted from January to June 2022 among Bhutanese adults aged 55 and older in Ohio, USA. This age classification is consistent with previous studies on refugee populations, which generally define older individuals as those aged 55 and above [11, 20]. Such a classification reflects global variations in life expectancy, which affect how old age is defined and measured worldwide [21]. The detailed description of study methods is also presented elsewhere [22].

Due to the lack of a sampling frame, snowball sampling was utilized to identify potential participants. This method involved collaborating with local community leaders and Bhutanese organizations in the selected cities. Snowball sampling is commonly used in research when finding study participants is challenging, and the researcher has limited knowledge about the population of interest [23]. A total of 276 participants were recruited from four major cities in Ohio, where substantial Bhutanese resettlement populations are present [24]: Akron (n = 28, 10.1%), Cincinnati (n = 53, 19.2%), Cleveland (n = 75, 27.2%), and Columbus (n = 120, 43.5%). The study excluded non-refugee Bhutanese immigrants, individuals with speech, language, or hearing disorders, and those residing in nursing homes or assisted living facilities.

Data collection

Pre-validated scales were used to prepare the study tool for this study [25–29]. The questionnaire was translated into Nepali and pretested among older Bhutanese adults in Cincinnati, resulting in only minor typographical and wording corrections without significant content changes. Data collection was conducted through a combination of in-person and telephone interviews, as well as an online survey administered via Qualtrics [30], a secure online survey and research platform. Such an approach accommodated participants’ preferences and pandemic-related restrictions. The surveyors, who were proficient in Nepali and had prior experience in health and social research, conducted these interviews using the Nepali version of the questionnaire. They received orientation on the study’s objectives, survey methodology, study tools, and the use of Qualtrics for data entry.

Study measures

Dependent variable

Multimorbidity was defined as the co-occurrence of two or more chronic diseases and served as the dependent variable. Participants were asked whether they had ever been diagnosed by a healthcare professional or were currently taking medication for any of the following eight chronic conditions: hypertension, arthritis, diabetes, high cholesterol, heart disease, chronic obstructive pulmonary disease, kidney disease, and cancer. We focused on chronic diseases that are widely recognized as the main chronic conditions among Bhutanese refugees and in the US, consistent with prior reports and literature [31, 32]. While the survey tool allowed participants to indicate additional conditions, most of the additional responses were either acute or pain-related conditions, which did not meet the definition of chronic diseases. For this reason, they were excluded from the analysis. The total number of self-reported chronic diseases for each participant was calculated, ranging from 0 to 8. This cumulative count was then categorized into a binary format: ‘Yes’ for multimorbidity (indicating two or more chronic diseases) and ‘No’ for those reporting either none or only one chronic disease.

Independent variables

Sociodemographic variables

Sociodemographic variables in the study included: age (categorized as 55–64, 65–74, 75 + years), gender (male/female), marital status (married/without a partner), religion (Hindu/other than Hindu), received formal education (Yes/No), English proficiency (ability to read/write/speak in English: Yes/No), employment status at the time of the survey (Yes/No), years spent in a refugee camp (< 20 or ≥ 20 years), and years living in the USA (≤ 10 or > 10 years).

Health and healthcare utilization

Self-reported health status was assessed using a single-item question with Likert scale responses: poor, fair, good, very good, and excellent. Due to low response counts (only four participants reported excellent health), the ‘very good’ and ‘excellent’ categories were combined. Other health-related variables examined included access to and utilization of health services: time since the last health facility visit (never visited or visited more than a year ago/visited within the past year), having a regular doctor for healthcare visits (yes/no), and type of health insurance (Medicaid, Medicare, dual coverage, employment-based, no insurance, or unknown).

Health behaviors

Health behaviors were assessed through self-reported smoking status, alcohol consumption, and tobacco use, each categorized as never, former, or current. Physical activity levels were also self-reported and categorized as high, medium, or low.

