Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 28.
Published in final edited form as: J Racial Ethn Health Disparities. 2020 Jun 11;8(2):375–383. doi: 10.1007/s40615-020-00791-x

Inequities in Access to Medical Care Among Adults Diagnosed with Diabetes: Comparisons Between the US Population and a Sample of US-Residing Marshallese Islanders

Samuel D Towne 1,2,3,4,5, Karen H Kim Yeary 6, Marie-Rachelle Narcisse 7, Chris Long 7, Zoran Bursac 8, Rachel Totaram 1, Elisa M Rodriguez 6, Pearl McElfish 7
PMCID: PMC9972993  NIHMSID: NIHMS1876934  PMID: 32529423

Abstract

Objective

We examined barriers to accessing medical care for migrant US-residing Marshallese Islanders.

Methods

Cross-sectional analyses were conducted to identify potential inequities. Surveys from largely migrant diabetic Marshallese Islanders (n = 255) were compared with nationally representative data. Two major outcomes were assessed including 1—whether or not one reported having forgone medical care in the past year because of cost—and 2—whether or not one reported not having a usual source of care.

Results

Overall, 74% and 77% of Marshallese Islanders reported forgone care and no usual source of care, respectively, versus 15% and 7% of the US diabetic population. In multivariable analyses, being younger; uninsured; unemployed; male; of lower education; Native American or Hispanic (versus White); and residing in the South were associated with forgone care nationwide, whereas only lacking insurance was associated with forgone care among Marshallese Islanders. Nationwide being younger; uninsured; unmarried; female; of lower education; Native American or Hispanic (versus White); and residing in the South were associated with not having a usual source of care, whereas only being younger and uninsured were associated with not having a usual source of care among Marshallese Islanders.

Conclusion

The largest group of diabetic Marshallese Islanders in the continental US faces severe healthcare access inequities necessitating policies that increase access to health insurance options and associated resources.

Keywords: Health disparities, Social determinants of health, Health inequities, Immigrant health

Introduction

Barriers to accessing necessary health care can be burden-some, especially for individuals trying to manage complex chronic diseases, such as diabetes. The World Health Organization (WHO) estimated that 422 million adults had type-2 diabetes (T2D) worldwide as of 2014, a nearly fourfold increase over the estimated 108 million in 1980 [1], making this both a critical and rapidly intensifying issue in need of amelioration. The burden associated with diabetes is also severe. Diabetes is linked to mortality (1.5 million deaths in 2012) [1] and causes serious complications that can lead to premature death [1]. Health care access is a critical component for persons with diabetes to prevent complications and optimize health [2]. Factors that serve as barriers to accessing health care are particularly important to identify among at-risk populations with diabetes. Past evidence has reinforced this, where among those with diabetes, being uninsured and lacking a usual source of care have been shown to be linked to both a greater likelihood of delays in seeking urgent care and having unmet medical needs [3]. Identifying high-risk populations that can be targeted by tailored policy and community-based interventions complements the many critical components necessary to lower diabetes-related complications and diabetes mortality rates. High-risk groups can be classified in many ways including, but not limited to, education, race and ethnicity, and income [4]. The current study largely focusses on identifying potential barriers in accessing medical care with major comparisons across race and ethnicity highlighting a particularly high-risk and relatively understudied population. Marshallese Pacific Islanders with diabetes residing in the continental US have some of the highest rates of diabetes worldwide; however, there are gaps in what is known regarding access to care for this population.

The Republic of the Marshall Islands (RMI), a cluster of Pacific islands and atolls [5, 6], is located between Hawaii and Australia. Our focus on the largest population of mostly migrant Marshallese Islanders residing in the continental US is driven by their unique standing on the global stage in terms of T2D. Using data from the World Bank’s online data tool (population aged 20–79 years), we find that the rate of diabetes in the Marshall Islands, at 30.5%, was more than 3 times the worldwide rate (8.5%) [7], making it the highest of any nation highlighted in the data. Therefore, migrants from the Marshal Islands to the USA may be a particularly at-risk population. It is also relevant to note that the rates of diabetes in the USA, while lower than that of the Marshall Islands, were also higher than the worldwide rate reaching 10.8% as of 2017 [7]. Among Organisation for Economic Co-operation and Development (OECD) nations, the USA has among the highest rates of diabetes [8], thereby making study of diabetes among individuals of these two nations of particular interest. In addition, the USA and the Marshall Islands have a history that links these nations with a recent migration of Marshallese Islanders leaving the Marshall Islands to live and work in the USA [9].

In the 1940s and 1950s, the US Nuclear Testing Program tested a huge payload (many times that dropped in Hiroshima) in the RMI [10–15]. Inhabitants of the bombed islands and atolls were relocated; however, residents of nearby atolls were not, thereby experiencing nuclear fallout [16–25]. Natural food supplies on these atolls became and remain contaminated. Subsequently, the Marshallese diet shifted from one rich in native foods that facilitated healthy dietary choices to foods common in western cultures, those heavily processed and high in sugar and fat [13, 14]. Further, past studies have also reported low levels of physical activity among Marshallese Islanders, with 11.5% and 22.5% of respondents reporting engaging in moderate-intensity and high-intensity physical activity, respectively [26]. In addition, previous qualitative research, while limited by a small sample size and restricted to those already living in the USA, has demonstrated that some US-residing Marshallese individuals reported feeling that diabetes is inevitable, while also not aware of potential positive benefits of diabetes self-management if diagnosed [27]. This has led to an increase in diabetes rates and preventable conditions [28].

