Development and evaluation of a blood glucose monitoring YouTube video for Marshallese patients using a community-based participatory research approach

Abstract

The purpose of this study was to (a) describe the development of a culturally appropriate glucose monitoring video using a community-based participatory research (CBPR) approach and (b) assess the cultural appropriateness and effectiveness of the video. The topic of the video – using a glucometer and the importance of performing blood glucose checks – was chosen by Marshallese community stakeholders. The video was produced in Marshallese with English subtitles and disseminated through YouTube. Participants were recruited from August 16, 2016, to September 12, 2016, in a diabetes clinic that serves Marshallese patients in northwest Arkansas. Fifty participants completed a survey at pre and post intervention, with questions capturing demographic information and questions on glucose monitoring self-efficacy using an adapted version of the Stanford Patient Education Research Center’s Diabetes Self-Efficacy Scale. Twenty of those participants who completed the survey also completed semi-structured interviews that assessed cultural appropriateness and effectiveness of the video. Participants reported significant increases in self-efficacy related to glucometer use and the importance of performing blood glucose checks (p < .001) and a 1.45% reduction in A1C between pre-intervention and 12 weeks post-intervention (p = .006). Qualitative results indicated the video was both culturally appropriate and effective. The findings of this study were consistent with evidence in the literature, which shows health education videos can be effective at improving health behaviors. Using a CBPR approach to prioritize video topics, and including members of the community in the creation and dissemination of the videos, could aid in ensuring the videos are effective and culturally appropriate.

Introduction

Marshallese are Pacific Islanders from the Republic of the Marshall Islands. The Republic of the Marshall Islands was a former Trust Territory of the US and the primary location of the US nuclear testing program (Barker, 2012; McElfish, Hallgren, & Yamada, 2015). The Republic of the Marshall Islands is now an independent nation that is part of the US Affiliated Pacific Islands through a Compact of Free Association signed in 1986 between the US and the Republic of the Marshall Islands (Barker, 2012). Marshallese migration from the Republic of the Marshall Islands to the US has increased steadily since 1986, and Arkansas has the largest Marshallese population in the continental US (Hixson, Hepler, & Kim, 2012). Through the policies outlined in the Compact of Free Association, Marshallese can come to the US without a visa (McElfish, Purvis, et al., 2016). When the Compact of Free Association was signed, Marshallese were eligible for Medicaid, but were not included in the eligibility requirements after the passage of the Personal Responsibility and Work Opportunity Reconciliation Act in 1996 (McElfish, Purvis, et al., 2016).

Since 2013, researchers at the University of Arkansas for Medical Sciences (UAMS) have participated in a community-based participatory research (CBPR) partnership with Marshallese community stakeholders in northwest Arkansas, where the majority of Marshallese Arkansans reside (McElfish, Kohler, et al., 2015). CBPR engages nontraditional partners and honors their unique contributions at all phases of the research process from prioritizing the research needs to disseminating the findings (Cornwall & Gaventa, 2001; Gaventa & Cornwall, 2006; Israel, Schulz, Parker, & Becker, 1998; Mendenhall et al., 2010; Minkler, M., & Wallerstein, N., (eds.), 2008; Minkler, Blackwell, Thompson, & Tamir, 2003; O’Toole, Aaron, Chin, Horowitz, & Tyson, 2003; Viswanathan et al., 2004). Nurse educators and nurse researchers have a long and successful history of implementing CBPR in a manner that allows community knowledge to be integrated in to interventions and the research process so that it is more culturally-acceptable (Amiri & Zhao, 2019; Arlotti, Cottrell, Lee, & Curtin, 1998; Mahrer-Imhof, Hediger, Naef, & Bruylands, 2014; Prieto, Zuleta, & Rodríguez, 2016; Rossman, Greene, & Meier, 2015).