Depressive symptoms

To screen for depressive symptoms, the study utilized a pre-validated Nepali version of the Geriatric Depression Scale (GDS-15) [25]. The GDS-15 exhibited high reliability, with a Cronbach’s alpha of 0.85 in this study. The scale assesses various depressive symptoms experienced over the past week, including sadness, loss of interest and energy, feelings of emptiness, helplessness, and guilt. It comprises 15 items, each with binary responses (1 = “Yes” and 0 = “No”). The total score, which ranges from 0 to 15, is calculated by summing the responses to the items, with a score of 5 or higher indicating the presence of depressive symptoms [33].

Life satisfaction

Life satisfaction was assessed using the 5-item Satisfaction with Life Scale (SWLS-5) [26], which has shown strong convergent validity with other measures of subjective well-being [34]. In this study, the SWLS-5 demonstrated high reliability, with a Cronbach’s alpha of 0.89. The scale comprises five statements that evaluate various aspects of life, including life ideality, conditions, satisfaction, and personal goals. Participants rated each statement on a 7-point Likert scale, from 1 (‘very strongly disagree’) to 7 (‘very strongly agree’). The total score, derived by summing the ratings for all five items, ranges from 5 to 35. Scores were then categorized into two groups: ‘Satisfied’ (scores of 20 and above) and ‘Dissatisfied’ (scores below 20), consistent with previous research [35, 36].

Religious coping

Religious coping was assessed using the Hindu Religious Coping Scale [27]. This scale has demonstrated good internal consistency, with alpha coefficients above 0.80 in previous studies [27]. In this study, Cronbach’s alpha was 0.88, indicating high internal consistency. The scale used a 4-point Likert response format, ranging from 1 (haven’t been doing this at all) to 4 (been doing this a lot). A cumulative score was derived by summing responses across 17 items [27], with higher scores reflecting greater use of reliance on religious coping strategies grounded in Hindu beliefs and practices. Due to the non-normal distribution and skewness towards higher coping levels, the total score was categorized into three groups based on tertiles: low, moderate, and high levels of coping.

Social support

Social support was assessed using the Multidimensional Scale of Perceived Social Support (MSPSS), a validated instrument measuring perceived support from family, friends, and significant others [29]. Previous studies have demonstrated strong reliability for the MSPSS, with Cronbach’s alpha coefficients ranging from 0.87 to 0.94 [37]. In this study, the MSPSS also showed high internal consistency, with a Cronbach’s alpha of 0.92. The MSPSS comprises 12 items rated on a 7-point Likert scale, from 1 (“very strongly disagree”) to 7 (“very strongly agree”). The mean MSPSS score was calculated by averaging responses across all 12 items, resulting in scores ranging from 1 to 7. According to the categorization guidelines proposed by the developers [38], mean scores between 1 and 2.9 were initially categorized as indicating low support, scores between 3 and 5 as moderate support, and scores between 5.1 and 7 as high support. However, due to the observed distribution in this study (with only one individual reporting “low support”), the categories of “low support” and “moderate support” were combined, resulting in a two-level categorical variable for the analyses.

Resilience

Resilience was assessed using the pre-validated Nepali version of the Connor-Davidson Resilience Scale (CD-RISC). This scale has demonstrated strong reliability, with a Cronbach’s alpha of 0.89 in previous research [28]. In this study, it exhibited even greater internal consistency, with a Cronbach’s alpha of 0.96. The CD-RISC comprises 10 items rated on a 5-point Likert scale, ranging from 0 (“Not true at all”) to 4 (“True nearly all the time”). Cumulative scores range from 0 to 40, with higher scores reflecting greater resilience. Due to the non-normal distribution and skewness towards higher resilience levels, the total scores were categorized into three groups based on tertiles: low, moderate, and high resilience.

Data analyses

The data were analyzed using SAS version 9.4 software [39]. Given that all variables were categorical, descriptive statistics, including frequencies and percentages, were computed. Bivariate analysis was conducted using chi-square tests or Fisher’s exact tests to compare differences between participants with and without multimorbidity. Additionally, a heatmap was created to visually represent the co-occurrence patterns of various chronic conditions, with darker color intensities indicating the most prevalent combinations of chronic diseases.

