In this qualitative study of older persons with multiple conditions, their experiences with adverse medication effects provided some with readily understood examples of the concepts of trade-offs and the need to prioritize among potentially competing outcomes. When asked about medications with potentially competing outcomes, participants shifted from thinking about outcomes in disease-specific terms, such as achieving a target blood pressure or lipid level, and instead considered more global, cross-condition health outcomes, such as survival, symptoms, and function. In contrast to their perceptions of medications, participants had little awareness that the presence of multiple conditions themselves placed them at risk for adverse outcomes. In addition, they frequently had overly optimistic beliefs regarding the benefits of the therapies they were receiving and generally conceptualized outcomes in absolute terms.
The participants' lack of awareness regarding the potential for competing outcomes arising from the presence of multiple conditions stands in contrast to an earlier qualitative study which asked persons with 2 or more chronic conditions to list barriers to self-management of their conditions.20
In that study, whose population was younger, in poorer health, and of lower socioeconomic status as compared to the current study population, participants frequently cited aggravation of one condition by the symptoms or treatment of another as a barrier. It is difficult to know whether the difference in findings of these two studies is a result of differences in the populations and/or the questions posed to the participants. Nonetheless, because participants in the current study most readily understood the concept of competing outcomes in terms of medication adverse effects, we asked them to describe how they had made decisions regarding their own medications and how they would make decisions regarding medications with competing outcomes. The way in which participants approached these decisions provides some fundamental insights into how older persons might be engaged in the process of thinking about what is most important to them when faced with potentially competing outcomes.
Central to the process of prioritizing their concerns was a shift in how participants thought about the outcomes of treatment. Their initial conception of these outcomes was in disease-specific terms, such as achieving a given blood pressure or lipid level. However, when faced with the need to prioritize among a number of different outcomes, many participants discussed these outcomes in more general, cross-disease terms. For example, when asked about taking a medication to treat hypertension that caused dizziness, participants spoke of weighing the adverse effect on quality of life caused by dizziness against the possibility of dying from a stroke or heart attack. Yet other participants, when considering whether they would take a medication that would increase their risk for heart attack in order to treat pain, defined thresholds of decreased function at which they would become willing to assume a greater risk of cardiovascular mortality. When taken to its logical extreme, this process allows all outcomes to be reduced to a single common denominator, as illustrated by the participant who would consider the effect of all of her diseases and their treatments on her function and then choose the therapeutic option that would prevent her from becoming incapacitated.
These results suggest that one approach to eliciting older persons' values when faced with competing outcomes would be to ask patients to prioritize among a small number of global outcomes by selecting the outcomes that are most important to them to achieve or avoid. Physicians could then determine the course of care that would be most likely to meet these priorities. Asking participants to prioritize among a set of outcomes potentially offers several advantages compared to the alternative and more common strategy of asking patients to choose among a given set of treatment options. First, the results of this study suggest that the process of prioritizing outcomes may be easier for patients to do than to understand all of the complexities inherent in the specific risks and benefits of alternative treatment options. Second, such an approach may have greater acceptability to the sizeable proportion of patients who do not want to participate directly in the process of making medical decisions.12
Third, patients' priorities regarding these global treatment outcomes may be applicable to a wide range of specific treatment decisions,21
so that they would not have to engage in a separate decision-making process for each of many decisions regarding competing outcomes that they may face. Finally, it has been argued that decision-making based on identifying desired outcomes and selecting the path that can best achieve those outcomes is superior to decision-making based on choosing among a set of predetermined alternatives.22
The results of this study also highlight additional challenges to the task of engaging patients with multiple conditions in the process of medical decision-making. First, many patients and physicians may not recognize that they cannot achieve desired outcomes without risking undesired outcomes. This lack of recognition is reflected in the belief expressed by participants that adverse effects of one medication could be treated by additional medications. In addition, this approach would require a shift in physician decision-making from its current focus, fostered by the existing specialty orientation of medical care, on individual diseases and disease-specific outcomes to a consideration of global, cross-disease outcomes. Despite the ever increasing recognition of competing risks, they are being measured as disease –specific outcomes.23, 24
The patient who is wishes to maximize her likelihood of independence but suffers from both diabetes and falls can participate in the decision whether to take a thiazolidinedione, for example, with its increased risk of hip fracture,23
only if the effects of both diabetes under different treatment strategies and hip fracture on function are known. Second, it will need to be determined how sensitive the outcome priorities of older persons are to transient, and particularly affective, factors such as anxiety or pain, in order to understand the extent to which they do or do not reflect core, long-term values.25, 26
Third, for patients who can understand the concept, the prioritization of outcomes would need to be done in the context of acknowledging and assessing patients' attitudes toward uncertainty. The finding in this study of a belief in the absolute benefits of certain interventions suggests the importance of an accurate understanding of outcomes for patient-centered decision-making. Previous studies have demonstrated that many people want greater benefit than can be achieved by preventive medications in order to choose to take the medication.27, 28
In addition, it has been shown that patients are more likely to prefer a medication when they believe it eliminates the risk of a given outcome rather than merely reducing the risk, even if the absolute risk reduction is equivalent.29
As a qualitative study, the results of this analysis can be used only to understand how older persons with co-existing conditions think about their illnesses and interventions and not to draw conclusions about the frequency of their knowledge or perceptions. By enrolling older persons from different recruitment sites, we sought to elicit the opinions of persons from a range of ethnic/racial and sociodemographic backgrounds. However, because we conducted the focus groups only in English we may have not included the full spectrum of perceptions.
Health care decision-making for older persons with multiple conditions is complicated by the potential for competing outcomes, and involving patients in the decision-making process is challenging. Asking older persons to prioritize among a set of global outcomes that can be applied across a spectrum of specific diseases may be one easily understood approach to eliciting values in a manner that can inform a range of health care decisions.