Date of Award

January 2014

Document Type

Thesis

Degree Name

Medical Doctor (MD)

Department

Medicine

First Advisor

Terri R. Fried

Subject Area(s)

Gerontology, Medicine

Abstract

This study's purpose is to explore the use of universal health outcome prioritization tools to facilitate the incorporation of older persons' values into medical decision-making. The study aims are therefore to investigate: 1) the psychometric properties of a new health outcome prioritization tool; 2) older adults' attitudes about using such tools, and 3) older adult's perspectives on tradeoffs in their care and discussing priorities with providers.

An observational cohort of 357 community-dwelling older adults age 65 and older used three new health outcome prioritization tools that focus on tradeoffs between quality and quantity of life, and present and future health. We did a detailed analysis of one tool, the Time and Outcome Preference (TOP) scale, including principal components analysis, construct validity, internal consistency, and test-retest reliability. For all three tools, we coded responses about difficulty of use and potential for changing care, and did bivariate and multivariate analyses of relationships between participant characteristics and preferred tool or tool difficulty. In addition, we interviewed 50 patients over age 65 from Veterans Administration clinics about tradeoffs in their care, the role of health outcome prioritization tools, and barriers to discussing priorities, and analyzed responses with descriptive statistics.

The TOP scale analysis resulted in two subscales, each addressing one tradeoff. Subscale score medians fell near the middle, but a substantial number of participants scored at the extremes. Construct validity and internal consistency were good and test-retest reliability was fair. Approximately 40% of participants preferred the TOP scale, and 40% preferred the Health Outcome Prioritization tool. Using tools was not difficult for the majority, and difficulty was not greater among participants with lower health literacy, education, or health status. Few participants believed tools would change care and the most commonly cited reason was satisfaction with existing care. In interviews about tradeoffs, many gave an example, but some did not offer two true alternatives and most described higher risk situations. Only 36% recalled having specific conversations about their priorities and 90% thought that tools could be helpful in this process. The main barriers to discussion were time and uncertainty about priorities or how priorities would change care.

Overall, this study supports the idea of offering the new universal health outcome prioritization tools as a valid way for older adults to express their priorities when faced with a tradeoff in their clinical care. While the TOP scale's test-retest reliability was modest, its high validity suggests that it can be used to familiarize people with common tradeoffs as part of an ongoing conversation. The variation in preferred tools and lack of difficulty for subpopulations could mean that providing a choice between tools may be helpful for a broad range of patients. Finally, while most participants understood the concept of tradeoffs in medical decisions, some may not be familiar with how tradeoffs apply to decisions about everyday management. The lack of specific conversations about priorities and multiple barriers to these conversations emphasize the need for further support around preference-sensitive decisions.

Comments

This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.

Share

COinS