Background: Unrelieved pain is highly prevalent and devastating for Veterans in VA’s 134 Community Living Centers (CLCs). Imminent removal of pain as one of the CLC quality measures offers an opportunity, per VA’s Office of Geriatrics and Extended Care (GEC), to develop new, risk-adjusted measures that more accurately characterize CLC pain management. These measures can identify CLCs successful at pain management while minimizing biased underestimates for CLCs with the sickest residents. Then, by diving deeply into structures and processes of high performers, we can learn how to intervene. My background in gerontology, quantitative methods, and implementation science partially prepares me for this work. But I need additional training in risk adjustment, qualitative research, cutting-edge analytic methods, and intervention study designs for the study and my health services research career to succeed. Specific Aims: The proposed CDA simultaneously fills the considerable gaps in my background and provides VA with rigorous, actionable research on which to ground future quality improvement efforts. A social- ecological model frames the work. GEC commits to serving as an invested partner. I have 3 aims, which I will achieve with my mentors and training. 1. Evaluate how risk adjustment changes judgements of CLC pain management performance. 2. Use mixed methods to perform in-depth studies of CLCs with high outlying performance. 3. Adapt an existing, evidence-based intervention comprising lessons learned from “positive deviants.” Methods: Aim 1: Using VA administrative data of CLC residents, I will (1) calculate unadjusted pain measures, (2) apply risk adjustment, (3) assess the measures’ reliability and validity, and (4) identify high and low outlying performance on pain management. Aim 2: I will use quantitative (survey) and qualitative data from staff and residents at 5 top-performing CLCs, contrasted with qualitative data from 5 low-performing CLCs, to develop hypotheses of contextual factors and pain management practices unique to positive deviants. I will test causal relationships using configurational comparative analytic methods. Aim 3: I will adapt an existing nursing home pain management intervention for use in VA CLCs, using empirical evidence from Aim 2 about necessary conditions for optimal pain management. A modified e-Delphi panel of CLC stakeholders and pain management experts will provide feedback on the intervention package’s design. I will use a developmental formative evaluation of qualitative data from staff at 1 low-performing CLC to assess the intervention’s feasibility and acceptability, in preparation for rigorous testing in future work. Expected Results and Next Steps: I will provide GEC with interim deliverables to enable assessment of CLC pain management quality, guide CLC policy, and support clinical practice in CLCs struggling with pain management. Knowledge from this CDA will lead me to develop studies to refine risk adjustment methods for quality measurement in other critical areas and to rigorously evaluate, using a hybrid type II design, clinical effectiveness and implementation of the intervention. Significance & Relevance to Veterans’ Health: Coming at a critical juncture in my VA research career, this timely study responds to the VA priority of Greater Choice for Veterans, ORD’s priority to increase substantial real-world impact of VA research, and HSR&D’s Long-term Care and Opioid/Pain priority domains. Although pain is highly prevalent and debilitating for the 42,000 vulnerable Veterans CLCs serve, almost nothing is systematically known about CLCs’ pain management quality. And current pain measures are about to disappear. This study seizes this opportunity, developing nuanced, VA-specific approaches that are custom- made to reflect the accurate state of CLC pain management and help improve VA long-term care.
External Links for this Project
Grant Number: IK2HX003184-01A1
None at this time.
Health Systems, Aging, Older Veterans' Health and Care
Prevention, Technology Development and Assessment
None at this time.