Diabetes Island: Preliminary Impact of a Virtual World Self-Care Educational Intervention for African Americans With Type 2 Diabetes

Abstract

Background Diabetes is a serious worldwide public health challenge. The burden of diabetes, including prevalence and risk of complications, is greater for minorities, particularly African Americans. Internet-based immersive virtual worlds offer a unique opportunity to reach large and diverse populations with diabetes for self-management education and support. Objective The objective of the study was to examine the acceptability, usage, and preliminary outcome of a virtual world intervention, Diabetes Island, in low-income African Americans with type 2 diabetes. The main hypotheses were that the intervention would: (1) be perceived as acceptable and useful; and (2) improve diabetes self-care (eg, behaviors and barriers) and self-care related outcomes, including glycemic control (A1C), body mass index (BMI), and psychosocial factors (ie, empowerment and distress) over six months. Methods The evaluation of the intervention impact used a single-group repeated measures design, including three assessment time points: (1) baseline, (2) 3 month (mid intervention), and (3) 6 month (immediate post intervention). Participants were recruited from a university primary care clinic. A total of 41 participants enrolled in the 6 month intervention study. The intervention components included: (1) a study website for communication, feedback, and tracking; and (2) access to an immersive virtual world (Diabetes Island) through Second Life, where a variety of diabetes self-care education activities and resources were available. Outcome measures included A1C, BMI, self-care behaviors, barriers to adherence, eating habits, empowerment, and distress. In addition, acceptability and usage were examined. A series of mixed-effects analyses, with time as a single repeated measures factor, were performed to examine preliminary outcomes. Results The intervention study sample (N=41) characteristics were: (1) mean age of 55 years, (2) 71% (29/41) female, (3) 100% (41/41) African American, and (4) 76% (31/41) reported annual incomes below US $20,000. Significant changes over time in the expected direction were observed for BMI (P<.02); diabetes-related distress (P<.02); global (P<.01) and dietary (P<.01) environmental barriers to self-care; one physical activity subscale (P<.04); and one dietary intake (P<.01) subscale. The participant feedback regarding the intervention (eg, ease of use, interest, and perceived impact) was consistently positive. The usage patterns showed that the majority of participants logged in regularly during the first two months, and around half logged in each week on average across the six month period. Conclusions This study demonstrated promising initial results of an immersive virtual world approach to reaching underserved individuals with diabetes to deliver diabetes self-management education. This intervention model and method show promise and could be tailored for other populations. A large scale controlled trial is needed to further examine efficacy.

The Use of Videogames, Gamification, and Virtual Environments in the Self-Management of Diabetes: A Systematic Review of Evidence

“…1), and 10 studies met all the inclusion criteria. Four studies were based on videogames, [19][20][21][22] three studies used VEs, [23][24][25] and three studies used principles of gamification for the management of diabetes. [26][27][28] Three studies reviewed used a randomized control trial (RCT) design.…”

“…[26][27][28] Three studies reviewed used a randomized control trial (RCT) design. 19,22,25 Five were quasi-experimental, 23,24,[26][27][28] one was a qualitative focus group discussion, 20 and one study was of an independent measures experimental design. 21 In addressing RQ1, a summary of the reviewed studies' characteristics is presented in Table 1.…”

“…Three studies adopted the Second Life program as a medium for intervention. [23][24][25] Two used mobile phones. 20,27 One study used a humanoid robot called ''Nao,'' 26 one used modified exercise cycles with a videogame attachment, 19 one used the Nintendo Ò (Kyoto, Japan) WiiÔ, 22 and one developed an Flash videogame.…”

“…Sample sizes for a majority studies reviewed were small, having fewer than 50 participants. [19][20][21]23,24,26,27 Only three studies had sample sizes above 50. 22,25,28 Age.…”

“…21,[26][27][28] The remaining studies recruited adult and elderly participants, with individuals ranging in age from 38 to 88 years. 19,20,[22][23][24][25] Study attrition rates. Dropout rate statistics were not available for most studies.…”