Abstract
Background:
Cinematic-virtual reality (cine-VR) has demonstrated improvements in cultural self-efficacy, diabetes attitudes, and empathy among healthcare providers, but its impact on health professional students is unknown. The purpose of the single-arm pre-post study was to examine the feasibility of this cine-VR diabetes training program as well as to assess changes in cultural self-efficacy, diabetes attitudes, and empathy among health professional students.
Method:
Participants viewed 12 cine-VR 12 simulations about a 72-year-old patient with type 2 diabetes. Pre-training and post-training, they completed the Transcultural Self-Efficacy Tool, Diabetes Attitude Scale-3, and Jefferson Scale of Empathy.
Results:
All 92 participants completed the full training. No participants reported technological difficulties or adverse events. For the assessment, 66 participants completed the pre-post measures for a response rate of 71.7% (mean age = 21.1 ± 1.9 years, 82.6% [n = 57] women; 84.1% [n = 58] white). We observed positive improvements in all three cultural self-efficacy subscales: “Cognitive” (t value = −4.705, P < .001), “Practical” (mean change = −.99, t value = −4.240, P < .001), and “Affective” (t value = −2.763, P = .008). Similarly, we observed positive improvements in four of the five diabetes attitude subscales: “Need for special training” (Z = −4.281, P < .001), “Seriousness of type 2 diabetes” (Z = −3.951, P < .001), “Value of tight glucose control” (Z = −1.676, P = .094), “Psychosocial impact of diabetes” (Z = −5.892, P < .001), and “Attitude toward patient autonomy” (Z = −2.889, P = .005). Finally, we observed a positive improvement in empathy (t value = −5.151, P < .001).
Conclusions:
Findings suggest that the cine-VR diabetes training program has the potential to improve cultural self-efficacy, diabetes attitudes, and empathy among health professional students. A randomized controlled trial is needed to confirm its effectiveness.
Over the next five years, the global market for medical education is expected to grow by $173.4 billion, with virtual reality (VR) propelling much of this growth. 1 Improved accessibility and affordability of VR head-mounted displays has led to its widespread adoption in medical education. 2 Moreover, demonstrated benefits in memory retention and the application of new knowledge increase the appeal of using VR in the classroom. 3 The immersive nature of VR promotes engagement and motivation, which in turn, improves cognitive processing of learned material.4,5
VR has the potential to enhance the patient-provider relationship. A high-quality patient-provider relationship is associated with improved diabetes outcomes, including improved self-care behaviors6,7 and lower hemoglobin A1c levels in type 2 diabetes. 8 Interventions that address the patient-provider relationship are needed considering some providers and providers-in-training view diabetes more negatively than other chronic conditions.9 -11 Moreover, several providers and providers-in-training still do not consider type 2 diabetes to be as serious as type 1 diabetes.11,12 Research shows that providers who view type 2 diabetes as less serious than type 1 diabetes perform fewer screenings for complications 13 and are less likely to intensify treatment.10,14 Patients who perceive negative attitudes from their providers report feeling stigmatized, which in and of itself can increase the risk for psychological distress and lower quality of life.15,16
Empathy, which encompasses knowing, comprehending, and perceiving what another person is experiencing, has been rated as a top priority in the patient-provider relationship 17 and has been shown to decrease all-cause mortality amongst research participants who reported higher empathy scores in their first year after diagnosis. 18 Cultural humility also plays a role in the patient-provider relationship because diabetes disproportionately affects certain populations. 19 The more providers understand and apply cultural constructs in diabetes care, the better patients will follow their diabetes management plan. 20 Thus, improving cultural awareness amongst providers has the potential to improve patient-provider relationships.21,22
Cinematic-virtual reality (cine-VR) is a unique form of VR that uses live images captured using a camera as opposed to computer-generated avatars and worlds. It leverages the techniques of cinema, including narrative storytelling, scripts, actors, lighting, framing, lens choices, and set design. For this study, we created a cine-VR program focused on type 2 diabetes and social drivers of health. The cine-VR simulations follow a patient, Lula Mae Tate, a 72-year-old woman with type 2 diabetes living in rural Ohio, USA. We previously conducted a pilot study with this cine-VR diabetes training program with health care providers. Findings from that pilot study suggested that the cine-VR diabetes training program has the potential to improve cultural self-efficacy and diabetes attitudes among health care providers; however, more research is needed to confirm its effectiveness.23,24 What is not known is how this cine-VR training impacts the cultural self-efficacy, diabetes attitudes, and empathy of health professional students.
