Assessment of Diabetes Self-Care Behaviors and Knowledge among Marshallese Adults with Type 2 Diabetes in the Republic of the Marshall Islands

Abstract

Aims:

The aim of this study is to assess and document engagement in type 2 diabetes mellitus (T2DM) self-care behaviors and self-reported diabetes knowledge among Marshallese adults living in the Republic of the Marshall Islands (RMI).

Methods:

The study uses data from a T2DM health screening study completed in the RMI; survey and biometric data were captured as part of the health screenings. Study objectives were examined using descriptive statistics to describe the characteristics of the participants, their diabetes self-care behaviors, and their levels of self-reported diabetes knowledge.

Results:

Results indicate many Marshallese diagnosed with T2DM did not engage in adequate self-care behaviors, including blood sugar checks and foot examinations. Participants reported having forgone needed medical care and medication due to issues with cost and/or access, and participants reported low levels of diabetes knowledge.

Conclusions:

The results demonstrate the need for further work in improving engagement in diabetes self-care by Marshallese living in the RMI. Increased engagement in self-care and diabetes education programs may help Marshallese with T2DM to improve control of their glucose and avoid long-term health complications, as well as reduce costs to the healthcare system.

Introduction

The Republic of the Marshall Islands (RMI) is a small grouping of atolls located between Hawai’i and New Zealand with a population of approximately 58,000 people. The Marshallese population experiences extremely high prevalence of overweight and obesity-related illness, particularly type 2 diabetes mellitus (T2DM); the RMI has the highest age-adjusted T2DM prevalence in the world (30.5%), compared to the United States (US) (13.3%) and globally (9.3%) [1–5].

The health disparities experienced by Marshallese are related to historical trauma resulting from extensive nuclear weapons testing conducted by the US government. From 1946 to 1958, the US military conducted more than 60 nuclear tests in the RMI [6]. The resulting nuclear fallout dramatically altered the Marshallese lifestyle [7, 8]. Before the nuclear testing, inhabitants of the RMI relied on a traditional diet of fresh fruits, vegetables, and fish sourced from the local agricultural systems, which required high levels of physical activity to acquire. Nuclear radiation from the explosions contaminated the local food ecosystems that were central to the sustenance farming lifestyle of the Marshallese people. As a result, inhabitants of the RMI involuntarily shifted their diet to one reliant on high-carbohydrate commodity foods imported by the US, and physical activity related to food acquisition was greatly reduced [6, 8–11]. The high prevalence of T2DM in the Marshallese population is likely attributable to this sudden shift in nutrition and physical activity.

In addition to healthy diet and regular exercise, proper management of T2DM requires numerous self-care behaviors that should be integrated into an individual’s daily routine, including monitoring glucose levels, managing stress, regularly taking medications, and routinely seeing a doctor [12, 13]. If unmanaged, T2DM can lead to life-threatening irreversible health complications like cardiovascular diseases, stroke, nephropathy, neuropathy, retinopathy, and amputation [14, 15]. Studies have found that Marshallese individuals experience socioeconomic barriers to engagement in self-care behaviors, particularly cost-related medication non-adherence and forgoing medical care due to cost [13, 16]. The minimum wage in the RMI is extremely low and is currently set at $3 per hour [17]. Due to the low minimum wage, healthcare is not always affordable. All Marshallese citizens carry basic insurance under the national healthcare plan; however, under this plan, they pay $5 for a full outpatient visit including laboratory, diagnostics, and pharmacy, $17 for an Emergency Room visit, and $10 per hospital admission [18]. Public transportation costs range between $1 and $5 each way [18].

Engagement with diabetes self-care behaviors has been documented among Marshallese populations living in the US and among other racial and ethnic minorities [13, 19]. However, few studies have examined engagement in self-care behaviors among Marshallese populations living in the RMI. The current study seeks to fill this gap in the literature by examining self-reported diabetes self-care behaviors and diabetes self-management knowledge among Marshallese adults living in the RMI. Understanding diabetes self-care behaviors and diabetes self-management knowledge may help to inform future interventions and programs to overcome barriers to appropriate T2DM management in the RMI.

Materials and Methods

Community-Based Participatory Research

This study utilized a community-based participatory research (CBPR)^20-23 approach, the use of which is important given the historical trauma experienced by the Marshallese people [6]. CBPR engages community partners, honors their unique contributions at all stages of research, and ensures Marshallese cultural knowledge informs the process [20–22]. The study team worked closely with the RMI Ministry of Health and Human Services (MOHHS), the Marshallese Educational Initiative, Kora In Jiban Lolorjake Ejmour, and a community advisory board (CAB) to develop the study protocol and activities. A more detailed description of the study development and protocol can be found elsewhere [23].