CBPR has demonstrated effectiveness in building alliances with minority, immigrant communities to improve health when there are disparities resulting from systematic disadvantage, racism, and historical trauma (Vaughn, Jacquez, Lindquist-Grantz, Parsons, & Melink, 2016). UAMS has worked with Marshallese community stakeholders in a CBPR partnership to identify the community’s top health concerns and collaboratively address them by involving Marshallese community stakeholders in the research process. This CBPR partnership is described in multiple article elsewhere (McElfish, Kohler, et al., 2015; McElfish, Long, et al., 2017; McElfish et al., 2017; McElfish, Moore, Laelan, & Ayers, 2018).

Marshallese suffer from disproportionate rates of type 2 diabetes mellitus (T2DM) compared to the general US population. A pilot study that included church-based health screenings demonstrated that more than a third (38.4%) of a sample (n = 401) of Marshallese adults in northwest Arkansas have glycated hemoglobin (A1C) levels indicative of T2DM (McElfish, Rowland, et al., 2017), a rate more than three times as high as the general US adult population (12.2%) (Centers for Disease Control and Prevention, 2017). The same study documented that 46.4% did not have insurance (McElfish, Rowland, et al., 2017). As part of the CBPR approach, the Marshallese community identified addressing T2DM as a top priority (Hallgren, McElfish, & Rubon-Chutaro, 2015; McElfish, Kohler, et al., 2015; McElfish, Moore, et al., 2016; McElfish et al., 2017). The CBPR team discussed a number of possible solutions for addressing T2DM, resulting in a request from Marshallese stakeholders that the UAMS research team partner with them to develop short health education videos focused on T2DM self-management behaviors.

Self-management behaviors, including daily monitoring of blood glucose, have been shown to be effective at improving glycemic control and preventing or delaying the development of diabetes-related complications and mortality (Holman, Paul, Bethel, Matthews, & Neil, 2008; Jones et al., 2003; Powers et al., 2015). However, rates of self-management behaviors and glycemic control are low in the general T2DM patient population (Ali et al., 2013), and even lower in many racial and ethnic minority groups (Campbell, Walker, Smalls, & Egede, 2012; Harris, Eastman, Cowie, Flegal, & Eberhardt, 1999; Karter, Ferrara, Darbinian, Ackerson, & Selby, 2000; Mayberry, Bergner, Chakkalakal, Elasy, & Osborn, 2016; Oster et al., 2006). Studies have demonstrated that culturally appropriate diabetes educational programs developed for racial and ethnic minority groups have resulted in improved T2DM knowledge self-management behaviors, and glycemic control (Ricci-Cabello et al., 2014; Ryan, Jennings, Vittoria, & Fedders, 2013; Sinclair et al., 2013). Therefore, a culturally appropriate, Marshallese language education video focused on glucometer use and the basics of blood glucose monitoring and control was chosen by Marshallese community stakeholders as the topic for the first health education video.

YouTube, one of the most visited websites in the world (second only to Google) (Alexa, 2017), is increasingly used as a platform for sharing videos on public health topics including immunizations, human papillomavirus, prostate cancer, obesity, and heart attack (Madathil, Rivera-Rodriguez, Greenstein, & Gramopadhye, 2015). Videos uploaded to YouTube are user-generated, meaning anyone can make a health-related video and post it to the site. Without formal content screening processes, the health information contained in YouTube videos is often misleading or incorrect (Madathil et al., 2015). However, YouTube videos produced by nurses and other health professionals, have the potential to communicate evidence-based health information to larger audiences while using fewer resources than traditional health education mediums. Video interventions targeted at changing health behaviors have been shown to be effective for a range of behaviors including self-management adherence, disease treatment compliance, prostate cancer screening, HIV testing, and breast self-examination (Tuong, Larsen, & Armstrong, 2014). Video interventions may be especially effective for populations with low health literacy (Sobel et al., 2009). While there have been limited video interventions developed for immigrant populations, the available research is promising. For example, health education videos produced in immigrant participants’ native language have been shown to increase knowledge regarding coronary artery disease, contraceptive use, mammography, and infant feeding (Garbers et al., 2015; Goel & O’Conor, 2016; Latif, Ahmed, Amin, Syed, & Ahmed, 2016; Scheinmann, Chiasson, Hartel, & Rosenberg, 2010). Furthermore, videos that depict real people performing tasks, often referred to as practice presentations, have been shown to be more effective at modifying health-related behaviors than didactic presentations (Abu Abed, Himmel, Vormfelde, & Koschack, 2014).