The purpose of this single-arm pre-post study was to measure changes in cultural self-efficacy, diabetes attitudes, and empathy among health professional students before and after diabetes cine-VR training program. We also examined the feasibility and acceptability of implementing the cine-VR training program in the classroom as well as the retention of the training with health professional students.
Methods
Research Design
We utilized a single-arm pre-post study design to assess changes in health professional students’ cultural self-efficacy, diabetes attitudes, and empathy before and after participating in the diabetes cine-VR training program. In addition, we examined the feasibility of the cine-VR diabetes training program, acceptability of the training, retention with the health professional students, and utility of our assessment procedures.
Cine-VR Simulations
The fictional patient in our cine-VR simulations is Lula Mae Tate, a 72-year-old woman living in rural Ohio. She has had type 2 diabetes for 22 years. She also has hypertension, hyperlipidemia, and a body mass index (BMI) of 43.5 kg/m2. She is a widow; her husband died 27 years ago from a myocardial infarction. She has three adult children and seven grandchildren. She cares full-time for her adult son after he suffered a traumatic brain injury form serving in the army. Lula Mae and her adult son live in a house originally belonging to her grandparents. Her two adult daughters and grandchildren live on the same plot of family land in two trailer homes. Lula Mae and her family experience multiple social drivers of health, including food insecurity, lack of transportation, financial instability, unsafe housing, lack of access to care, and lack of social support. Her vitals are as follows: blood pressure 164/90 mmHg, height 64 inches, and weight 264 pounds. Her recent laboratory numbers are a hemoglobin A1c of 12.7% (high), total cholesterol 198 mg/dL (borderline), high-density lipoprotein cholesterol 32 mg/dL (low), triglycerides 260 mg/dL (high).
Participants viewed the cine-VR simulations using Pico G2 4K head-mounted displays. Specifically, participants viewed two traditional short films and ten cine-VR simulations that followed Lula Mae and her interactions with a primary care physician, pharmacist, family members, and community.
Throughout the series, participants experience Lula Mae’s struggles as well as the many ways her rural community came together to support her. The cine-VR episodes were accompanied by 12 brief debriefs, lasting three to five minutes in length, on the following topics: (1) Diabetes Bias and Burnout, (2) Food Insecurity, (3) Appalachian Cultural Strengths, (4) Transportation Barriers in Rural Areas, (5) Person-Centered Care, (6) Psychosocial Issues in Diabetes, (7) Financial Insecurity and the Cost of Diabetes Medications, (8) Lack of Social Support, (9) Appalachian Cultural Values, (10) Diabetes Complications, (11) Diabetes Comorbidities, and (12) Effective Patient-Provider Communication. Additional information about the cine-VR content was published previously.23,25
Ethics Approval
Ethics approval for the study was obtained from the Ohio University Office of Research Compliance Institutional Review Board (approval number: 19-X-211). In complying with federal, state, and local laws and regulations for human subjects, we ensured our research met the requirements set forth in the regulations on public welfare in Part 46 of Title 45 of the Code of Federal Regulations (45 CFR 46); the principles set forth in “The Belmont Report,” and the Helsinki Declaration of 1975.
Sample Size
We conducted an a priori power analysis using G*Power 3 software, 26 which determined a sample size of 54 students was estimated for 80% power at a 5% significance level (P < .05) to detect an effect size of d = 0.5.
Participants
We recruited adults aged 18 years and older who were able to read and write in English. Participants included health professional students enrolled at the Ohio University during the 2019-2020 academic year. Professional programs included doctor of osteopathic medicine, speech-language pathology, nutrition/dietetics (any level), physical therapy (any level), nursing (any level), social work (any level), child life specialist, athletic training, exercise physiology, physician assistant, psychology (any level), and pharmacy. There were no exclusion criteria.
Enrollment for the cine-VR diabetes training program opened on January 1, 2020 and closed on March 7, 2020. The cine-VR diabetes training program was screened at five health professional courses, including the following classes: Psychology of Aging (n = 42), Introduction to Interprofessional Education and Practice in Healthcare (n = 12), Nutrition: Trends in Diabetes (n = 18), Exercise Physiology Graduate Seminar (n = 5), and Area Health Education Center Scholars Seminar (n = 15). Each student participated in the cine-VR diabetes training with their own head-mounted display, but in a group setting with their classmates. All students enrolled in these five health professional programs were emailed an electronic, anonymous survey by their professor and not the research team. The research team had no association with the health professional students. This step was taken so that the participants would not feel any undue pressure to participate in the study. To link participants’ pre-response and post-response, we included three questions at the beginning of the survey that served as a unique identifier; this unique identifier has been successfully employed with previous research studies to protect participant anonymity. While all students viewed the diabetes VR training, participation in the study was completely voluntary. Note, recruitment ended in March 2020 due to the COVID-19 pandemic and the restrictions placed on in-person human subjects’ research per university mandate.