Study Population

The study uses primary data from a health screening study completed in the RMI. Individuals who met the inclusion criteria and consented to participate in the study provided survey and biometric data. Participant inclusion criteria included: (1) self-identified as Marshallese and (2) over 18 years of age. The CBPR partners recommended verbal consent with a written document outlining the study information. Previous work with the Marshallese community has highlighted participants’ desire for a verbal discussion of study requirements and a simplified consent process [24, 25]. Therefore, a waiver of consent documentation was granted by the Institutional Review Board (IRB #262557) and approved by the RMI MOHHS, and verbal consent was utilized.

Community health workers and CAB members assisted with recruiting churches to participate in the screenings. Recruitment efforts also included announcements on the local radio station, through text messaging, and through invitations extended to non-participating churches who inquired about participation after hearing of the health screenings taking place elsewhere. Once a church agreed to participate, town hall-style question and answer sessions were conducted to inform and recruit individual participants. Additionally, the health screenings were advertised via postings on local websites, social media, and study flyers.

At the health screening events, participants were provided a study information sheet in Marshallese and were allowed time to read and review the information sheet with a community health worker. Participants were encouraged to ask any questions they may have had before verbally acknowledging consent to participate in the study. Participants received a copy of the study information sheet in Marshallese for their records, and those who consented to the study were given $10 as compensation for their participation. All participants received counseling regarding their results at the health screenings.

Measures

Survey and biometric data were captured as part of the health screenings. All staff had previous experience with the collection of biometric and survey data in the RMI and had completed human subjects, ethics, research privacy and confidentiality, blood-borne pathogen, biometric data collection, and study-specific trainings. Study instruments were chosen collaboratively with Marshallese stakeholders, translated into Marshallese, and validated with Marshallese participants. The data collection methods and survey items have been utilized in prior studies with Marshallese participants [26–31].

Sex, educational attainment, and age of the participant were collected via self-report. Biometric measures collected included weight and height (to calculate body mass index [BMI]) and glycated hemoglobin (HbA1c). Participants’ weight (without shoes) was measured to the nearest 0.5 pound using a calibrated digital scale. Height (without shoes) was measured to the nearest 0.25 inch using a stadiometer. BMI was calculated using collected height and weight [(weight in pounds/height in inches²) X 703]. HbA1c was collected via finger prick using aseptic techniques and analyzed using PTS Diagnostic’s A1CNOW+ point-of-care tests. HbA1c was recorded as percent of glycated hemoglobin (NGSP %). HbA1c was categorized as controlled if the participant’s HbA1c was less than 9.0% and uncontrolled if their HbA1c was 9.0% or greater.

To determine the extent of the participants’ diabetes knowledge and to document participants’ diabetes self-care behaviors and ability to access healthcare services, participants completed a survey instrument adapted from the Behavioral Risk Factor Surveillance System (BRFSS) survey’s Diabetes and Healthcare Access Modules [32] and the validated Michigan Diabetes Care Profile [33]. All responses were self-reported. The Diabetes and Healthcare Access Modules of the BRFSS include standardized questions for sociodemographic information, when the participant last saw a healthcare provider, forgoing care due to cost, the frequencies of blood sugar monitoring, and the frequencies of feet checks. The Diabetes Care Profile [33] module called “Understanding” was used, which asks participants about their knowledge/understanding of diabetes and diabetes care. Questions were translated into Marshallese and edited as needed to fit within the context of the RMI.

Analysis

Study objectives were examined using descriptive statistics to describe the participants, their self-reported diabetes self-care behaviors, and their self-reported level of diabetes knowledge. Analysis was performed using STATA SE v17 (Statacorp, LLC., College Station, TX) [34].

Results

Demographics

Five hundred and twenty eight (528) Marshallese adults from 20 churches located on Majuro Atoll participated in the health screenings. Of the 528 participants who participated in the health screenings, 152 indicated they had a previous diagnosis of T2DM by responding yes to the question, “Has a doctor, nurse, or other health professional EVER told you that you had diabetes?” Individuals with a previous diagnosis of T2DM were administered an additional survey which asked participants about their knowledge of diabetes and their diabetes-related self-care behaviors. The 152 participants who completed this survey were included in the analysis.

Table 1 provides the descriptive statistics of the participants’ demographic characteristics and self-care behaviors. Participants were predominately female (67.8%) and between the ages of 45 and 64 (59.2%), with a mean age of 52.8 years (±11.7 years). Two-thirds of the participants had completed some high school (38.2%) or an elementary-level (22.4%) education. The majority of the participants were overweight (35.3%) or obese (51.3%) and had an HbA1c indicative of uncontrolled T2DM (60.7%).