Despite promising early results for video interventions, few studies have tested the effectiveness of YouTube videos in providing health education and evidence is even sparser on the effectiveness of YouTube-based health education interventions specifically designed for minority immigrant populations whose first language is not English.

To fill this gap, we developed and tested a health education video focused on blood glucose monitoring and control. The health education video was designed for Marshallese with diabetes as part of a CBPR partnership,(McElfish, Kohler, et al., 2015; McElfish et al., 2017; McElfish et al., 2018) and the video was disseminated through YouTube. We hypothesized that participants who viewed the video would report increased understanding and self-efficacy related to checking their blood glucose and the meaning of their results. Furthermore, we hypothesized that participants would find the video culturally appropriate due to the use of a Marshallese family practice physician and the integration of CBPR principles throughout the conception and production of the video.

Methods

Developing the video through CBPR

The lead author collaboratively developed the video script and format with five Marshallese community members, two Marshallese community health workers, a Marshallese family practice physician, and a Pacific Islander Health Education researcher. The primary learning objectives for the video were to increase self-efficacy and participants’ understanding of: (a) how to use a glucometer to check blood glucose, (b) what the numbers on the glucometer mean in the context of managing T2DM, (c) information health care providers need about blood glucose check results to help best manage T2DM, (d) how diet and exercise can affect blood glucose check results, and (e) what to do if blood glucose levels are too high or too low.

The video was produced in Marshallese with English subtitles, using plain language for those with low health literacy (Stableford & Mettger, 2007). The video depicted a Marshallese physician explaining the importance of performing regular blood glucose checks. Next, the video showed a Marshallese patient actor performing a blood glucose check while explaining each step in detail and showing close-ups of all glucometer equipment. The actor then explained what the resulting numbers meant and how blood glucose numbers can fluctuate throughout the day and after eating. Finally, the video depicted the physician going over the patient’s results and explaining how important it is to keep track of results and share them with their health care provider. Marshallese community members, Marshallese community health workers, a Marshallese physician, Pacific Islander health researchers, a nurse health educator, a certified diabetes educator, and the lead author collaboratively edited the video for both medical accuracy and cultural appropriateness. The video is 3 minutes, 31 seconds in length. The video was viewed during the clinic visit and then could be viewed as many times as the participants wanted to after the clinical visit.

Study Setting

Participants were recruited from the UAMS North Street Clinic from August 16, 2016, to September 12, 2016. Marshallese have limited access to health care services due to language barriers and high uninsured rates (Hagiwara, Yamada, Tanaka, & Ostrowski, 2015; McElfish, Hallgren, et al., 2015), and the North Street clinic provides care regardless of patients’ ability to pay. Bilingual (English and Marshallese) Marshallese staff are an integral part of the clinic team, assisting patients and providers to ensure all services are effectively communicated and culturally appropriate (McElfish, Hudson, et al., 2017). All patients treated for T2DM in the clinic, regardless of their participation in research, are provided a glucometer and glucometer strips free of charge.

Recruitment and consent

Patients meeting the study inclusion criteria (at least 18 years of age, Marshallese, T2DM diagnosis by a health care provider) were approached by a bilingual research staff member in the clinic before their visit with their health care provider, provided information about the study, and asked if they were interested in participating in the study. Bilingual research staff completed at least 15 hours of training that addressed consent procedures, quality control, patient confidentiality, data security, and human participant protections. The recruiting study staff reviewed the consent information orally with potential participants and gave them a copy of the consent document. Participants were provided the opportunity to have any study-related questions answered by research staff. The individuals who agreed to participate provided written consent. The consent included a Health Insurance Portability and Accountability Act (HIPAA) release for medical record abstraction. All materials, including consent, were provided in the participants’ language of choice (English or Marshallese). All study procedures were reviewed and approved by the UAMS Institutional Review Board (IRB #205428). Those who took part in the study were provided a $20 gift card at each data collection event.