Binary logistic regression was employed to examine the relationships between the dependent variable (multimorbidity) and the independent variables. Both unadjusted and adjusted logistic regression models were developed. The Akaike Information Criterion (AIC) was used to identify the best-fitting model; the model with the lowest AIC value indicated the best fit [40]. The initial model included all variables listed in Table 1, while the final model retained only religion, formal education, employment, depressive symptoms, life satisfaction, religious coping, social support, and resilience based on the lowest AIC value. Multicollinearity was assessed using the “proc reg” statement in SAS by evaluating the variance inflation factor (VIF). All variables had a VIF of less than 2.5, indicating no significant multicollinearity issues [41]. Advanced diagnostics, including dfbetas, influence, and leverage options in the “proc logistic” statement, were used to check for influential observations, and none were detected. Odds ratios and 95% confidence intervals were reported to quantify the odds of multimorbidity for specific levels of independent variables. Statistical significance was determined using a p-value threshold of less than 0.05.

Table 1.

Characteristics of the study participants by Multimorbidity status

Variables	Total (n = 276; 100%)	Multimorbidity		p-value
		No (n = 105; 38.0%)	Yes (n = 171; 62.0%)
	n (%)	n (%)	n (%)
Socio-demographics
Age in years				0.021
55–64	81 (29.4)	41 (39.0)	40 (23.4)
65–74	111 (40.2)	36 (34.3)	75 (43.9)
75+	84 (30.4)	28 (26.7)	56 (32.7)
Gender				0.032
Male	135 (48.9)	60 (57.1)	75 (43.9)
Female	141 (51.1)	45 (42.9)	96 (56.1)
Marital status				0.568
Married	205 (74.3)	80 (76.2)	125 (73.1)
Without partnera	71 (25.7)	25 (23.8)	46 (26.9)
Religion				0.001
Hindu	217 (78.6)	72 (68.6)	145 (84.8)
Other than Hindub	59 (21.4)	33 (31.4)	26 (15.2)
Received formal education				< 0.001
Yes	42 (15.2)	26 (24.8)	16 (9.4)
No	234 (84.8)	79 (75.2)	155 (90.6)
Able to read/write/speak in English (n = 273)				0.090
Yes	33 (12.0)	18 (17.1)	15 (8.8)
No	243 (88.0)	87 (82.9)	156 (91.2)
Current Employment				< 0.001 c
Yes	22 (8.0)	19 (18.1)	3 (1.8)
No	254 (92.0)	86 (81.9)	168 (98.2)
Years spent in a refugee camp (n = 267)				0.354
Less than 20	142 (53.2)	49 (49.5)	93 (55.4)
20 or more	125 (46.8)	50 (50.5)	75 (44.6)
Years in the USA (n = 266)				0.131
10 or less	109 (41.0)	46 (46.9)	63 (37.5)
More than 10	157 (59.0)	52 (53.1)	105 (62.5)
Health and healthcare utilization
Self-reported health				< 0.001
Very good/excellent	24 (8.7)	20 (19.0)	4 (2.3)
Good	84 (30.4)	47 (44.8)	37 (21.6)
Fair	102 (37.0)	28 (26.7)	74 (43.3)
Poor	66 (23.9)	10 (9.5)	56 (32.8)
Time since the last health facility visit				0.003 c
Within the last year	260 (94.2)	93 (88.6)	167 (97.7)
Never/more than a year	16 (5.8)	12 (11.4)	4 (2.3)
Has a regular doctor				0.029 c
Yes	267 (96.7)	98 (93.3)	169 (98.8)
No	9 (3.3)	7 (6.7)	2 (1.2)
Insurance type				< 0.001
Medicare	28 (10.1)	16 (15.2)	12 (7.0)
Medicaid	202 (73.2)	69 (65.7)	133 (77.8)
Dual (Medicare and Medicaid)	21 (7.6)	3 (2.9)	18 (10.5)
Employment-based	11 (4.0)	9 (8.6)	2 (1.2)
Unknown/others	11 (4.0)	5 (4.8)	6 (3.5)
No insurance	3 (1.1)	3 (2.9)	0 (0.0)
Health behaviors
Smoking status				0.845
Never	180 (65.2)	69 (65.7)	111 (64.9)
Former	80 (29.0)	29 (27.6)	51 (29.8)
Current	16 (5.8)	7 (6.7)	9 (5.3)
Tobacco consumption				0.330
Never	174 (63.0)	64 (61.0)	110 (64.3)
Former	70 (25.4)	25 (23.8)	45 (26.3)
Current	32 (11.6)	16 (15.2)	16 (9.4)
Alcohol consumption				0.006
Never	238 (86.2)	83 (79.1)	155 (90.6)
Former	31 (11.2)	16 (15.2)	15 (8.8)
Current	7 (2.6)	6 (5.7)	1 (0.6)
Self-reported physical activity level				< 0.001
High	37 (13.4)	24 (22.9)	13 (7.6)
Medium	127 (46.0)	50 (47.6)	77 (45.0)
Low	112 (40.6)	31 (29.5)	81 (47.4)
Depressive symptoms (n = 274)				< 0.001
Absent	187 (68.2)	88 (85.4)	99 (57.9)
Present	87 (31.8)	15 (14.6)	72 (42.1)
Life satisfaction (n = 269)				0.011 c
Satisfied	246 (90.1)	4 (3.9)	23 (13.5)
Dissatisfied	27 (9.9)	99 (96.1)	147 (86.5)
Religious coping (n = 273)				0.051
High	95 (34.8)	27 (26.2)	68 (40.0)
Moderate	83 (30.4)	33 (32.0)	50 (29.4)
Low	95 (34.8)	43 (41.8)	52 (30.6)
Social support (n = 273)				0.110
High	244 (89.4)	96 (93.2)	148 (87.1)
Moderate	29 (10.6)	7 (6.8)	22 (12.9)
Resilience (n = 273)				0.418
High	98 (35.9)	40 (38.8)	58 (34.1)
Moderate	85 (31.1)	34 (33.0)	51 (30.0)
Low	90 (33.0)	29 (28.2)	61 (35.9)