Measures
Feasibility and Acceptability: To assess the feasibility of the diabetes cine-VR training program, we examined recruitment and retention rates, length of time required to recruit, adherence rates, rate of completion of the cine-VR training program, and feasibility of the data collection measures, including completion rates of the measures. 27 To assess acceptability of the cine-VR training program, we assessed affective attitude of the participants and perceived burden to participate in the training program. 28
Participants completed the following measures:
Demographic form: participants self-reported age, gender, ethnicity, race, health professional program, year in program, and desired specialty choice.
Transcultural Self-Efficacy Tool-Multidisciplinary Healthcare Provider (TSET-MHP) 29 : a 83-item scale that assesses changes in self-efficacy for cultural knowledge, cultural practical skills, and cultural awareness. The 83 items are answered on a ten-point Likert-type scale ranging from 1 (not confident) to 10 (totally confident). This scale identifies three domains of learning: (1) Cognitive (ie, confidence in knowing the ways culture influences the care provided to patients of different cultural backgrounds), (2) Practical (ie, confidence in interviewing patients of different cultural backgrounds to learn about values and preferences for care), and (3) Affective (ie, confidence in accepting similarities and differences between cultural groups). 29 The internal consistency of each factor ranges from .94 to .98, and the reliability of the total instrument is .99. 29
Diabetes Attitude Scale-3 (DAS-3) 30 : a 33-item scale that measures five diabetes-related attitudes: (1) need for special training (Cronbach’s α = .67), (2) seriousness of type 2 diabetes (Cronbach’s α = .80), (3) value of tight glucose control (Cronbach’s α = .72), (4) psychosocial impact of diabetes (Cronbach’s α = .65), and (5) attitude toward patient autonomy (Cronbach’s α = .76). These 33 items are rated on a five-point Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree). The scale demonstrates good internal consistency and high content validity. 30
Jefferson Scale of Empathy Health Care Provider Students Version 31 : a 20-item scale that measures empathy. The 20 items are answered on a seven-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). Of these 20 questions, ten questions are worded positively and ten questions are worded negatively. The scale demonstrates good internal consistency (Cronbach’s α = .85).
Data Collection
Participants completed the demographic form and validated measures via the online questionnaire service Qualtrics (Provo, UT: Qualtrics). All participants provided online informed consent. Completion of the measures took approximately 15 to 20 minutes. Participants received a $15.00 gift card as compensation for participating in the study. To receive the gift card, participants clicked on a new Qualtrics link that was not connected to their responses; this ensured that their data were not linked to their identity. However, given the anonymous nature of the study, it was possible that participants received compensation without completing all of the pre-measure and post-measure.
Statistical Analysis
We assessed sociodemographic factors using descriptive statistics and presented them as means and standard deviations or sample size and percentages. Prior to conducting our analyses, we verified normality of the distribution using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Next, we verified homogeneity of variance with a Levene’s test. We performed paired t tests and Wilcoxon sign-rank tests to examine changes in cultural self-efficacy, diabetes attitudes, and empathy scores before and after the cine-VR training program. In addition, we calculated effect sizes using Hedges’ g, with a small effect = 0.2, medium effect = 0.5, and a large effect = 0.8. Finally, we conducted linear regression models examining the association of age, gender, race, type of program, and year in program with mean change scores in cultural self-efficacy, diabetes attitudes, and empathy. We defined statistical significance as a P value less than .05 and conducted analyses in SPSS statistical software version 28.0 (Chicago, IL: SPSS Inc.).
Results
We successfully recruited 92 health professional students from health professional classes in two months. All 92 health professional students participated in the cine-VR diabetes training and completed the full training with 100% retention. No participants reported technological difficulties or adverse effects (eg, nausea, dizziness) from wearing the head-mounted displays. We completed the trainings within the time frames of each class/seminar, which was approximately 90 minutes. Finally, acceptability of the cine-VR was reflected by positive feedback from the participants. We received no negative feedback from the participants. Furthermore, none of the participating students or professors perceived the cine-VR training to require too much effort or burden to participate during their class time. In fact, all five professors invited us to deliver the same training in their future classes.