Table 1.

Participant Demographics, Biometrics, and Engagement in Self-Care Behaviors

	Obs	N (Mean ± SD)	% of Sample	Range
Sex	152
Female		103	67.8
Male		49	32.2
Education	152
Never attended or elementary education		34	22.4
Some high school		58	38.2
High school graduate or GED		31	20.4
Some college or technical school		23	15.1
College graduate		6	4.0
Age (continuous)	152	52.8 ± 11.7	--	23 - 77
Age (categorical)	152
18-24		2	1.3
25-44		35	23.0
45-64		90	59.2
>65		25	16.5
BMI 1	150
Normal weight		20	13.3
Overweight		53	35.3
Obese		77	51.3
Blood sugar control 2	140
Controlled (HbA1c < 9.0%)		55	39.3
Uncontrolled (HbA1c > 9.0%)		85	60.7
Number of times per day checked glucose	133
Zero		28	21.1
Once		76	57.1
Twice		20	15.0
Three or more times		9	6.8
Number of days checked feet	151
Less than 7 days per week		102	67.6
Seven days per week		49	32.5
Number of days checked shoes
Less than 7 days per week		83	55.0
Seven days per week		68	45.0
Last visit with a doctor, nurse, or health professional	152
Within the past 6 months		114	75.0
More than a year ago		22	14.5
More than 2 years ago		7	4.6
None		4	2.6
Don’t know/not sure		5	3.3
Forgone medical care	152
No		42	27.6
Yes		110	72.4
Forgone medication	152
No		87	57.2
Yes		65	42.8

Notes: Only valid percentages shown. Percentages may not total 100 due to rounding.

SD, standard deviation; GED, graduate equivalency degree; BMI, body mass index; HbA1c, hemoglobin A1c

¹Missing data due to participant limitations

²Missing data due to supply chain issues obtaining HbA1c test kits

Self-Care Behaviors

Slightly more than half of participants reported checking their glucose levels once per day (57.1%), followed by zero times per day (21.1%) and twice per day (15.0%). The majority of participants did not check their feet (67.6%) or their shoes (55.0%) every day as recommended.

Healthcare Access

Three-fourths of participants (75.0%) stated they had seen a physician or other healthcare provider within the past six months, 14.5% stated it had been more than a year, and 4.6% reported seeing a physician or other healthcare provider more than two years ago. A small number of participants stated they had never seen a physician or other healthcare provider (2.6%) or they did not know the last time they had seen a provider (3.3%). Nearly three-fourths (72.4%) of participants reported having forgone medical care, and 42.8% had forgone needed medications due to cost or lack of access (e.g., no transportation, out of stock, closed/no staff available).

Diabetes Knowledge

Table 2 reports the descriptive results of participants’ diabetes knowledge. Overall, the mean score for diabetes knowledge was 17.3 (±7.3) with a range of 0-26. More than half of participants stated they knew ‘a lot’ about taking medications correctly (60.5%), taking care of their feet (55.9%), the benefits of improving blood sugar control (55.9%), the role of exercise in diabetes (51.3%), and how to prevent and treat low blood sugar (51.3%). Areas where less than half of participants reported knowing ‘a lot’ were regarding how to manage pregnancy and diabetes (36.2%), how to use the results of blood sugar monitoring (45.4%), and how to eat for blood sugar control (46.7%).

Table 2.

Participant Perceived Diabetes Knowledge by Category (n = 152)

How well do you understand…	N (Mean ± SD)	%	Range
Diabetes knowledge scale score	17.3 ± 7.3		0 - 26
How to manage your diabetes?
Not at all	24	15.8
A little	52	34.2
A lot	76	50.0
How to manage your stress?
Not at all	26	17.1
A little	50	32.9
A lot	76	50.0
How to eat for blood sugar control?
Not at all	23	15.1
A little	58	38.2
A lot	71	46.7
The role of exercise in diabetes?
Not at all	21	13.8
A little	53	34.9
A lot	78	51.3
How to take your medications correctly?
Not at all	20	13.2
A little	40	26.3
A lot	92	60.5
How to use the results of blood sugar monitoring?
Not at all	39	25.7
A little	44	29.0
A lot	69	45.4
How diet, exercise, and medicines affect blood sugar levels?
Not at all	22	14.5
A little	54	35.5
A lot	76	50.0
How to prevent and treat high blood sugar?
Not at all	27	17.8
A little	55	36.2
A lot	70	46.1
How to prevent and treat low blood sugar?
Not at all	21	13.8
A little	53	34.9
A lot	78	51.3
How to prevent long-term complications of diabetes?
Not at all	24	15.8
A little	53	34.9
A lot	75	49.3
How to take care of your feet?
Not at all	18	11.8
A little	49	32.2
A lot	85	55.9
The benefits of improving blood sugar control?
Not at all	18	11.8
A little	49	32.2
A lot	85	55.9
How to manage pregnancy and diabetes?
Not at all	51	33.6
A little	46	30.3
A lot	55	36.2