Note: p-values significant at < 0.05 are bolded

^aSeparate/divorced, widow/widower, and unmarried were combined into “without a partner”

^bBuddhist, Kirati, Christian, and others were combined into “other than Hindu”

^cp-value from Fisher’s exact test; all other p-values are from the Chi-square test

Results

Background characteristics

Table 1 provides an overview of the study participants’ characteristics, stratified by multimorbidity status. The study included a total of 276 resettled Bhutanese older adults in Ohio. Overall, the majority of participants were in the 65–74 age group (40.2%), with an almost equal distribution between males (48.9%) and females (51.1%). Most participants were married (74.3%) and identified as Hindu (78.6%). A significant proportion (84.8%) had not received formal education, and 88.0% were unable to read, write, or speak English. The vast majority (92.0%) were not employed at the time of the survey.

In terms of health status, only 8.7% reported very good or excellent health, while 37.0% reported fair and 23.9% reported poor health. Many of the participants (94.2%) had visited a health facility within the last year, and 96.7% had a regular doctor. Medicaid was the most common type of health insurance (73.2%). Regarding health behaviors, most participants reported never smoking (65.2%), never using tobacco (63.0%), and never drinking alcohol (86.2%). Additionally, 40.6% reported low physical activity. Depressive symptoms were reported in 31.8% of participants, while 90.1% reported being satisfied with life and had high social support (89.4%).