Data Collection

Using a CBPR approach and extensive input from Marshallese community stakeholders and a Marshallese physician, a culturally appropriate Glucose Monitoring Self-Efficacy Scale was developed and pilot tested. The survey instrument assessed participants’ self-efficacy related to glucometer usage, performing regular blood glucose checks, and understanding blood glucose check results. The instrument was adapted from the eight-item Stanford Patient Education Research Center’s Diabetes Self-Efficacy Scale (Scheinmann et al., 2010) to include language that is relevant to the Marshallese community. The Stanford survey was originally developed and tested in Spanish and has been shown to be reliable in both Spanish and English (Lorig, Ritter, Villa, & Piette, 2008; Lorig, Ritter, Villa, & Armas, 2009).

Survey items begin with “How confident do you feel that you…” and assess specific aspects of performing blood glucose checks (e.g., “How confident do you feel that you have the time to check your blood glucose daily?”; “How confident do you feel that you can understand the blood glucose numbers and what to do with those numbers?”). The instrument was translated from English into Marshallese by bilingual research staff. The eight self-efficacy questions were measured on a three-point Likert scale. Participants were asked to express their level of confidence regarding the use of glucometers and understanding blood glucose check results (Not at all Confident, 0; Moderately Confident, 1; Totally Confident, 2). The eight-item survey yields a single cumulative scale score (pre- and post-intervention Cronbach’s α’s > .90). Participants were also asked to report the number of times per day/per week/per month they check their blood glucose. For purposes of analysis, we converted answers reported as number of times per week or as number of times per month into number of times per day.

The survey also included demographic information on sex, age, and race/ethnicity. The survey was administered before and after participants viewed the video (pre- and post-intervention) in a private room in the North Street clinic. On the same day that we collected survey data, we obtained body mass index (BMI) and A1C as part of their regularly scheduled check-up and abstracted from their medical records.

Twenty participants were seen again at the clinic for a check-up between 12 and 17 weeks after initial data collection. For those 20 participants, a second A1C was obtained during their visit and abstracted from their medical records. These 20 participants also participated in follow-up semi-structured interviews, which had additional open-ended questions to elicit information regarding the effectiveness and cultural appropriateness of the video. The semi-structured interview guide encouraged participants to speak in-depth about their perceptions of and experiences with the video and ensured that study staff were consistent in the questions asked of all participants. The interview guides is provided as supplemental material. Interviews were conducted by bilingual Marshallese study staff trained in qualitative interviewing techniques. Interviews were approximately 10–15 minutes in length, audio recorded, transcribed verbatim, and then translated from Marshallese to English by a trained Marshallese translator and then verified by a second translator.

Data Analysis

Descriptive statistics for all demographic and biometric items were calculated. Pre/post-intervention self-efficacy survey responses were analyzed using paired sample t-tests. Mean changes for individual survey items were examined to determine the topics for which the video produced the greatest changes in self-efficacy. Changes in A1C were also analyzed using paired sample t-tests.

The interviews were coded using content analysis, and summarized by the lead author and a second co-investigator. Content analysis is an effective strategy for the qualitative descriptive design utilized for this study (Neergaard, Olesen, Andersen, & Sondergaard, 2009; Sandelowski, 2000). It allows the researchers to document participants’ experiences with and perceptions of the YouTube video in their own words. Initial coding began by naming each data segment with short summations to organize the data for more focused codes. Data was coded for both a priori and emergent themes and collaboratively discussed to ensure scientific rigor and intercoder agreement. Codes were confirmed by a third co-investigator. A summary of the qualitative responses is provided with quotes that best represent the feedback from participants.

