Communication Skills Assessment Using Human Avatars: Piloting a Virtual World Objective Structured Clinical Examination

INTRODUCTION

Proficiency in communication skills is a core competency of residency training. We evaluated the feasibility, acceptability and applicability of a virtual world objective structured clinical examination that enables practice based learning and assessment of resident communication skills.

METHODS

A virtual clinical encounter situated in 2 practice settings was developed that uses a human avatar physician and a standardized patient. Following an online tutorial house staff participated in 4 communication tasks, including shared decision making, delivering bad news, obtaining informed consent and disclosing a medical error. Validated instruments and semistructured interviews were used to assess house staff acceptability and applicability of the platform. Three faculty members used ACS (Affective Competency Scale) and communication specific assessment instruments to evaluate house staff performance.

RESULTS

A total of 12 urology house staff completed the simulation. Direct costs were approximately $1,000. The virtual world was easy to use and immersive. Applicability directly correlated with presence (Pearson r = 0.67, p = 0.01) and co-presence (Pearson r = 0.8, p = 0.002). House staff identified problems with 1) limited nonverbal cues, 2) too much information presented and 3) a lack of immediate feedback. The ICC (intraclass correlation) of faculty assessments was high for ACS at 0.53 (95% CI 0.36-0.69) for single measures, 0.77 (95% CI 0.63-0.86) for average measures and less for other assessment instruments.

CONCLUSIONS

A virtual world objective structured clinical examination is a feasible, acceptable and applicable method of communication skills assessment. Improving nonverbal cues, focusing on individual skill sets and providing immediate feedback are measures to be adopted in future iterations of this platform.

Use of an online personal health record's Track Health function to promote positive lifestyle behaviors in Veterans with prediabetes

This pilot 3-month clinical trial investigated the feasibility, effectiveness, and acceptability of using the Track Health function of the Veterans Health Administration's personal health record for eliciting a more positive physical activity and dietary intake lifestyle in a sample of 38 overweight and obese Veterans with prediabetes. Comparisons between baseline and 3 months post-intervention indicated significant improvements in weight, physical activity, abdominal circumference, and blood pressure. Use of a personal health record that users can identify with and find usable and useful coupled with instruction targeting critical functionalities could potentially promote healthy behavioral lifestyle changes.

The Association of Graph Literacy With Use of and Skills Using an Online Personal Health Record in Outpatient Veterans

Personal health records (PHRs) are intended to increase patients' access to and ownership over their health care information for self-management purposes. The purpose of this study was to determine the association of graph literacy with adoption of an online PHR and, among adopters with self-reported skills, the frequency of use and intent to return to use the PHR . We conducted a cross-sectional survey of veterans receiving outpatient care. We measured health literacy, numeracy, graph literacy, and Internet and PHR adoption and use. We compared subgroups of veterans using analyses of covariance. We used hierarchical logistic regression models to estimate the effects of the literacy variables on PHR use. A total of 600 veterans (age = 22-94) participated in the survey. After we adjusted for known covariates, we found that adopters of a PHR were more likely to demonstrate higher health and graph literacy than nonadopters. Among PHR adopters, self-reported frequent and skillful users were more likely to have higher graph literacy than lower frequency and less skillful users. Adopters with higher intentions to return to use the PHR were more likely to show lower graph literacy than those less likely to return to use the PHR. Inadequate graph literacy was associated with lower adoption of a PHR and, among users, with lower self-reported frequent use and skills . As PHR use becomes more widespread, stakeholders will need to consider patients' levels of graph literacy when implementing PHRs.

Effects of an Avatar-Based Anti-Smoking Game on Smoking Cessation Intent

The purpose of this study was to compare the effects of a computer-based anti-smoking game on the intent and motivation to quit tobacco. Smokers with nicotine dependence were briefly exposed to an anti-smoking game with or without an avatar resembling the smoker's self. The computer-based anti-smoking game improved participants' immediate intent and motivation to quit smoking. Embedding an avatar resembling self into the game did not result in added benefits.