For our assessment, a total of 69 of the 92 health professional students consented to participate in the study; however, only 66 participants completed both the pre-measure and post-measure (response rate of 71.7%). The mean age of participants was 21.1 ± 1.9 years, 57 (82.6%) identified as women, 12 (17.4%) identified as men, three (4.3%) identified as Hispanic or Latinx, two (2.9%) identified as Asian/Pacific Islander, three (4.3%) identified as black or African American, three (4.3%) identified as mixed race, and 58 (84.1%) identified as white (see Table 1). The most common health professional program was psychology (n = 17, 24.6%), and the majority of participants were in their fourth year (n = 27, 39.1%) of their program (see Table 1).
Participant Demographic Characteristics in Cinematic-Virtual Reality (Cine-VR) Study (n = 69).
Missing value “Year in Program,” immersive cine-VR group (n = 2).
Cultural Self-Efficacy
We observed positive improvements post-training in all three cultural self-efficacy subscales: “Cognitive” (mean change = −.89, t value = −4.705, P < .001, see Table 2), “Practical” (mean change = −.99, t value = −4.240, P < .001), and “Affective” (mean change = −.37 t value = −2.763, P = .008). The subscale with the largest magnitude of change was the “Practical” subscale, with a Hedges’ g of .54 suggesting a medium effect. We also conducted regression models with the mean change score for each subscale controlling for age, gender, race, type of program, and year in program. Age, gender, race, year in program, and type of program were not independently associated with the mean change scores in “Cognitive,” “Practical,” or “Affective” subscales (see Table 3).
Cultural Self-Efficacy, Diabetes Attitudes, and Empathy Scale Means Before and After the Cinematic-Virtual Reality (Cine-VR) Training (n = 66).
Regression Models Examining Associations Among Demographic Variables and Mean Change Scores in Cultural Self-Efficacy, Diabetes Attitudes, and Empathy (n = 66).
Diabetes Attitudes
We observed improvements post-training in four of the five diabetes attitude subscales: “Need for special training” (Z = −4.281 P < .001, see Table 2), “Seriousness of type 2 diabetes” (Z = −3.951, P < .001), “Psychosocial impact of diabetes” (Z = −5.92, P < .001), and “Attitude toward patient autonomy” (Z = −2.818, P = .005). We observed no change in “Value of tight glucose control” (Z = −1.676, P = .094). The subscale with the largest magnitude of change was “Psychosocial impact of diabetes,” which had a Hedges’ g of .98 indicating a very large effect. We conducted regression analyses with mean change scores for each subscale to examine associations by age, gender, race, type of program, and year in program. Only type of program was associated with the mean change score of “Value of tight glucose control” (b = −.055, P = .002); no other subscales had significant associations (see Table 3).
Empathy
We observed increases in empathy scores post-training (mean change = −9.02, t value = −5.151, P < .001). The Hedges’ g for the change in empathy was .59, indicating a medium effect. We also conducted a regression analysis examining the associations among age, gender, race, type of program, and year in program with the mean change in empathy; none of these variables were associated with the change in empathy scores (see Table 3).
Discussion
In this single-arm pre-post study, we examined the potential effectiveness of a cine-VR diabetes training program in improving cultural self-efficacy, diabetes attitudes, and empathy scores among health professional students. In addition, we examined the feasibility of conducting the cine-VR training with health professional students. Overall, our findings showed that the cine-VR diabetes training program was feasible to conduct in the classroom with health professional students. All participants completed the full training with 100% retention in the program. Also, we experienced no technological issues and participants reported no adverse effects. Importantly, the cine-VR was widely accepted by the participating health professional students as well as their professors. In our assessment study, findings suggest that that the cine-VR diabetes training may have contributed to improved cultural self-efficacy, four of the five diabetes attitudes scores, and empathy post-training. Future research with a larger, more diverse sample with a proper control condition is needed to determine the effectiveness of the cine-VR diabetes training in improving cultural self-efficacy, diabetes attitudes, and empathy.