Results

Participant Enrollment

From August 16, 2016, to September 12, 2016, 77 patients were identified as eligible for the study. Of the 77 eligible patients, 52 were approached during their regularly scheduled clinic visits and offered information about the study and the opportunity to participate. Twenty-five eligible patients were not approached due to patient flow constraints and a lack of study staff availability. Of the 52 patients approached for participation, two gave soft refusals (e.g. “not this time,” “too busy today”), and 50 (96%) agreed to participate and provided informed consent. Twenty participants returned for their regularly scheduled appointments approximately 12 weeks later and participated in qualitative interviews (See Figure 1).

Figure 1.

Recruitment and enrollment flow diagram

Participant Characteristics

All study participants identified as Marshallese and were at least 18 years of age. The average age of the participants was 53 years (SD = 11.5). Most (64.0%) of the participants were female. The average BMI of the participants was 31.3 (SD = 6.1). Less than a quarter (18.6%) of participants had a healthy/normal BMI (between 18.5 and 24.9), while all others were overweight (27.9%) or obese (53.5%). The average A1C of the participants was 10.8% (95 mmol/mol). More than two-thirds (70.0%) of participants had an A1C greater than 9.0% (75 mmol/mol), indicating poor glycemic control. The majority (72.0%) of participants indicated they check their blood glucose at least twice per day (See Table 1).

Table 1.

Participant demographics

Characteristic	N (%) or Mean ± SD
Age	53.0 ± 11.5
18–29	0 (0.0)
30–39	7 (14.0)
40–49	17 (34.0)
50–59	9 (18.0)
60–69	13 (26.0)
70+	4 (8.0)
Sex
Female	32 (64.0)
Male	18 (36.0)
Race/Ethnicity
Marshallese/Pacific Islander	50 (100.0)
BMI	31.3 ± 6.1
Underweight (< 18.5)	0 (0.0)
Normal (18.5 – 24.9)	8 (18.6)
Overweight (25.0 – 29.9)	12 (27.9)
Obese (≥ 30.0)	23 (53.5)
A1C (%)	10.8 ± 2.6
A1C (mmol/mol)	95 ± 28.4
How often do you check your blood glucose?
Never	5 (10.0)
Less than once per day	4 (8.0)
Once per day	5 (10.0)
Twice or more per day	36 (72.0)

Note: N = 50. Only valid percentages are shown. SD = standard deviation. BMI = body mass index. A1C = glycated hemoglobin.

Quantitative results

Results of the pre/post-intervention surveys indicated statistically significant increases between participants’ pre-intervention scores and their post-intervention scores on the self-efficacy scale (See Table 2). According to paired sample t-tests, the participants’ post-intervention scores (M = 14.41, SD = 2.83) were significantly higher than their pre-intervention scores (M = 10.93, SD = 4.34), t(45) = 6.540, p < .001, d = .95. Furthermore, analysis of pre/post-intervention means of the eight survey items indicated the video intervention produced statistically significant changes in all but one item. Specifically, the item assessing knowledge of what to do when blood glucose is too high or too low produced a non-significant increase (p = .073). All other items showed significant gains in self-efficacy between pre- and post-intervention (p’s ≤ .003) (Table 2). Results also indicated statistically significant decreases in A1C among the 20 participants who returned to the clinic for a follow-up visit. According to the paired sample t-test, the participants’ follow-up A1C’s (M = 10.2% [88 mmol/mol], SD = 2.01% [22.0 mmol/mol]) were significantly lower than their pre-intervention A1C’s (M = 11.69% [104 mmol/mol], SD = 2.36% [25.8 mmol/mol]), t(19) = 3.066, p = .006, d = 0.66 (See Table 3).

Table 2.

