Abstract
OBJECTIVES: To assess whether patient perceptions of treatments for diabetes mellitus differ according to clinical criteria such as limited life expectancy and functional decline (i.e., vulnerability). DESIGN: Cross-sectional survey. SETTING: Clinics affiliated with two Chicago-area hospitals. PARTICIPANTS: Patients aged 65 and older living with type 2 diabetes mellitus (N=332). MEASUREMENTS: Utilities (quantitative measures of preference on a scale from 0 to 1, with 0 representing a state equivalent to death and 1 representing life in perfect health) were assessed for nine hypothetical treatment states using time trade-off questions, and patients were queried about specific concerns regarding medications. Vulnerability was defined according to the Vulnerable Elders Scale. RESULTS: Thirty-six percent of patients were vulnerable. Vulnerable patients were older (77 vs 73) and had diabetes mellitus longer (13 vs 10 years; P<.05). Vulnerable patients reported lower utilities than nonvulnerable patients for most individual treatment states (e.g., intensive glucose control, mean 0.61 vs 0.72, P<.01), but within group variation was large for both groups (e.g., standard deviations >0.25). Although mean individual state utilities differed across groups, no significant differences were found in how vulnerable and nonvulnerable patients compared intensive and conventional treatment states (e.g., intensive vs conventional glucose control). In multivariable analyses, the association between vulnerability and individual treatment state utilities became nonsignificant except for the cholesterol pill. CONCLUSION: Older patients' preferences for intensity of treatment for diabetes mellitus vary widely and are not closely associated with vulnerability. This observation underscores the importance of involving older patients in decisions about treatment for diabetes mellitus, irrespective of clinical status.