VR has been referred to as the “ultimate empathy machine.” This term was first popularized by Christopher Milk, a filmmaker and immersive designer, during his 2015 TED talk. 32 It is theorized that cine-VR generates empathy through the depiction of emotionally charged content from the perspective of another. 33 Cine-VR offers users the unique opportunity to experience events viscerally from another’s point of view, which may serve as a technological trigger for empathy. 34 For example, in our study, the cine-VR depicted a compelling narrative about a 72-year-old woman with type 2 diabetes who experienced numerous barriers (ie, economic instability, food insecurity, lack of transportation, unhealthy housing) in small-town America. The storyline was crafted to elicit emotional reactions from the participant. We achieved this through Lula Mae’s actions and feelings. For example, she displayed traits valued in Appalachian culture, including loyalty, caregiving, and generosity. Participants watched Lula Mae search her kitchen cabinets trying to find enough food to feed her children and grandchildren. Lula Mae realizes that she does not have enough food, so she tells her family to go ahead and finish what is on the table because she made herself a separate plate. Participants also witnessed Lula Mae being mistreated by provider for an elevated Hemoglobin A1c. They see her change in disposition, withdrawn posture, poor eye contact, and hesitation to answer her provider’s questions during that medical visit. Participants did not need to relate to or like everything about Lula Mae to identify with her humanness. The goal of this storyline was to drive participants towards empathic concern, or a type of empathy that leads people to care about another person’s welfare.33,35 Building empathic concern among health care professionals is critical because it is the major contributing source of altruistic motivation, 36 or the desire to enhance the welfare of others. 37 In diabetes, demonstrating empathy to patients with type 2 diabetes is critical. A 2019 population-based cohort study found that patients newly diagnosed with type 2 diabetes who perceived high levels of empathy from their health care provider had a 40% to 50% lower risk of all-cause mortality at ten-year follow-up. 18
We observed potential improvements in cultural self-efficacy and diabetes attitudes post-training, with the exception of diabetes attitude “Value of Tight Glucose Control.” These findings echo the current literature on VR. A recent meta-analysis found that VR interventions at varying levels of immersion improved multiple social attitudes (eg, racial bias, gender bias, ageism). 38 Similarly, another recent review evaluating the effectiveness of low and highly immersive cine-VR studies in cultural learning, found that both contributed to the development of positive attitudes, greater interest in culture, and increased knowledge about visible and invisible cultural attributes. 39 Prior research combined with our preliminary findings support a full-scale study examining the effectiveness of the cine-VR diabetes training program to a proper control condition. Furthermore, conducting a cine-VR trial at different levels of immersion is another promising area of research. We have developed the cine-VR diabetes training as an online interactive website. Future research is needed to examine its effectiveness as well as compare it to the in-person delivery of the cine-VR with head-mounted displays. If both delivery modes are determined to be effective, the online interactive website will be an accessible educational tool for health care providers and trainees at no cost to users.
Limitations
Limitations of this study include the small, homogenous sample, data collected from one site, selection bias, social desirability bias, lack of a control group, and no long-term follow-up. While a final sample of 66 is small, it was sufficient for our single-arm pre-post study. Future research will require a larger, more diverse sample to reflect the current health care workforce. Relatedly, we collected data from one university in the Midwest; therefore, the findings are not generalizable to all health professional students. Next, our assessment findings may be susceptible to selection bias, as health professional students who volunteered to complete the measures may have been more willing or motivated to participate compared with other students. For example, students with diabetes or with close relationships with people with diabetes may have been more willing to participate. However, we did not collect this information, but we plan to for future research. The study may also be susceptible to social desirability bias if the participants felt they had to provide answers in accordance with society’s expectations rather than their own beliefs and attitudes. Finally, we did not include a proper control condition with randomization to groups and we did not include a long-term follow-up assessment to determine the sustained impact of the cine-VR diabetes training on cultural self-efficacy, diabetes attitudes, and empathy.
Conclusions
We demonstrated that our cine-VR diabetes training program was acceptable and feasible with health professional students at our Midwestern university. Furthermore, findings from our assessment study show that our cine-VR diabetes training program potentially improves cultural self-efficacy, diabetes attitudes, and empathy among health professional students. Future research should determine the effectiveness of this cine-VR diabetes training program via a randomized controlled study design with comparisons to a proper control condition. In addition, the randomized controlled trial should include a larger, more diverse sample recruited from multiple sites with long-term follow-up measures.
Footnotes
Abbreviations
Cine-VR, cinematic-virtual reality; VR, virtual reality.
Data Availability
The data are available upon request to the corresponding author.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was supported by the Osteopathic Heritage Foundation Ralph S. Licklider, D.O. Endowed Professorship in Behavioral Diabetes awarded to Elizabeth A. Beverly, PhD.
This study was part of the Medicaid Simulation Project funded by the Ohio Department of Medicaid and administered by the Ohio Colleges of Medicine Government Resource Center. The views expressed in the cine-VR simulations and this manuscript are solely those of the creators and do not represent the views of the state of Ohio or federal Medicaid programs.
Research Ethics and Patient Consent
The study was determined to be expedited by the Ohio University Office of Research Compliance (Institutional Review Board 19-X-211). Online informed consent with a downloadable document as well as written informed consent were available to participants prior to participation in the study. The Ohio University Office of Research Compliance reviewed and approved all procedures and materials. The Ohio University Office of Research Compliance conforms to the recognized guidelines of the US Federal Policy for the Protection of Human Subjects.