Differences in pre-intervention and post-intervention scores for the Glucose Monitoring Self-Efficacy Scale and individual scale items

Measure	Pre-intervention score (SD)	Post-intervention score (SD)	n	t	p
Glucose Monitoring Self-Efficacy Scale	10.93 (4.34)	14.41 (2.83)	46	6.540	<.001
Individual Scale Items
1. How confident do you feel that you understand the purpose of checking your blood glucose?	1.34 (0.69)	1.84 (0.37)	50	5.466	<.001
2. How confident do you feel that you know how to use your glucometer?	1.36 (0.72)	1.78 (0.42)	50	4.628	<.001
3. How confident do you feel that you can get the glucometer strips and lancet that you need?	1.16 (0.82)	1.78 (0.42)	50	5.280	<.001
4. How confident do you feel that you have the time to check your blood glucose daily?	1.42 (0.68)	1.88 (0.33)	48	5.143	<.001
5. How confident do you feel that you can understand the blood glucose numbers and what to do with those numbers?	1.33 (0.69)	1.77 (0.47)	48	4.449	<.001
6. How confident do you feel that you understand how to use your blood glucose numbers to manage diabetes; for example, different foods and exercise?	1.34 (0.72)	1.78 (0.42)	50	4.610	<.001
7. How confident do you feel that you know what to do when your blood glucose level goes higher or lower than it should be?	1.54 (0.61)	1.70 (0.54)	50	1.830	.073
8. How confident do you feel that you can judge when the changes in your blood glucose readings mean you should visit the doctor?	1.40 (0.70)	1.72 (0.50)	50	3.175	.003

Note: Analyses conducted with paired sample t-test. Analyses are based on the number of participants who fully completed each measure. SD = standard deviation.

Table 3.

Differences in pre-intervention and post-intervention A1C

Measure	Pre-intervention mean (SD)	Post-intervention mean (SD)	n	t	p
A1C (%)	11.69 (2.36)	10.24 (2.01)	20	3.066	.006
A1C (mmol/mol)	104 (25.8)	88 (22.0)

Note: Analysis conducted with paired sample t-test. SD = standard deviation. A1C = glycated hemoglobin.

Qualitative results

When participants were asked “What did you learn from the video?” all mentioned monitoring or controlling blood glucose and/or proper glucometer use. For example, one participant stated they learned “how to manage my diabetes and how to take my medications, especially how to control my blood glucose,” and “how to use the glucometer properly.” Another participant summarized, “it helps me to better understand how to manage my diabetes.” Another stated, “the video shows us how to use the glucometer machine properly.”

When participants were asked if there were parts of the video that were confusing, most responded that there were not. Three participants noted, however, that parts of the video were confusing. Among those, one stated, “I can’t recall what numbers were good and not,” and another shared that they were confused about “the part that showed the number dial on the lancet device.”

When asked about parts of the video that were most helpful or easy to understand most responses focused on how “it shows me how to check my blood glucose properly,” with several participants mentioning the importance of recording their blood glucose check results. As one participant summarized, “[the video] explained clearly how to use the log and write down the numbers.” Participants noted that the video helped them understand the process “step by step.” As one participant noted, “[the video] is helpful because it reminds us all of the right way to use the glucometer correctly from steps one, two, three.” Participants also noted that the video helped them understand “the results of our blood glucose [checks].”

In addition to the responses regarding blood glucose monitoring, more than half of the participants noted that the information about the importance of diet when monitoring and controlling blood glucose was helpful. One participant stated “the part that mentioned a healthy diet” was particularly helpful. The participant continued, “I didn’t know that it was important to manage my diet…and how it is important to my blood glucose.”

When asked “Did you find the video to be culturally appropriate and effective?” all participants answered affirmatively, with many expressing appreciation for the inclusion of a Marshallese physician and Marshallese community members in the video. When asked “Do you think the length of the video was right – was it too short or too long?” most participants stated it was the right length. Two participants noted that the video should be longer. Finally, when asked “Would you recommend the video to others?” all of the participants stated that they would recommend it to others.

Discussion

Both quantitative and qualitative results support our hypotheses. Quantitative results demonstrated significant improvements in participants’ self-efficacy scores, thus supporting our first hypothesis. Qualitative results demonstrated that participants understood the purpose and content of the video and found it culturally appropriate, thus supporting our second hypothesis. The vast majority of participants stated that they found the content and length appropriate; however, some thought the video should be longer. Furthermore, both quantitative and qualitative results showed that there was still a level of uncertainty surrounding what the blood glucose numbers mean and what to do if they are too high or too low. There may be a potential to improve the video by lengthening it in order to add more detailed content.