After World War II, the Marshall Islands was a Trust Territory of the USA, and in the 1980s, became a sovereign nation through a Compact of Free Association (COFA) with the USA. The COFA allows Marshallese to move to the USA (e.g., live, work, study) without requiring a visa or permanent resident card; [29] as a result, Marshallese Islanders are free to seek employment in the USA, which they have been doing at increasing rates since the late 1980s due to a lack of employment opportunity in the RMI [29]. According to data from the 2010 US Census measuring the total number of Native Hawaiian and Other Pacific Islander (NHPI), approximately 22,434 individuals reporting “detailed NHPI group alone or in any combination” reported Marshallese [30]. Over the course of the next few decades, it is expected that more Marshallese will live in the USA than in the Marshall Islands [31]. Many Marshallese have migrated to the northwest part of Arkansas. Arkansas now has the largest Marshallese population within the continental US with ~ 10,000–12,000 Marshallese [32].

Although COFA migrants have increased exponentially in the USA, COFA migrants are not eligible for Medicaid [9]; it is unknown the extent to which this considerable health care access barrier affects the health of COFA migrants, as many receive private insurance through employment. Within the USA, Marshallese Islanders are disproportionately affected by T2D which accounted for one third of Marshallese deaths from 1996 to 2000 [6]. While research is limited, pilot studies (N = 401) report that 38.4% of Arkansas-residing Marshallese have diabetes and 32.6% have prediabetes [33]. Ensuring adequate and equitable access to health care among this vulnerable population is critical for health equity, particularly given the unique health care access barriers Marshallese US migrants face. However, to the authors’ knowledge, healthcare access has not been thoroughly examined among US-residing Marshallese Islanders with T2D.

Objectives

Diabetes is clearly a critical issue facing both the Marshall Islands and the USA, and the recent migration of Marshallese Islanders to the USA provides a unique opportunity to assess factors associated with barriers to accessing medical care among this particularly vulnerable population. Therefore, we aimed to examine barriers to accessing medical care across two items: (1) forgone medical care due to cost and (2) not having a usual source of care. We present analyses from a sample of Marshallese Islanders with diabetes in Arkansas and the US population with diabetes.

Methods

Data

Two datasets were used in analyses for comparison between Marshallese Islanders with T2D in the USA and the general US population with diabetes, of which approximately 90–95% have T2D [34].

National Estimates

The Behavioral Risk Factor Surveillance System (BRFSS) database (2016) was used to serve as a comparison to data collected (using the same survey items as the BRFSS) among a sample of Marshallese Islanders residing in the continental US collected during a similar timeframe. As a nationally representative database of non-institutionalized adults, the BRFSS provides a national snapshot (cross-sectional design) of health and related outcomes. The unweighted sample size for the 2016 sample included 486,303 respondents. Only those diagnosed with diabetes (n = 66,053) were included in analyses.

Marshallese Islander Estimates

Data from Marshallese Islanders were collected in 2015–2017 from participants recruited through a community-based diabetes education program. Those with a diagnosis of T2D were included (n = 255). Based on comparisons with our sample and published data by the US Department of the Interior (e.g., indicating somewhat similar rates of educational attainment [35]), data published in reports based on the American Community Survey (e.g., somewhat similar rates for marital status such as the percent married [36]), and the experience of several members of the research team, who have been conducting research within this population for several years, it was estimated that this sample was likely representative of the larger population of Marshallese Islanders residing in Arkansas—which represents the largest population of Marshallese Islanders within the continental US [37].

Outcomes of Interest

Forgone Medical Care due to Cost in the Past 12 Months

The BRFSS survey item in both the general survey and the Marshallese survey included the following: Was there a time in the past 12 months when you needed to see a doctor but could not because of cost? [38], coded as yes or no.

Usual Source of Care

The BRFSS survey item in both the general survey and the Marshallese survey included the following: Do you have one person you think of as your personal doctor or health care provider? [38], coded as yes or no.

Variables of Interest

Age, a categorical variable in the BRFSS, was coded as 18–44 years, 45–54 years, 55–64 years, and 65 years and older. Insurance status was coded as insured or not insured. Marital status was coded as married or not married. Employment (employed for wages versus not employed) and education (less than high school graduate or equivalent versus higher) were included as a proxy for socioeconomic status. We also included race (non-Hispanic American Indian or Alaska Native, non-Hispanic Asian, non-Hispanic Black or African American, non-Hispanic White, other non-Hispanic), and ethnicity (Hispanic) in the national comparison. Additionally, given the regional location of the Marshallese Islanders sample, we included region in the national comparison.