The Communication of Global Cardiovascular Risk by Avatars

Communicating numerical estimates of cardiovascular risk (CVR) to patients encourage risk reduction actions. Avatars may enhance the risk messages ability to improve persuasion to adhere to healthy behaviors. We compared the efficacy of a computer-based aid communicating CVR with and without animated avatars for improving intention to adhere to lifestyle changes. Males with intermediate to high CVR received their risk message in 2 versions: an avatar using voice; voice only. Forty-one participants completed the study. Intent to change lifestyle showed a significant effect favoring the avatar (moderate effect size). Intent to follow medical treatments also showed a significant effect favoring the avatar (moderate effect size). An avatar-based computer aid significantly increased participants' intention to adhere to positive behavioral changes.

Group-based differences in anti-aging bias among medical students

Medical students (MS) may develop ageist attitudes early in their training that may predict their future avoidance of caring for the elderly. This study sought to determine MS' patterns of explicit and implicit anti-aging bias, intent to practice with older people and using the quad model, the role of gender, race, and motivation-based differences. One hundred and three MS completed an online survey that included explicit and implicit measures. Explicit measures revealed a moderately positive perception of older people. Female medical students and those high in internal motivation showed lower anti-aging bias, and both were more likely to intend to practice with older people. Although the implicit measure revealed more negativity toward the elderly than the explicit measures, there were no group differences. However, using the quad model the authors identified gender, race, and motivation-based differences in controlled and automatic processes involved in anti-aging bias.

Communicating global cardiovascular risk: are icon arrays better than numerical estimates in improving understanding, recall and perception of risk?

OBJECTIVE

Experts recommend that adults have their global cardiovascular risk assessed. We investigated whether icon arrays increase understanding, recall, perception of CVR, and behavioral intent as compared with numerical information.

METHODS

Male outpatient veterans, at an intermediate to high cardiovascular risk participated in a randomized controlled trial of a computer tutorial presenting individualized risk. Message format was presented in 3 formats: percentages, frequencies, and frequencies with icon arrays. We assessed understanding immediately (T1) and recall at 20 min (T2) and 2 weeks (T3) after the intervention. We assessed perceptions of importance/seriousness, intent to adhere, and self-efficacy at T1. Self-reported adherence was assessed at T3.

RESULTS

One-hundred and twenty male veterans participated. Age, education, race, health literacy and numeracy were comparable at baseline. There were no differences in understanding at T1 [p = .31] and recall at T3 [p = .10]. Accuracy was inferior with frequencies with icon arrays than percentages or frequencies at T2 [p ≤ .001]. There were no differences in perception of seriousness and importance for heart disease, behavioral intent, self-efficacy, actual adherence and satisfaction.

CONCLUSION

Icon arrays may impair short-term recall of CVR.

PRACTICE IMPLICATIONS

Icon arrays will not necessarily result in better understanding and recall of medical risk in all patients.

Computer-Based Programmed Instruction Did Not Improve the Knowledge Retention of Medication Instructions of Individuals With Type 2 Diabetes Mellitus

Purpose

The purpose of this study was to investigate whether computer-based programmed instruction (CPI) with a dynamic avatar (DA) improves retention of medication information better than text (controls) or better than CPI with text (T-CPI), CPI with voice (V-CPI), or CPI with static avatar with text (SA-CPI).

Methods

Participants with type 2 diabetes mellitus were enrolled in a randomized controlled trial comparing TDS with CPI in 4 conditions (T-CPI, V-CPI, SA-CPI, and DA-CPI). CPI sequentially delivers segmented information in text or voice followed by a multiple-choice question. Immediately after the user selects an option, CPI delivers elaborated feedback. Satisfaction was measured immediately after the interventions, and medication knowledge was measured at 2 weeks.

Results

One-hundred fifty individuals (30 per group) with a mean age 62 years (standard deviation [SD] 7.99 years) participated. There were no baseline differences in race, body mass index, education, and health literacy. Medication knowledge retention at 2 weeks was not significantly different between the groups, df(4), F = 0.17, P = .95 (TDS, mean = 25.43, SD = 5.11; T-CPI, mean = 25.07, SD = 4.98; V-CPI, mean = 25.77, SD = 4.89; SA-CPI, mean = 25.83, SD = 5.31; and DA-CPI, mean = 24.93, SD = 6.25). Satisfaction scores were significantly lower for TDS, df(4), F = 3.11, P = .01.

Conclusions

CPI did not improve medication knowledge retention at 2 weeks. CPI led to higher patient satisfaction compared with controls.

