Adaptation and Evaluation of the Digital Elder Abuse Intervention “VOICES” in the Primary Care-Based Setting

Abstract

Purpose Elder abuse is a growing international problem where prevalence ranges globally from 2.2% to 36.2%¹. Traditional methods for screening and identification are limited, and healthcare professionals face a variety of barriers to reporting such as time constraints, misinformation regarding adult protective service (APS) reporting and lack of elder abuse education². We aimed to address these limitations by developing a self-administered, tablet-based digital health intervention to assist older adults in identifying with and reporting elder abuse³. The VOICES elder abuse intervention (EAI) was created to be self-administered by the patients and offers a customized experience guided by answers provided by the patient. The VOICES EAI is presented on a tablet device and includes multimedia elements such as automated text-to-speech and animated videos. In a previous study, we evaluated VOICES in a real-world emergency department (ED) setting (N=1002) where we established proof of acceptability, feasibility, and preliminary efficacy. Method In this study we adapted and refined the VOICES EAI to be used by older adults in primary care settings. To evaluate the adapted EAI, we recruited 50 participants in a feasibility study from the Yale Internal Medicine Associates (YIMA) primary care clinic in New Haven, CT, USA. We recruited English-speaking, community-dwelling, cognitively intact individuals aged 60 years and older who visited the clinic. Recruitment sought diversity in age, race, and sex to reflect the patient population receiving care at the study site. The study session was conducted entirely on an iPad tablet and took 45 minutes on average to complete. Once the VOICES EAI was completed, participants answered a series of validated questions related to feasibility, satisfaction, ease of use and appropriateness of the tool. Study data were derived from the following sources: participant demographics, participant responses from VOICES (abuse screener, self-identification, self-reporting), participant responses from post-survey, and researchers’ observations. The Yale University Human Investigation Committee (IRB) approved all study protocols and participants were compensated for their time. Results and Discussion Among the participants who completed the study (N=50, 73y+/-7.3, 62%F, 83% white), 72% were not employed and 98% received a high school education or higher.  Eighty-three percent of participants used the internet daily and 98% had easy access to or owned a computer, tablet, and/or smartphone. On a 5-point Likert scale, 96% of participants were satisfied with the level of confidentiality and privacy VOICES provided, as well as their ability to complete the tool on their own without assistance. Ninety-two percent of participants were satisfied with the time it took to complete using VOICES (Figure 1). Our findings demonstrate that older adults in primary care settings found the VOICES EAI to be feasible and acceptable, and do not significantly differ from our emergency department findings. We believe that VOICES is best adapted into existing primary care protocols for identifying elder abuse, rather than replacing them. Additional studies exploring VOICES with cognitively impaired individuals are underway. Future work involving this tool will include Spanish-speaking individuals, as well as exploring the use of artificial intelligence for expanding psychoeducational modules.