Statistical Analyses

SAS 9.4 (Cary, NC) was used for analyses of the national BRFSS data, while Stata SE 15 was used to analyze the Marshallese Islander sample. SAS survey procedures were used to account for the complex sampling frame of the BRFSS. Given the dependent variables were binary, logistic regression was used to assess the likelihood of our outcomes with odds ratios and corresponding 95% confidence intervals (alpha = 0.05).

Results

Descriptive Statistics

Diabetes

Overall, the rate of T2D in the BRFSS sample at approximately 10.8% was roughly a third the rate at 30.5% [7] reported among Marshallese Island residents (national estimates from the Republic of the Marshall Islands) and 38.4% reported in Arkansas in previous research [33]. In terms of age, a similar proportion, roughly 1-in-4 were aged 55–64; however, whereas the largest segment of the US population with diabetes were aged 65 and older, among the sample of Marshallese Islanders residing in the USA with diabetes, the largest segments were those aged 18–44 years followed closely by those aged 45–54 years (see Table 1).

Table 1.

Distribution of the target populations by key characteristics

US noninstitutionalized population with diagnosed diabetes, 2016 US-residing Marshallese Islanders with diagnosed diabetes* US noninstitutionalized population with diagnosed diabetes and reporting forgone medical care, 2016 US-residing Marshallese Islanders with diagnosed diabetes and reporting forgone medical care US noninstitutionalized population with diagnosed diabetes and reporting no usual source of care, 2016 US-residing Marshallese Islanders with diagnosed diabetes and reporting no usual source of care
Percent Percent Percent Percent Percent Percent
Total 10.8% * 14.6% 73.5% 6.6% 76.8
Age 18–44 years 12.0 30.6 21.1 28.5 27.6 35.9
45–54 years 17.1 29.4 24.6 26.9 20.2 30.3
55–64 years 27.3 25.5 32.6 28.5 30.0 21.5
65 and older 43.6 14.5 21.7 16.1 22.1 12.3
Insurance Status Insured 92.7 66.7 74.5 23.4 65.2 24.3
Not insured 7.3 33.3 25.5 76.6 34.8 75.7
Marital status Married 53.2 67.8 48.2 68.8 43.6 70.8
Not married 46.8 32.2 51.8 31.2 56.4 29.2
Employment Employed for wages 32.6 36.5 35.1 31.2 41.9 37.4
Not employed 67.4 63.5 64.9 68.8 58.1 62.6
Sex Male 50.1 41.2 45.6 36.0 58.1 40.0
Female 49.9 58.8 54.4 64.0 41.9 60.0
Education Did not graduate High School 22.4 58.8 33.4 60.8 36.6 56.4
Graduated High School 77.6 41.2 66.6 39.2 63.4 43.6
Race/ethnicity Hispanic 17.0 N/A 28.3 N/A 34.7 N/A
Other 1.9 N/A 2.1 N/A 2.3 N/A
American Indian or Alaska Native 1.4 N/A 1.9 N/A 3.2 N/A
Asian 4.3 N/A 3.7 N/A 3.0 N/A
Black or African American 15.5 N/A 16.7 N/A 16.2 N/A
White 60.1 N/A 47.3 N/A 40.7 N/A
Census region South 41.6 N/A 51.1 N/A 48.8 N/A
North 16.8 N/A 13.3 N/A 10.4 N/A
Midwest 20.3 N/A 16.7 N/A 14.0 N/A
Western/Pacific 21.3 N/A 18.8 N/A 26.9 N/A
Open in a new tab
*

Sample of the largest population of Marshallese Islanders residing in the USA, not necessarily nationally representative

Forgone Medical Care due to costs

Nationwide, approximately 15% of those with diabetes reported forgone medical care due to cost (see Table 1). In contrast, approximately 74% of sampled Marshallese Islanders with diabetes residing in the USA reported forgone medical care due to costs.

Lacking a Usual Source of Care

Nationwide, approximately 7% of those with diabetes reported they did not have a usual source of care. In contrast, approximately 77% of sampled Marshallese Islanders with diabetes residing in the USA reported they did not have a usual source of care.

Unadjusted Analyses

Forgone Medical Care due to Costs

Nationwide, age, insurance status, marital status, employment, sex, education, race and ethnicity, and region were all significantly associated with forgone medical care due to costs (see Table 2). In particular, being in younger ages including ages 18–44 (versus older ages); not being insured (versus insured); not being married (versus being married); being employed (versus unemployed including retired); being female (versus male); having less than a high school education (versus high school or equivalent or higher); being American Indian or Alaska Native, Hispanic, Black or African American, or Other (versus being Non-Hispanic White); and residing in the South (versus all other regions) were associated with a higher likelihood of reporting forgone medical care due to cost among those with diabetes.

Table 2.