Effects on performance of individual versus dyadic practice during an avatar-based three-dimensional virtual home safety simulation

This pilot study tests the effects on individual performance of dyadic versus individual practice in a 3D virtual world (VW) home safety assessment. Sixty medical students in three conditions (dyadic spatially separated with paired avatars DPA; individual with avatar IND; and dyadic spatially together with single avatar DSA) participated in a geriatric home safety simulation. The participants, via avatars, conducted an assessment of physical hazards. Participants then worked individually in a separate 3D VW home assessment. Dyadic practice, spatially together with a single avatar (DSA), improved individual performance in the subsequent 3D VW home assessment.

Health literacy, numeracy, and graphical literacy among veterans in primary care and their effect on shared decision making and trust in physicians

Studies reveal high levels of inadequate health literacy and numeracy in African Americans and older veterans. The authors aimed to investigate the distribution of health literacy, numeracy, and graph literacy in these populations. They conducted a cross-sectional survey of veterans receiving outpatient care and measured health literacy, numeracy, graph literacy, shared decision making, and trust in physicians. In addition, the authors compared subgroups of veterans using analyses of covariance. Participants were 502 veterans (22-82 years). Low, marginal, and adequate health literacy were found in, respectively, 29%, 26%, and 45% of the veterans. The authors found a significant main effect of race qualified by an age and race interaction. Inadequate health literacy was more common in African Americans than in Whites. Younger African Americans had lower health literacy (p <.001), graph literacy (p <.001), and numeracy (p <.001) than did Whites, even after the authors adjusted for covariates. Older and younger participants did not differ in health literacy, objective numeracy, or graph literacy after adjustment. The authors found no health literacy or age-related differences regarding preferences for shared decision making. African Americans expressed dissatisfaction with their current role in decision making (p =.03). Older participants trusted their physicians more than younger participants (p =.01). In conclusion, African Americans may be at a disadvantage when reviewing patient education materials, potentially affecting health care outcomes.

Medical students' attitudes toward obese patient avatars of different skin color

Physicians' biases for skin color and obesity may negatively affect health-care outcomes. Identification of these biases is the first step to address the problem. We randomized 128 U.S medical students into one of four animated videos of avatar physician-patient counseling sessions, varying the weight and skin color of an elderly patient avatar: white-thin, black-thin, white-obese and black-obese. Medical students viewed white obese avatars as unattractive, ugly, noncompliant, lazy, and sloppy. Medical students' comments suggested a paternalistic attitude toward avatar patients. Avatar-mediated experiences can elicit medical students' bias potentially enabling medical educators to implement bias reduction interventions.

Using anthropomorphic avatars resembling sedentary older individuals as models to enhance self-efficacy and adherence to physical activity: psychophysiological correlates

The prevalence of obesity and associated health complications are currently at unprecedented levels. Physical activity in this population can improve patient outcomes. Virtual reality (VR) self-modeling may improve self-efficacy and adherence to physical activity. We conducted a comparative study of 30 participants randomized to 3 versions of a 3D avatar-based VR intervention about exercise: virtual representation of the self exercising condition; virtual representation of other person exercising and control condition. Participants in the virtual representation of the self group significantly increased their levels of physical activity. The improvement in physical activity for participants in the visual representation of other person exercising was marginal. The improvement for the control group was not significant. However, the effect sizes for comparing the pre and post intervention physical activity levels were quite large for all three groups. We did not find any group difference in the improvements of physical activity levels and self-efficacy among sedentary, overweight or obese individuals.

Simulating geriatric home safety assessments in a three-dimensional virtual world

Virtual worlds could offer inexpensive and safe three-dimensional environments in which medical trainees can learn to identify home safety hazards. Our aim was to evaluate the feasibility, usability, and acceptability of virtual worlds for geriatric home safety assessments and to correlate performance efficiency in hazard identification with spatial ability, self-efficacy, cognitive load, and presence. In this study, 30 medical trainees found the home safety simulation easy to use, and their self-efficacy was improved. Men performed better than women in hazard identification. Presence and spatial ability were correlated significantly with performance. Educators should consider spatial ability and gender differences when implementing virtual world training for geriatric home safety assessments.