Prevalence of single conditions and multimorbidity

Table 1 also presents the prevalence of multimorbidity overall across several socio-demographic and health-related variables. The overall prevalence of multimorbidity was 62.0%, meaning that three out of five participants had two or more chronic conditions. The bivariate tests showed that several variables, such as age, gender, religion, formal education, employment status, self-reported health, health facility visits, doctor availability, health insurance, alcohol consumption, physical activity, depressive symptoms, and life satisfaction, were significantly associated with multimorbidity (see Table 1). Among those with multimorbidity, the highest prevalence was observed in older adults aged 65–74 (43.9%), women (56.1%), Hindus (84.8%), those with no formal education (90.6%), and those unemployed (98.2%).

Figure 1 presents a heatmap illustrating the combination of chronic diseases among the study participants. As shown in Fig. 1, hypertension (63.0%), high cholesterol (43.5%), diabetes (42.8%), and arthritis (42.0%) were the most common individual conditions, while cancer (n = 5, 1.8%) was the least prevalent. Regarding co-occurrence patterns, the most prevalent combination was hypertension and high cholesterol (n = 101, 36.6%), indicated by a darker color intensity at their intersection. This suggests that a significant number of participants suffer from both conditions simultaneously. Other notable combinations include hypertension with diabetes (n = 92, 33.3%), hypertension with arthritis (n = 87, 31.5%), and diabetes with high cholesterol (n = 81, 29.4%), as evidenced by the relatively darker shades in these areas of the heatmap.

Fig. 1.

Heatmap Illustrating Co-occurrence of Chronic Diseases among Study Participants

Factors associated with multimorbidity

Table 2 presents the results of the binary logistic regression analyses, highlighting the factors associated with multimorbidity among the study participants. The table includes both unadjusted and adjusted odds ratios; the adjusted odds ratios account for all variables listed in Table 2, selected based on the AIC criteria.

Table 2.

Unadjusted and adjusted odds ratio from the binary logistic regression for factors associated with Multimorbidity

	Unadjusted OR (95% CI)	Adjusted OR (95% CI)a
Religion
Other than Hindu	Ref.	Ref.
Hindu	2.61 (1.45–4.70)**	2.50 (1.26–4.96)**
Received formal education
Yes	Ref.	Ref.
No	3.08 (1.56–6.11)**	1.95 (0.87–4.35)
Current Employment
No	Ref.	Ref.
Yes	0.08 (0.02–0.28)***	0.15 (0.04–0.58)**
Depressive symptoms
Absent	Ref.	Ref.
Present	4.21 (2.25–7.87)***	3.13 (1.46–6.72)**
Life satisfaction
Dissatisfied	Ref.	Ref.
Satisfied	0.26 (0.09–0.77)*	0.53 (0.15–1.83)
Religious coping
High	Ref.	Ref.
Moderate	0.60 (0.32–1.13)	0.72 (0.36–1.43)
Low	0.48 (0.26–0.88)	0.66 (0.30–1.47)
Social support
High	Ref.	Ref.
Moderate	2.04 (0.84–4.96)	1.00 (0.34–2.96)
Resilience
High	Ref.	Ref.
Moderate	1.03 (0.57–1.87)	0.89 (0.45–1.77)
Low	1.45 (0.80–2.64)	0.78 (0.33–1.83)

Abbreviation: OR= Odds Ratio, CI= Confidence Interval

Significant odds ratios are bolded

^aThe model is adjusted for all variables in the table, which were selected based on the Akaike Information Criterion (AIC)

*p-value <0.05, **p-value <0.01, ***p-value <0.001

The results showed that after adjusting for all other variables in the model, religion, employment status, and depressive symptoms were significantly associated with multimorbidity (Table 2). Hindu participants had 2.50 times increased odds of multimorbidity compared to those from other religions (aOR: 2.50, 95% CI: 1.26–4.96, p < 0.01). Employed participants had 85% lower odds of having multimorbidity compared to those unemployed (aOR: 0.15, 95% CI: 0.04–0.58, p < 0.01). Participants with depressive symptoms had three-fold increased odds of multimorbidity compared to those without depressive symptoms (aOR: 3.13, 95% CI: 1.46–6.72, p < 0.01).