Unadjusted analysis for forgone care and forgone medication among individuals age 18 years of age and older with a diagnosis of diabetes

US noninstitutionalized population with diagnosed diabetes and reporting forgone medical care, 2016 US-residing Marshallese Islanders with diagnosed diabetes and reporting forgone medical care US noninstitutionalized population with diagnosed diabetes and reporting no usual source of care, 2016 US-residing Marshallese Islanders with diagnosed diabetes and reporting no usual source of care
OR 95% confidence intervals OR 95% confidence intervals OR 95% confidence intervals OR 95% confidence intervals
Age 45–54 years versus 18–44 years 0.76* 0.65 0.90 0.91 0.46 1.79 0.47* 0.38 0.58 0.42 0.17 1.05
55–64 years versus 18–44 years 0.61* 0.52 0.72 2.18 0.97 4.90 0.44* 0.35 0.55 0.22* 0.09 0.53
65 and older versus 18–44 years 0.23* 0.19 0.26 1.94 0.75 5.04 0.19* 0.16 0.23 0.21* 0.08 0.57
Insurance status Not insured versus insured 7.83* 6.62 9.27 5.17* 2.84 9.41 9.37* 7.72 11.39 5.47* 2.92 10.26
Marital status Not married versus married 1.27* 1.14 1.41 1.88 0.95 3.71 1.51* 1.29 1.77 1.46 0.73 2.91
Employment Not employed versus employed 0.88* 0.78 0.98 2.41* 1.36 4.27 0.65* 0.56 0.76 0.86 0.46 1.58
Sex Female versus male 1.24* 1.12 1.37 2.32* 1.32 4.11 0.71* 0.61 0.83 0.66 0.36 1.22
Education Less than high school versus high school or equivalent and above 1.95* 1.72 2.21 1.33 0.76 2.34 2.13* 1.79 2.54 1.63 0.91 2.92
Race/ethnicity Hispanic vs White 2.48* 2.14 2.88 N/A N/A N/A 3.35* 2.77 4.05 N/A N/A N/A
Other vs White 1.49* 1.10 2.02 N/A N/A N/A 1.88* 1.39 2.55 N/A N/A N/A
American Indian or Alaska Native vs White 1.99* 1.54 2.56 N/A N/A N/A 3.89* 2.86 5.28 N/A N/A N/A
Asian vs White 1.11 0.77 1.59 N/A N/A N/A 1.03 0.64 1.66 N/A N/A N/A
Black or African American vs White 1.45* 1.27 1.65 N/A N/A N/A 1.59* 1.26 1.99 N/A N/A N/A
Census region North vs South 0.60* 0.52 0.69 N/A N/A N/A 0.51* 0.41 0.63 N/A N/A N/A
Midwest vs South 0.62* 0.55 0.70 N/A N/A N/A 0.57* 0.47 0.69 N/A N/A N/A
Western/Pacific vs South 0.68* 0.58 0.78 N/A N/A N/A 1.08 0.90 1.30 N/A N/A N/A
Open in a new tab
*

Significantly different (alpha = 0.05)

Among US-residing Marshallese Islanders with diabetes, insurance status, employment, and sex were all associated with forgone medical care due to costs. In particular, not being insured (versus insured), not being employed (versus employed), and being female (versus male) were associated with a higher likelihood of reporting forgone medical care due to cost.

Lacking a Usual Source of Care

Nationwide, age, insurance status, marital status, employment, sex, education, race and ethnicity, and region were all significantly associated with reporting not having a usual source of care. In particular, being age 18–44 (versus older ages); not being insured (versus insured); not being married (versus being married); being employed (versus unemployed including retired); being male (versus female); having less than a high school education (versus high school or equivalent or higher); being American Indian or Alaska Native, Hispanic, Black or African American, or Other (versus being Non-Hispanic White); and residing in the South (versus the North and Midwest regions) were associated with a higher likelihood of reporting not having a usual source of care among those with diabetes.

Among US-residing Marshallese Islanders with diabetes, age and insurance status were associated with reporting not having a usual source of care. In particular, being age 18–44 (versus age 55–64 and 65 and older) and not being insured (versus insured) were associated with a higher likelihood of reporting not having a usual source of care.

Multivariable Analyses

Forgone Medical Care due to Costs

Nationwide, age, insurance status, employment, sex, education, race and ethnicity, and region were all significantly associated with forgone medical care due to costs among those with diabetes (see Table 3). In particular, being age 18–44 (versus those age 55–64 and 65 and older), not being insured (versus insured), not being employed (versus employed), being female (versus male), having less than a high school education (versus high school or equivalent or higher), being American Indian or Alaska Native or Hispanic (versus being Non-Hispanic White), and residing in the South (versus all other regions) were associated with a higher likelihood of reporting forgone medical care due to cost.

Table 3.