Avatar-mediated home safety assessments: piloting a virtual objective structured clinical examination station

BACKGROUND

Avatars and virtual worlds offer medical educators new approaches to assess learners' competency in home-safety assessments that are less time-consuming and more flexible than traditional home visits. We sought to evaluate the feasibility and acceptability of implementing an avatar-mediated, 3-dimensional (3-D) home simulation as a virtual objective structured clinical examination station for geriatric medicine fellows.

METHODS

We developed a 3-D home simulation in the virtual world Second Life (Linden Lab, San Francisco, CA) containing 50 safety hazards that could affect the safety of an elderly person at home. Eight geriatric medicine fellows participated in a 16-station objective structured clinical examination, with one station assigned to the 15-minute 3-D virtual world simulation, where the fellow's "home visit" was performed by navigating his or her avatar in the virtual world simulation. The fellows were instructed to find the home safety hazards in the simulated environment and then provide specific written recommendations. Two reviewers independently scored the fellows' written findings against an inventory-based checklist.

RESULTS

The geriatric medicine fellows scored a mean of 43% ± SD 9 on the inventory-based checklist. The scoring of the 2 reviewers showed a high interrater reliability (88%). Six of the 8 participants (75%) rated the simulation as "excellent." Four of the 5 women (80%) and none of the 3 men (20%) participating in the virtual objective structured clinical examination needed navigation assistance in the 3-D virtual house.

CONCLUSION

The 3-D, avatar-based, virtual geriatric home safety objective structured clinical examination is a practical and acceptable alternative to the traditional home safety visits in an objective structured clinical examination setting.

Avatar-mediated training in the delivery of bad news in a virtual world

BACKGROUND

Delivering bad news to patients is an essential communication skill for physicians. Educators commonly use standardized patient (SP) encounters to train physicians in the delivery of bad news. It is expensive to use actors, for logistical reasons such as travel and scheduling, and there are limits to the characters and conditions an actor can portray in teaching encounters. Thus we studied the feasibility of creating SP avatars in a virtual world for the task of training medical trainees to deliver bad news. The SP avatars are easily customized for different scenarios and amenable to distance learning.

METHODS

We recruited 10 medical trainees to interact with a standardized female avatar in a three-dimensional simulated clinic, where the trainee was to inform the avatar of her newly diagnosed breast cancer. The trainee evaluated his or her self-efficacy in delivering bad news via an affective competency score (ACS) before and after the encounter. Two palliative care specialists evaluated each trainee's performance using the Bad News Assessment Schedule and the performance ACS.

RESULTS

The self-efficacy ACS scores of the trainees improved overall: before, 20 ± 4, versus after, 24 ± 3, p = 0.001 (maximum score = 30). All participants considered the experience positive and commended the novel approach, although noting that the avatars were not able to portray body language cues.

CONCLUSION

Participants viewed the avatar-mediated training as an excellent approach for learning how to deliver bad news but believed it could not substitute for real patient interactions. However, participant self-efficacy improved, which suggests that avatar-mediated training in a virtual world is a viable educational approach for skill training in delivering bad news.

Preparation and characterization of activated carbon from rice bran

A study on the preparation of rice bran-based activated carbon was conducted, with and without an acid treatment step prior the activation process. The influence of the activation time on the structure of the activated carbons was evaluated. The acid treatment had a significant positive influence on sorption properties. The rice bran-activated carbon presented a BET surface area of 652m(2)g(-1) and a pore volume of 0.137cm(3)g(-1), with mesopores predominance (ca. 55%). These experimental results indicated the potential use of rice bran as a precursor in the activated carbon preparation process, thus representing an economically promising material.

Omega-3 fatty acids in baked freshwater fish from south of Brazil

Lipid and fatty acid levels in the edible flesh of 17 baked freshwater fish from Brazil's southern region were determined. Analyses of fatty acids methyl esters were performed by gas chromatography. Palmitic acid (C16:0) was the predominant saturated fatty acid, accouting for 50-70% of total saturated acids. Linoleic acid (C18:2 omega 6), linolenic acid (C18:3 omega 3), and docosahexaenoic acid (C22:6 omega 3) were the predominant polyunsatured fatty acids (PUFA). The data revealed that species such as barbado, corvina, pintado, and truta were good sources of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), and that most freshwater fish examined were good sources of PUFA-omega 3.