Perhaps most exciting, among the 20 participants for which a second A1C was obtained through their regularly scheduled visit, a 1.45% (15.8 mmol/mol) reduction in A1C was achieved relative to pre-intervention. This reduction was both clinically and statistically significant.³² Although a reduction in A1C was not hypothesized, the observed improvement is encouraging. It should be noted, however, the reduction in A1C cannot be solely attributed to the video intervention. Additional testing, such as a randomized controlled trial, would be needed to determine the video’s unique effect on glycemic control.

The results of this pilot study are consistent with existing literature that shows health education videos, including those utilizing YouTube, can be effective in improving health behaviors (Tuong et al., 2014). Furthermore, this study adds to the limited but growing body of literature that demonstrates educational videos may be particularly effective for low health literacy populations (Sobel et al., 2009), as well as immigrant populations who speak English as a second language (Garbers et al., 2015; Goel & O’Conor, 2016; Latif et al., 2016; Scheinmann et al., 2010). The results also provide evidence that a CBPR approach may be effective when developing educational videos, which is consistent with the general literature regarding the use of CBPR in developing health interventions (Cornwall & Gaventa, 2001; Gaventa & Cornwall, 2006; Israel et al., 1998; Minkler et al., 2003; Minkler, Wallerstein, & (eds.), 2008; O’Toole et al., 2003; Viswanathan et al., 2004), and consistent with literature that demonstrates that CBPR can be helpful when developing interventions to address T2DM (McElfish et al., 2018; Mendenhall et al., 2010).

Limitations

The study has several limitations that should be acknowledged. The sample size was small, with only 50 participants. Furthermore, all of the participants were recruited from a single clinic serving Marshallese patients in northwest Arkansas. This limits the generalizability of the research to other racial groups and other regions. The pilot study was a pre/post assessment with no control group; therefore, the reduction in A1C among the 20 participants who returned to the clinic for their regularly scheduled visit cannot be attributed solely to the video. Despite the study’s limitations, it provides promising evidence that short health education videos provided to participants in their native language may be effective at increasing self-efficacy and improving glycemic control.

Contributions to the Literature.

This study makes a significant contribution to the literature and may be particularly helpful to nurse health educators, as it is the first study to test the use of YouTube Videos to provide health education to Pacific Islander communities and to the Marshallese community specifically. It is also the first article to document the use of CBPR to develop a blood glucose monitoring video. This article adds an important contribution to the literature as the first article to document the use of the eight-item Stanford Patient Education Research Center’s Diabetes Self-Efficacy Scale to evaluate diabetes health education videos provided through YouTube and the first study to document the use of the Stanford Patient Education Research Center’s Diabetes Self-Efficacy Scale with Marshallese participants.

Future Research.

While the results of this pilot are encouraging and make a significant contribution to the literature, additional research is needed. Future fully powered randomized controlled trials should be conducted to understand the effectiveness of YouTube health education videos compared to a usual care control, additional face-to-face education, and/or written materials. Furthermore, future comparative effectiveness research should include cost-effectiveness aims to examine the potential to improve health outcomes at reduced cost compared to other health education methods.

Implications for Practice

Health education videos disseminated through YouTube may provide an inexpensive means of educating patients in their native language about self-management practices, such as performing regular blood glucose checks. YouTube videos have the added benefit of being able to be viewed multiple times and on demand, right when patients need the information without additional cost. Health education videos developed by nurses and other health care providers can provide accurate, timely, and cost-effective education that can be accessed whenever and wherever patients need that information and can also be viewed by family members and/or care givers who may be supporting the patients. The success of this study was predicated on using a CBPR approach to prioritize which videos should be created, and engaging Marshallese community stakeholders in the creation and dissemination of the videos. Nurse educators and other health care providers should utilize a CBPR approach to ensure health education videos are culturally appropriate and effective. Furthermore, the American Diabetes Association and American Association of Diabetes Educators should consider policy statements regarding the use of CBPR to address T2DM.