Multivariable analysis for forgone care and forgone medication among individuals age 18 years of age and older with a diagnosis of diabetes

US noninstitutionalized population with diagnosed diabetes and reporting forgone medical care, 2016 US-residing Marshallese Islanders with diagnosed diabetes and reporting forgone medical care US noninstitutionalized population with diagnosed diabetes and reporting no usual source of care, 2016 US-residing Marshallese Islanders with diagnosed diabetes and reporting no usual source of care
OR 95% confidence intervals OR 95% confidence intervals OR 95% confidence intervals OR 95% confidence intervals
Age 45–54 years versus 18–44 years 0.86 0.72 1.04 0.92 0.42 2.00 0.54* 0.42 0.69 0.48 0.18 1.29
55–64 years versus 18–44 years 0.67* 0.56 0.81 2.03 0.77 5.33 0.59* 0.46 0.77 0.20* 0.07 0.58
65 and older versus 18–44 years 0.28* 0.23 0.34 2.22 0.69 7.15 0.37* 0.28 0.48 0.18* 0.05 0.63
Insurance status Not insured versus insured 4.81* 4.02 5.76 5.00* 2.60 9.64 6.00* 4.75 7.58 6.05* 2.97 12.31
Marital status Not Married versus married 1.09 0.97 1.22 1.08 0.49 2.35 1.40* 1.17 1.67 1.63 0.72 3.69
Employment Not employed versus employed 1.28* 1.13 1.46 1.07 0.51 2.25 0.87 0.72 1.06 0.83 0.35 1.96
Sex Female versus male 1.21* 1.08 1.35 1.78 0.93 3.43 0.62* 0.52 0.74 1.12 0.54 2.32
Education Less than high school versus high school or equivalent and above 1.41* 1.22 1.63 1.27 0.67 2.40 1.36* 1.11 1.68 1.46 0.71 3.01
Race/ethnicity Hispanic vs White 1.52* 1.27 1.83 N/A N/A N/A 1.78* 1.41 2.24 N/A N/A N/A
Other vs White 1.12 0.77 1.63 N/A N/A N/A 1.47* 1.04 2.08 N/A N/A N/A
American Indian or Alaska Native vs White 1.63* 1.22 2.18 N/A N/A N/A 3.19* 2.31 4.40 N/A N/A N/A
Asian vs White 1.24 0.81 1.88 N/A N/A N/A 0.92 0.54 1.55 N/A N/A N/A
Black or African American vs White 1.05 0.90 1.21 N/A N/A N/A 1.23 0.95 1.59 N/A N/A N/A
Census region North vs South 0.68* 0.58 0.80 N/A N/A N/A 0.60* 0.47 0.75 N/A N/A N/A
Midwest vs South 0.73* 0.64 0.83 N/A N/A N/A 0.67* 0.55 0.83 N/A N/A N/A
Western/Pacific vs South 0.63* 0.53 0.76 N/A N/A N/A 1.09 0.87 1.38 N/A N/A N/A
Open in a new tab
*

Significantly different (alpha = 0.05)

Among US-residing Marshallese Islanders with diabetes, only insurance status was associated with forgone medical care due to costs. In particular, not being insured (versus insured) was associated with a statistically significant higher likelihood (OR = 5.00, 95% CI 2.60–9.64 for those not insured as compared with those who were insured) of reporting forgone medical care due to cost.

Lacking a Usual Source of Care

Nationwide, age, insurance status, marital status, sex, education, race and ethnicity, and region were all significantly associated with reporting not having a usual source of care among those with diabetes. In particular, being age 18–44 (versus older ages); not being insured (versus insured); not being married (versus being married); being male (versus female); having less than a high school education (versus high school/equivalent or higher); being American Indian or Alaska Native, Hispanic, or other (versus being Non-Hispanic White); and residing in the South (versus the North and Midwest regions) were associated with a higher likelihood of reporting not having a usual source of care.

Among US-residing Marshallese Islanders with diabetes, age and insurance status were associated with reporting not having a usual source of care. In particular, being age 18–44 (versus age 55–64 and 65 and older) was associated with a statistically significant higher likelihood (OR = 0.20, 95% CI 0.07–0.58 for those aged 55–64 as compared with those aged 18–44 years, and OR = 0.18, 95% CI 0.05–0.63 for those aged 65 and older as compared with those aged 18–44 years) of reporting not having a usual source of care. In addition, not being insured (versus insured) was associated with a statistically significant higher likelihood (OR = 6.05, 95% CI 2.97–12.31 for those not insured as compared with those who were insured) of reporting not having a usual source of care.

Discussion

We identified significant inequities facing migrants with T2D from the Marshall Islands residing in the continental US. Past studies using data from 2002 to 2005 have explored having a usual source of care among those with diabetes nationwide, concluding that lacking a usual source of care and being un-insured were associated with a lower odds of receiving health-related examinations (e.g., hemoglobin A1C tests, blood pressure [3]). Thus, the current study builds upon the strong connection between barriers in access to care and the likelihood of receiving potentially critical medical examinations that may lower the likelihood of diabetes-related complications.