SD, standard deviation

Discussion

Marshallese living in the RMI experience extremely high prevalence of T2DM [1–5]. To assist Marshallese individuals in managing the effects of an existing T2DM diagnosis, it is important to understand their T2DM-related self-care behaviors and the knowledge Marshallese patients with T2DM have of managing the disease. The current study sought to understand self-care behaviors and self-reported diabetes knowledge among a sample of Marshallese adults living in the RMI.

Overall, our results show many Marshallese previously diagnosed with T2DM did not engage in adequate self-care behaviors, including checking their blood sugar and examining their feet. These findings are similar to studies done previously in the US and the RMI, which showed low engagement with self-care activities [29, 35]. For example, one study assessing self-care behaviors among 111 Marshallese adults living in the US found low engagement in diabetes-related self-care behaviors. Among the seven self-care activities examined (maintaining a normal weight, exercising regularly, checking glucose daily, having a biennial eye exam, taking a diabetes course, having an annual foot exam, and having an annual doctor visit), Marshallese participants integrated less than an average of four behaviors into their routine [35]. This is particularly concerning given the high percentage of participants in the present study with an HbA1c indicative of uncontrolled diabetes. Almost two-thirds of the participants had an HbA1c greater than 9.0%; by comparison, 50% of US adults with uncontrolled T2DM have an HbA1c greater than 9.0% [36].

Additionally, participants reported having forgone needed medical care and medication due to issues with cost and/or access. Healthcare is remote and underfunded in the RMI, and in combination with a high unemployment rate and low minimum wage, many Marshallese face difficulties in accessing and paying for medical care and supplies [7, 37]. Transportation is also an issue, as many healthcare providers are not located within walking distance and many Marshallese do not own/are unable to access personal transportation [38, 39]. These findings are similar to studies assessing barriers to medical care access and medication adherence in the US and in the RMI, highlighting the importance of addressing financial and transportation-related barriers to care [16, 29, 40–43].

Further, many participants reported low levels of diabetes self-management knowledge. However, the results do show some higher levels of self-reported diabetes self-management knowledge when compared to a previous study done in the RMI [29]. The results provide evidence of the work being done in the RMI to educate Marshallese on diabetes self-management [27]. Although an improvement, the combination of a high proportion of individuals with uncontrolled T2DM, inadequate self-care, insufficient knowledge about diabetes self-management, and barriers to healthcare does place the participants at a greater risk of life-threatening irreversible health complications, such as cardiovascular diseases, stroke, nephropathy, retinopathy, neuropathy, and infections leading to amputations [14, 15]. Further, the expense of uncontrolled diabetes affects the economy of the RMI; in 2017, healthcare expenditures were 16.34% of the RMI’s gross domestic product, similar to the US [44]. More work is needed to educate Marshallese communities in the RMI on the importance of diabetes self-management and increase engagement in diabetes self-management behaviors, both to improve the health of Marshallese living in the RMI and to lower government healthcare spending.

Limitations

There are several limitations to the results presented here. The convenience sample limits the ability to generalize the results to broader Marshallese and Pacific Islander communities, and the sampling method limits comparisons to other studies. Further, diabetes self-care behaviors and diabetes self-management knowledge are self-reported by the participant. Self-reports do carry a risk of bias, including the adjustment of participant responses to be more socially desirable [45] if the interviewer knows the participant, as is often the case with CBPR [46]. Although social desirability may play a role in the responses, prior work has shown the effect is limited even for sensitive questions (e.g., substance use) [46].

Conclusion

The findings are a valuable addition to the literature on diabetes self-care behaviors and diabetes self-management knowledge among Marshallese living in the RMI. The results demonstrate the need for further work in improving engagement in diabetes self-care by Marshallese living in the RMI. It is important to consider the evidence of gains in diabetes self-management knowledge by additional resources to individuals with T2DM in an effort to increase engagement in self-care behaviors. Increased engagement in self-care may help Marshallese with T2DM to improve control of their T2DM and avoid long-term health complications.

Highlights.

• Adults with T2DM in the RMI did not engage in adequate diabetes self-care behaviors

• Participants reported forgoing medical care and medication due to cost and access

• Participants reported low levels of diabetes self-management knowledge

• Diabetes self-management knowledge was higher than in previous studies

• Further work is needed to continue improving diabetes self-care in the RMI