Fried TR, Tinetti ME, Iannone L, O’Leary JR, Towle V, Van Ness PH. Health Outcome Prioritization as a Tool for Decision Making Among Older Persons With Multiple Chronic Conditions. Arch Intern Med. 2011;171(20):1856-1858. doi:10.1001/archinternmed.2011.424
Author Affiliations: Department of Medicine (Dr Fried), School of Epidemiology and Public Health (Dr Tinetti), and Program on Aging (Mss Iannone and Towle, Mr O’Leary, and Dr Van Ness), Yale University School of Medicine, New Haven, Connecticut; and Clinical Epidemiology Research Center, VA Connecticut Healthcare System, West Haven (Dr Fried).
Older persons with multiple chronic conditions are at substantial risk for unintended adverse outcomes, such as medication adverse events. Less-severe adverse events are commonly referred to as “side effects,” implying that they are secondary to disease-specific benefits. However, patients consider these adverse events to be important outcomes in their own right.1 Such findings suggest that all possible benefits and harms resulting from different treatment options be considered as competing outcomes, among which older persons with multiple chronic conditions face trade-offs.
When treatments involve trade-offs, the best option depends on patients' preferences. The challenge for older persons with multiple conditions is that these trade-offs encompass both many different specific diseases and nondisease-specific health domains.2 One approach to this challenge is to consider treatment in terms of its effects on a set of universal, cross-disease outcomes and to use older persons' prioritization of these outcomes as an assessment of preferences. These outcomes, examples of which include length of life, physical and cognitive function, and symptoms, include basic domains recognized to be the key components of health.3 The goal of this study was to explore the use of a simple tool to elicit older persons' health outcome priorities.
Participants were recruited from 3 senior centers and 1 independent/assisted-living facility in the New Haven area in Connecticut. All volunteers were included in the study without exclusion except for residents in the assisted-living facility with a diagnosis of dementia. The protocol was approved by the Yale University School of Medicine Human Investigations Committee.
All data were collected by self-report in face-to-face interviews. The main outcome variable was participants' prioritization of 4 universal health outcomes: keeping you alive, maintaining independence, reducing or eliminating pain, and reducing or eliminating other symptoms (eg, dizziness, fatigue, shortness of breath). These priorities were elicited using a computerized tool, in which each of these outcomes was presented in a moveable box on a screen. Participants were asked to provide a rank order of priorities. Participants were also asked to provide a priority score by placing the boxes on a visual analog scale from 0 to 100, with higher scores indicating that the outcome was more important. The tool was accompanied by a script explaining the concept of competing outcomes (eAppendix 1). Initial development and testing of the tool has been previously described.4 In response to poor test-retest reliability, the script was modified prior to the start of the present study to emphasize the notion of trade-offs.
The methods and results for examining the test-retest reliability of the tool are available in eAppendix 2 and the eTable.
We examined distributions of health outcome priorities in 2 ways. We first examined the proportion of participants choosing different priority rankings. We next used the spread in scores assigned to each outcome to evaluate the strength of importance ascribed to each outcome. We divided the participants into 4 groups according to which of the outcomes they selected as most important. Within each of these groups, we calculated the mean score for each outcome.
Of the 357 participants, 75% were women and were white, 69% had 4 or more chronic conditions, and 49% were prescribed 4 or more medications. Maintaining independence was the health outcome ranked as most important by most participants (76%), with varying proportions of participants ranking each of the other outcomes as most important (Table). Within each outcome ranked as most important, there was variability in the rankings of the other outcomes, although patterns did emerge. Among those ranking maintaining independence as most important, most participants (67%) ranked pain and/or symptom relief as second and/or third and staying alive as least important. Among those ranking staying alive as most important, most (66%) ranked maintaining independence as second and pain and/or symptom relief as third and/or fourth. Among those ranking pain or symptom relief as most important, there was more variability in the rankings of the other outcomes.
The mean scores for the outcome ranked as most important were 97 to 98, with narrow standard deviations. The standard deviations around the mean scores for other health outcomes were large. For participants who ranked staying alive as most important, the other outcomes also received high mean scores (maintaining independence, 80; pain relief, 70; and symptom relief, 65). For all participants who ranked an outcome other than maintaining independence as most important, maintaining independence received mean scores close to the most highly scored outcome (80-84). In contrast, for participants who ranked an outcome other than staying alive as most important, staying alive received mean scores substantially lower than the most highly scored outcome (52-59).
The variability in universal health outcome rankings is similar to the variability seen in patients' valuations of disease-specific health outcomes.5,6 Nonetheless, the importance of maintaining independence for many participants has also been demonstrated in studies evaluating older persons' preferences in end-of-life decision making.7- 9 The similarities in findings across studies examining different clinical situations and using different methods support the key role that function should play in treatment decision making for older persons.
These findings illustrate a potential role for health outcome prioritization at both the individual and population level. At the population level, the effects of treatment interventions are currently generally assessed in terms of disease-specific outcomes or reduction in mortality. Moreover, nonmortality outcomes are frequently measured by a composite quality of life index, which averages the effects of the intervention across the different domains of functioning and symptoms. The variable importance of each of these individual domains to older persons argues for efforts to examine the effect of treatment on each domain. At the individual level, the study results suggest that the prioritization of outcomes can be used to begin a discussion about what is most important to older persons with multiple conditions. The variability in individual priorities will require the development of a range of treatment options, designed to increase the likelihood of wanted outcomes and decrease the likelihood of unwanted outcomes.
Correspondence: Dr Fried, VA Connecticut Healthcare System, 950 Campbell Avenue, Mail Stop CERC 151B, West Haven, CT 06516 (email@example.com).
Published Online: September 26, 2011. doi:10.1001/archinternmed.2011.424
Author Contributions: Dr Fried had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Fried and Tinetti. Acquisition of data: Fried, Iannone, O’Leary, and Towle. Analysis and interpretation of data: Fried, Tinetti, Iannone, O’Leary, Towle, and Van Ness. Drafting of the manuscript: Fried. Critical revision of the manuscript for important intellectual content: Tinetti, Iannone, O’Leary, Towle, and Van Ness. Statistical analysis: Van Ness. Obtained funding: Fried and Tinetti.
Financial Disclosure: None reported.
Funding/Support: This study was supported by grant Pioneer 58381 from the Robert Wood Johnson Foundation and by the Claude D. Pepper Older Americans Independence Center at Yale University School of Medicine (#P30AG21342 from the National Institutes of Health/National Institute on Aging [NIH/NIA]). Dr Fried is supported by grant K24 AG028443 from the NIH/NIA.
Role of the Sponsors: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
Additional Contributions: Andrea Benjamin, RN, provided skillful interviews of the participants.