Our findings indicated that approximately three-in-four Marshallese Islanders with diabetes reported forgone medical care. This is in stark contrast to the estimated 15% reported nationwide. Similarly, approximately three-in-four Marshallese Islanders with diabetes reported not having a usual source of care. Again, this is a major difference from that reported nationwide, where just 7% reported not having a usual source of care among those with diabetes as defined in the current analyses. Thus, Marshallese Islanders in the study show extreme health disparities that provide critical and timely information to key stakeholders. Therefore, identifying factors associated with both forgone medical care and not having a usual source of care can serve as targets for interventions (e.g., policy interventions, educational awareness). For example, the existing literature highlights diets characterized as heavily processed and high in sugar and fat [13, 14] in addition to low levels of physical activity [26] among Marshallese Islanders. Interventions targeting healthful diets and adequate physical activity can serve as both preventative measures (e.g., preventing those with pre-diabetes from developing diabetes) and tertiary measures (e.g., chronic disease self-management among those already diagnosed with diabetes). In addition, previous work highlighting limited awareness of potential positive benefits of diabetes self-management if diagnosed [27] is a potential target for clinical interventions including physician referral to community- and evidence-based programs focused on diabetes [39] or chronic disease self-management in general [40]. These behavioral interventions may hold promise for ameliorating the burden of diabetes among at-risk groups, such as Marshallese Islanders.

Among the factors associated with forgone medical care due to cost, lacking health insurance, not being employed, and being female all served as risk factors among Marshallese Islanders. When considering several factors simultaneously, only insurance status (lacking insurance) remained a significant factor associated with forgone medical care due to cost. This highlights the critical need for relevant public policies that are aimed at bridging significant gaps in access to health insurance, especially for those in occupations not provided health insurance through their employer. For example, while Marshallese Islanders are afforded the right to be in the USA, there appears to be a lack of targeted policies and information on how to access health insurance, potentially serving as barriers in accessing affordable health care. Policies limiting COFA migrants from being eligible for some federal benefits [41] namely Marshallese Islanders eligibility for Medicaid [9] may contribute to barriers in access to health care that may continue to contribute to health inequities [5]. Concerns of the potential for limited access health care expressed in past work [9, 41], are consistent with our study findings identifying existing gaps in multiple measures of access to care. For this reason, policies seeking to bridge the gap in access to care, namely policies seeking to equip Marshallese Islanders with health insurance, are likely necessary.

Among the factors associated with not having a usual source of care, lower age (being age 18–44) and lacking health insurance served as risk factors among Marshallese Islanders. When considering several factors simultaneously, each of these factors remained as significant factors associated with not having a usual source of care. Thus, Marshallese Islanders age 18–44 and those without health insurance are particularly vulnerable. Key stakeholders can use this information to target those most at risk of facing barriers to critically important medical care which may serve to reduce potentially preventable complications.

Limitations

Several limitations warrant mentioning. The self-reported nature of diabetes prevents a full characterization of all with diabetes, where those with undiagnosed diabetes may be misclassified. Further, the cross-sectional design does not allow for causal inference. Additionally, the relatively small sample of Marshallese Islanders may not be representative of all Marshallese Islanders nationwide, globally, or those without diabetes. Further, we were restricted to the definition of forgone medical care due to cost using the single survey item that does not allow for a measure of the intensity of forgone medical care (e.g., number of times/year). Past studies have evaluated this variable in terms of cost-related barriers in access to care including comparisons across race and ethnicity nationwide [42, 43], and among those with diabetes [44] using older data. Therefore, it was appropriate for comparative purposes among Marshallese Islanders residing in the USA. Further, having a usual source of care was also operationalized using a single survey item and as such may not be comparable to studies using other measures of having a usual source of care. These and other limitations should be considered in light of the implications in the current study. Future research should aim to recruit a larger sample of Marshallese Islanders (with/without diabetes) for more rigorous analyses (e.g., longitudinal studies) of potential health inequities facing this population. However, this was a sample drawn from the largest population of Marshallese Islanders residing in the continental US adding considerable strength to our approach.

Conclusions

Identifying inequities in accessing health care among this relatively understudied population provides actionable information that can be used, by policy makers and other key stake-holders, to implement public policy and community-based interventions aimed at ameliorating inequities in accessing necessary medical care.

Funding Information

A portion of the data used in the current study were collected using a previously funded project. That research is made possible because of our CBPR partnership with the Marshallese Consulate General in Springdale, Arkansas, the Arkansas Coalition of Marshallese, and the Gaps in Services to the Marshallese Task Force. The CBPR partnership support is provided from the University of Arkansas for Medical Sciences Translational Research Institute (TRI) grant UL1TR000039, which is funded through the NIH National Center for Research Resources and National Center for Advancing Translational Sciences. The current research that this article describes was supported through a Patient-Centered Outcomes Research Institute (PCORI) Award. All statements in this report, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the NIH nor of the PCORI.

Footnotes

Conflict of Interest The authors declare that they have no conflict of interest.

Ethical Approval Ethical approval provided from the University of Arkansas for Medical Sciences Northwest (UAMS) Institutional Review Board (IRB) UAMS-IRB-Approval-203282.

