Date of Award


Document Type


Degree Name

Medical Doctor (MD)

First Advisor

Mary E. Tinetti MD


Context: Although rarely discussed explicitly, there is an implicit tradeoff between cardiovascular versus fall-related and medication symptom risks in antihypertensive prescribing. Few studies have assessed priorities of elderly persons at risk for these specific outcomes and the reasons for these priorities. Objective: The primary objectives of this study were to determine how elderly persons at risk of falling and with diagnosed hypertension ranked avoidance of cardiovascular, fall, and medication-related symptoms and to ascertain participants reasons for their preferences. The study also looked at possible relationships between participant baseline characteristics and rankings and compared the ranking results with the conjoint analysis results obtained in the larger study. Secondary objectives were to evaluate the relationship between ranking and self-reported adherence, to ascertain the relationship between ranking and considering stopping medications, and to determine the relationship between ranking and perception of medication symptoms. Design and Setting: Single interview encounter with participants residing in age-aggregated housing (including public and private senior housing, life care communities, and retirement facilities). Participants: 103 persons 70 years or older at risk of falling and with diagnosed hypertension. Measurements: Demographic and health data; Card order ranking task; open-ended responses to questions about reasons for ranking; Morisky and visual analog scale measures of adherence, and self-reported adverse medication effects; conjoint analysis data from larger study. Results: Of the 103 participants, 83 (81%) ranked avoiding cardiovascular outcomes first; 20 (19%) prioritized avoiding fall injuries and medication symptoms. Concern about disability and personal experience were among the top reasons for both groups. Concern about head injuries was cited only by those who prioritized falls ( p < 0.01) while death was mentioned only by those prioritizing cardiovascular outcomes (p = 0.02). Those participants taking fewer anti-hypertensive medications were statistically more likely to rank avoiding falling or medication symptoms first (p = 0.02). There were trends suggestive of an increased preference for avoiding fall injury or medication-related symptoms in those who were older (p = 0.13), white (p = 0.17), and taking fewer medications (p = 0.19). There was fair agreement between the results of the ranking and conjoint analysis (kappa = 0.355; 95% confidence interval from 0.174 to 0.537). Analysis did not reveal a relationship between participant ranking and adherence, considering stopping medications, or reported adverse effects. Conclusions: While the majority of participants ranked avoidance of adverse cardiovascular events first there was inter-individual variability in ranked priorities. Participants were able to articulate reasons for these preferences. This supports treating in accordance with preference and may mean compromising on blood pressure reduction for some individuals. Reasons given for ranking reflect broad concerns about disability that should be considered by clinicians. The differences between ranking and conjoint analysis suggest that the way participants are asked about their priorities matters. Participants place less priority on avoidance of falling or medication symptoms in a ranking task than in a conjoint analysis task. Development of a reliable and clinically simple method of preference elicitation that incorporates elements of tradeoffs will be essential so that they can be acknowledged and priorities and reasoning elicited as part of clinical decision-making.


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