Discussion

This study assessed the prevalence of single and multiple chronic conditions and identified factors associated with multimorbidity among resettled Bhutanese older adults in Ohio. Hypertension emerged as the most prevalent condition and over half of the participants experienced multiple chronic conditions or multimorbidity. Common condition pairings included hypertension with high cholesterol, diabetes, and arthritis. Key factors linked to multimorbidity were religion, employment status, and depressive symptoms.

The study found a notable prevalence of multimorbidity among resettled Bhutanese older adults in Ohio, reflecting trends observed in other refugee populations globally. In particular, the high prevalence of hypertension aligns with global patterns, highlighting it as a significant health issue among adult populations [3] and is consistent with findings from previous studies on adult refugees [42, 43]. The combination of hypertension with other chronic conditions, such as high cholesterol and diabetes, is particularly concerning as it significantly increases the risk of severe health complications, including cardiovascular diseases, stroke, and kidney failure [44, 45]. The prevalence of a combination of hypertension and arthritis was also noted, aligning with nationally representative surveys in the USA that show an association between hypertension and arthritis, particularly rheumatoid arthritis, among individuals aged 60 and older [46]. Additionally, the combination of diabetes and high cholesterol was substantial in this study. Diabetes often reduces high-density lipoprotein or “good” cholesterol levels while increasing triglycerides and low-density lipoprotein or “bad” cholesterol levels, thereby raising the risk of heart disease and stroke—a condition known as diabetic dyslipidemia [47]. This high burden of disease underscores the heightened vulnerability of older refugees, who often face a convergence of factors such as socioeconomic hardship, cultural dislocation, limited access to healthcare, adverse environmental conditions, and psychological stress [48]. These factors, compounded over a lifetime and further exacerbated by the stresses of displacement and resettlement, contribute to the development of chronic conditions among these vulnerable populations.

Additionally, three in five participants (62.0%) had multimorbidity. To the best of our knowledge, this is the first study to examine multimorbidity among resettled Bhutanese older adults, so we lack comparative figures for this specific population. However, our findings reveal a considerably higher prevalence compared to the global prevalence of 51.0% among individuals over 60 years old [3]. This observed prevalence also exceeds that of the general adult population in the USA (58.4%) [49] and surpasses the prevalence reported among other older refugees (59.3%) [50]. Interestingly, Asian Americans are often perceived as a model minority with better health outcomes compared to other racial and ethnic groups in the USA. Consistent with this perception, a previous study found a lower prevalence of multimorbidity in Asian groups (41.3%) compared to other racial groups [49]. However, despite being part of this broadly heterogeneous Asian American category, Bhutanese individuals experience a disproportionately higher burden of chronic disease [51]. Therefore, the high prevalence of multimorbidity among study participants underscores the need for more detailed and segmented health studies within the diverse Asian population to better understand and address these health disparities.

In addition to the underlying biological mechanisms, several other factors contribute to the disproportionately high burden of multimorbidity observed among older Bhutanese. Socioeconomic disadvantages, along with the accumulated trauma and stressors associated with displacement, refuge, resettlement, and acculturation into the host country, play significant roles [52]. Resettled Bhutanese older adults have endured considerable trauma and stressors, including extended periods in refugee camps characterized by minimal resources, unsanitary conditions, and limited access to preventive care [53]. Such adverse conditions may have exacerbated their health risks and delayed necessary treatment for chronic conditions. Moreover, language and cultural barriers in the host country further complicate the process of navigating the healthcare system, making it even more challenging for this population to access appropriate care [52, 53]. According to life course theories, the cumulative socioeconomic disadvantages experienced over a lifetime can lead to poorer health outcomes in later years [7–10].

Bhutanese older adults who practice Hinduism were found to have higher odds of experiencing multimorbidity. Even within a displaced population, religion may act as a differentiating factor in health outcomes. Hindu Bhutanese refugees were historically more likely to experience religious persecution and prolonged stays in refugee camps [14, 52], which could contribute to cumulative trauma and stress-related health risks. In the resettlement context, Hindus may also face additional minority stress, cultural dissonance, and barriers to accessing culturally appropriate healthcare, which could exacerbate chronic disease burden [54]. Therefore, this finding highlights the importance of considering within-group heterogeneity in refugee health research.

Another potential explanation for this finding is that Hinduism is the predominant religion within this population [14, 16], with four-fifths of the study sample identifying as Hindus. Consequently, the higher prevalence of multimorbidity among Hindus might reflect the overall religious composition of the sample rather than indicating a direct association between Hinduism and multimorbidity. Previous studies have predominantly highlighted religion and spirituality as important resources for coping with chronic conditions, rather than establishing a direct link between religious affiliation and multimorbidity per se [55, 56]. This underscores the paucity of empirical evidence directly examining religion as a determinant of multimorbidity. Thus, a longitudinal study is needed to compare different religion and their practices to understand multimorbidity. In addition to that, exploring the relationship between religion and multimorbidity through the lens of translational epidemiology [57]—an emerging field that integrates epidemiological research with religious practices and beliefs to improve public health outcomes—can provide valuable insights. This approach suggests that religious characteristics can vary significantly across different sociodemographic groups, affecting their morbidity and mortality risks [57]. This perspective is especially pertinent for refugee populations, where religious and cultural beliefs play a crucial role. Displaced individuals, often with limited possessions and experiencing significant identity crises, frequently rely on their cultural and religious beliefs as essential coping mechanisms to navigate their new environments [58]. Understanding these dynamics can guide health programs for resettled Bhutanese older adults to address their unique needs, foster community support, enhance health education, and improve outcomes by leveraging the involvement of religious leaders and faith-based institutions in health promotion and prevention efforts.

While employment is not expected or necessary for all older adults, staying engaged in work-related activities in late life can significantly enhance both quality of life and health outcomes [59]. This study found that participants who were not engaged in paid employment had higher odds of experiencing multimorbidity compared to those who were employed, a finding consistent with a previous study conducted in Malaysia (aOR: 1.53; 95% CI:1.20 to 1.95; p = 0.001) [60]. Research on the relationship between employment and multimorbidity among older refugees is limited. However, the broader literature supports the various health benefits associated with remaining engaged in work-related activities. Continued employment, whether part-time or in a volunteer capacity, offers older adults more than just financial support—it provides them with a renewed sense of purpose and numerous opportunities for social interaction, which are essential for maintaining physical, mental, and emotional well-being [59]. Moreover, many workplaces offer health insurance and wellness programs that significantly improve access to preventive care, further supporting overall health [61]. Beyond these benefits, a supportive work environment fosters self-care and encourages positive health behaviors, which are vital for sustaining long-term well-being [59].

Participants with depressive symptoms were more likely to experience multimorbidity. Numerous studies support the connection between multimorbidity and depression, revealing that individuals with multiple chronic conditions are at an increased risk for depression [62–64]. This relationship is significant because depression can further complicate chronic diseases and exacerbate health complications, partly due to its influence on lifestyle behaviors such as poor diet, insufficient physical activity, and a sedentary lifestyle [65]. Research on the interplay between depression and multimorbidity in late life suggests a bidirectional relationship between these conditions [63, 65, 66]. Given these intertwined relationships, it is crucial to address mental health issues in older Bhutanese individuals who have chronic diseases. Ensuring that this population has access to integrated healthcare services that address both physical and mental health needs is essential for improving overall health outcomes and quality of life.

Strengths and limitations

Comprehensive data on the chronic disease profile of the Bhutanese population in the USA are limited and outdated [16]. This group is often included under broader Asian American categories, obscuring their unique health challenges [51]. Furthermore, research specifically focused on older Bhutanese adults is scarce. Thus, this study is crucial in addressing these significant gaps in literature. A major strength is that interviews were conducted in Nepali by trained researchers, facilitating effective communication between interviewers and participants. Additionally, validated Nepali-translated measurement scales were used. The study also achieved balanced gender representation, reducing potential gender biases in the analysis. Previous research underscores the importance of considering gender-related factors in multimorbidity, as gender is a significant determinant of chronic diseases [3, 67].

However, several limitations of this study should be noted. As a cross-sectional study, it can only provide a snapshot of data and is inherently limited in its ability to assess causal relationships between variables. Participant recruitment relied on snowball sampling with the help of local community leaders and Bhutanese organizations due to the lack of a comprehensive record of this population. This method may introduce selection bias, as the identified participants might not fully represent the broader population of older Bhutanese refugees. Despite this, snowball sampling is a common approach for reaching hard-to-access populations [23], and resettled Bhutanese in the USA are recognized as such a group [68]. The demographic variables were categorized with limited response options to reduce participant burden and enhance accessibility of the survey. However, this approach may have resulted in some information loss and potentially reduced statistical power. Data were collected through self-report, which introduces the possibility of measurement bias. This approach is vulnerable to several factors, such as memory recall issues, personal perceptions, and social desirability, all of which can impact the accuracy and reliability of the information provided.

Conclusions

The study highlights a significant prevalence of multimorbidity among resettled Bhutanese older adults in Ohio, emphasizing a critical public health issue for this vulnerable group. Hypertension emerged as the most common chronic condition, frequently occurring alongside high cholesterol, diabetes, and arthritis. Factors such as religion, employment status, and depressive symptoms play a significant role in determining the likelihood of multimorbidity.

Given the limited research on older Bhutanese populations in the US, this study offers essential baseline data for future research and program development at both local and state levels. The findings emphasize the need for policies that promote health equity and improve healthcare access for resettled refugees. Addressing the high rates of hypertension and multimorbidity requires targeted health research focused on chronic disease prevention and lifestyle management interventions aimed at preventing chronic conditions. The link between religion and multimorbidity highlights the importance of culturally sensitive healthcare practices and suggests involving faith-based institutions and leaders in health promotion and disease prevention efforts. Additionally, the influence of employment status and depressive symptoms on multimorbidity underscores the need to address social determinants of health. Enhancing community support, and social policies, and promoting late-life employment or volunteer opportunities can help alleviate the burden of multimorbidity. A comprehensive, holistic approach to healthcare—integrating physical and mental health through multidisciplinary teams and coordinated care—is essential for meeting the specific needs of resettled Bhutanese older adults.

Abbreviations

AIC

Akaike Information Criterion

aOR

Adjusted Odds Ratio

CDC

Centers for Disease Control and Prevention

CD-RISC

Connor-Davidson Resilience Scale

CI

Confidence Interval

COPD

Chronic obstructive pulmonary disease

GDS

Geriatric Depression Scale

MSPSS

Multidimensional Scale of Perceived Social Support

OR

Odds Ratio

SWLS

Satisfaction with Life Scale

UNHCR

United Nations High Commissioner for Refugees

VIF

Variance Inflation Factor

Authors’ contributions

IK and SG contributed to the conceptualization of this study. AS, IK, and SG collected the data. IK, SG, and MNH analyzed the data. IK wrote the original draft. SG, JS, AS, UNY, SKM, and MNH contributed to the initial review. IK and SG revised the manuscript. The final version of the manuscript was read, revised, finalized, and approved by all authors.

Funding

This work was supported by the Asian Resource Center for Minority Aging Research (RCMAR) at Rutgers University under Grant #AG0059304.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to privacy and confidentiality concerns. The data may contain sensitive information about individuals’ demographics, experiences, physical and mental health status, which must be protected to ensure the privacy and dignity of the participants. However, the de-identified datasets are available from the corresponding author on reasonable request with ethical approval.

Declarations

Ethics approval and consent to participate

The study was approved by the Institutional Review Board of Miami University (Protocol ID# 03942e). Verbal informed consent was obtained from all participants before their interviews. Participants were informed that their participation was voluntary, and confidentiality was strictly maintained throughout the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.