References

  • 1.Organization WH: Global report on diabetes: World Health Organization; 2016. [Google Scholar]
  • 2.Association AD. Standards of medical care for patients with diabetes mellitus. Diabetes Care. 2002;25(suppl 1):s33–49. [DOI] [PubMed] [Google Scholar]
  • 3.DeVoe JE, Tillotson CJ, Wallace LSJDC. Usual source of care as a health insurance substitute for US adults with diabetes? 2009;32(6): 983–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Solar O, Irwin A: A conceptual framework for action on the social determinants of health. 2007.
  • 5.McElfish PA, Hallgren E, Yamada S. Effect of US health policies on health care access for Marshallese migrants. Am J Public Health. 2015;105(4):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Reddy R, Shehata C, Smith G, Maskarinec GG. Characteristics of Marshallese with type 2 diabetes on Oahu: a pilot study to implement a community-based diabetic health improvement project. Calif J Health Promot. 2005;3(4):36–47. 10.1901/jaba.2005.3-36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.World Bank. International Diabetes Federation, Diabetes atlas: diabetes prevalence (% of population ages 20 to 79). Available at: https://data.worldbank.org/indicator/SH.STA.DIAB.ZS?year_high_desc=false. Accessed 9 Jun 2020.
  • 8.OECD. Diabetes prevalence, in Health at a glance 2017. Paris: OECD indicators, OECD Publishing; 2017. 10.1787/health_glance-2017-15-en. [DOI] [Google Scholar]
  • 9.McElfish PA, Purvis RS, Maskarinec GG, Bing WI, Jacob CJ, Ritok-Lakien M, et al. Interpretive policy analysis: Marshallese COFA migrants and the Affordable Care Act. Int J Equity Health. 2016;15(1):91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Cooke S In mortal hands: a cautionary history of the nuclear age. Bloomsbury: USA; 2010. [Google Scholar]
  • 11.Barker H Bravo for the Marshallese: regaining control in a post-nuclear, the case of Marshallese post-colonial world. Independence: Cengage Learning; 2012. [Google Scholar]
  • 12.Pollock NJ. Health transitions, fast and nasty: the case of Marshallese exposure to nuclear radiation. Pac Health Dialog. 2002;9(2):275–82. [PubMed] [Google Scholar]
  • 13.Guyer RL. Radioactivity and rights: clashes at Bikini Atoll. Am J Public Health. 2001;91(9):1371–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Gittelsohn J, Haberle H, Vastine AE, Dyckman W, Palafox NA. Macro- and microlevel processes affect food choice and nutritional status in the Republic of the Marshall Islands. J Nutr. 2003;133(1): 310S–3S. [DOI] [PubMed] [Google Scholar]
  • 15.Yamada S, Palafox N. On the biopsychosocial model: the example of political economic causes of diabetes in the Marshall Islands. Fam Med. 2001;33(9):702–4. [PubMed] [Google Scholar]
  • 16.Bogen KT, Conrado CL, Robison WL. Uncertainty and variability in updated estimates of potential dose and risk at a U.S. nuclear test site–Bikini Atoll. Health Phys. 1997;73(1):115–26. [DOI] [PubMed] [Google Scholar]
  • 17.Gilbert ES, Land CE, Simon SL. Health effects from fallout. Health Phys. 2002;82(5):726–35. [DOI] [PubMed] [Google Scholar]
  • 18.Lessard ET, Miltenberger RP, Cohn SH, Musolino SV, Conard RA. Protracted exposure to fallout: the Rongelap and Utirik experience. Health Phys. 1984;46(3):511–27. [DOI] [PubMed] [Google Scholar]
  • 19.Niedenthal J A history of the people of Bikini following nuclear weapons testing in the Marshall Islands: with recollections and views of elders of Bikini Atoll. Health Phys. 1997;73(1):28–36. [DOI] [PubMed] [Google Scholar]
  • 20.Noshkin VE, Robison WL, Wong KM, Brunk JL, Eagle RJ, Jones HE. Past and present levels of some radionuclides in fish from Bikini and Enewetak Atolls. Health Phys. 1997;73(1):49–65. [DOI] [PubMed] [Google Scholar]
  • 21.Robison WL, Bogen KT, Conrado CL. An updated dose assessment for resettlement options at Bikini Atoll–a U.S. nuclear test site. Health Phys. 1997;73(1):100–14. [DOI] [PubMed] [Google Scholar]
  • 22.Simon SL. A brief history of people and events related to atomic weapons testing in the Marshall Islands. Health Phys. 1997;73(1): 5–20. [DOI] [PubMed] [Google Scholar]
  • 23.Simon SL, Robison WL, Thorne MC, Toburen LH, Franke B, Baverstock KF, et al. A comparison of independently conducted dose assessments to determine compliance and resettlement options for the people of Rongelap Atoll. Health Phys. 1997;73(1):133–51. [DOI] [PubMed] [Google Scholar]
  • 24.Walker RB, Gessel SP, Held EE. The ecosystem study on Rongelap Atoll. Health Phys. 1997;73(1):223–33. [DOI] [PubMed] [Google Scholar]
  • 25.Robison WL, Noshkin VE, Conrado CL, Eagle RJ, Brunk JL, Jokela TA, et al. The northern Marshall Islands radiological survey: data and dose assessments. Health Phys. 1997;73(1):37–48. [DOI] [PubMed] [Google Scholar]
  • 26.Ichiho HM, deBrum I, Kedi S, Langidrik J, Aitaoto N. An assessment of non-communicable diseases, diabetes, and related risk factors in the Republic of the Marshall Islands, Majuro Atoll: a systems perspective. Hawaii J Med Public Health. 2013;72(5 Suppl 1):87–97. [PMC free article] [PubMed] [Google Scholar]
  • 27.McElfish PA, Hallgren E, Henry LJ, Ritok M, Rubon-Chutaro J, Kohler P. Health beliefs of Marshallese regarding type 2 diabetes. Am J Health Behav. 2016;40(2):248–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Ichiho HM, Seremai J, Trinidad R, Paul I, Langidrik J, Aitaoto N. An assessment of non-communicable diseases, diabetes, and related risk factors in the Republic of the Marshall Islands, Kwajelein Atoll, Ebeye Island: a systems perspective. Hawaii J Med Public Health. 2013;72(5 Suppl 1):77–86. [PMC free article] [PubMed] [Google Scholar]
  • 29.McElfish PA. Marshallese COFA migrants in Arkansas. The Journal of the Arkansas Medical Society. 2016;112(13):3. [PMC free article] [PubMed] [Google Scholar]
  • 30.Hixson L, Hepler B, Kim M: The Native Hawaiian and other Pacific Islander population: 2010 (2010 Census Briefs, C2010BR-12). 2012. Retrieved from US Census Bureau website: http://www.census.gov/prod/cen2010/briefs/c2010br-12.pdf. Accessed 9 Jun 2020. [Google Scholar]
  • 31.Marshall Islands Journal: Another donor country in the U.S.?. In: Marshall Islands Journal. 2013. [Google Scholar]
  • 32.McElfish PA. Marshallese COFA migrants in Arkansas. J Ark Med Soc. 2016;112(13):259–62. [PMC free article] [PubMed] [Google Scholar]
  • 33.McElfish PA, Rowland B, Long CR, Hudson J, Piel M, Buron B, et al. Diabetes and hypertension in Marshallese adults: results from faith-based health screenings. J Racial Ethn Health Disparities. 2017;4(6):1042–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Control CD. Prevention: national diabetes statistics report, 2017. Atlanta: Centers for Disease Control and Prevention, US Department of Health and Human Services; 2017. [Google Scholar]
  • 35.Joshua TH: Interior supports Marshallese community in Arkansas with grant to Marshallese resource and educational center. 2019. Available at: https://www.doi.gov/oia/interior-supports-marshallese-community-arkansas-grant-marshallese-resource-and-educational. Accessed 9 Jun 2020.
  • 36.Levin MJ: Marshallese migrants in the United States in 2015: a statistical profile based on the American Community Survey. Available at: http://www.rmiembassyus.org/images/pdf/Marshallese-Migrants-in-the-United-States-in-201570.pdf. Accessed 28 Jun 2019.
  • 37.Donoho G, McElfish P, Avants R, Hallgren E. A novel recruiting and surveying method: participatory research during a Pacific Islander community’s traditional cultural event. Gateways: International Journal of Community Research and Engagement. 2015;8(1):150–9. 10.5130/ijcre.v8i1.4227. [DOI] [Google Scholar]
  • 38.CDC. (2016) Behavioral Risk Factor Surveillance System (BRFSS). 2016 Codebook. Available at: https://www.cdc.gov/brfss/annual_data/2016/pdf/codebook16_llcp.pdf. [Google Scholar]
  • 39.Lorig K, Ritter PL, Villa FJ, Armas J. Community-based peer-led diabetes self-management: a randomized trial. Diabetes Educ. 2009;35(4):641–51. 10.1177/0145721709335006. [DOI] [PubMed] [Google Scholar]
  • 40.Lorig KR, Ritter P, Stewart AL, Sobel DS, Brown BW Jr, Bandura A, et al. Chronic disease self-management program: 2-year health status and health care utilization outcomes. Med Care. 2001;39(11): 1217–23. [DOI] [PubMed] [Google Scholar]
  • 41.Shek D, Yamada S. Health care for Micronesians and constitutional rights. Hawaii Med J. 2011;70(11 Suppl 2):4–8. [PMC free article] [PubMed] [Google Scholar]
  • 42.Towne SD Jr. Socioeconomic, geospatial, and geopolitical disparities in access to health care in the US 2011–2015. Int J Environ Res Public Health. 2017;14(6):573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Towne Jr. S: Differential effects of the great recession on minority populations. Department of Health Services, Policy, and Management, Arnold School of Public Health, University of South Carolina; 2013. [Google Scholar]
  • 44.Towne SD, Bolin J, Ferdinand A, Nicklett EJ, Smith ML, Ory MG. Assessing diabetes and factors associated with foregoing medical care among persons with diabetes: disparities facing American Indian/Alaska Native, Black, Hispanic, low income, and southern adults in the US (2011–2015). Int J Environ Res Public Health. 2017;14(5):464. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES