Little is known about the relationship of education level to utilities for health states of functional dependency such as the Activities of Daily Living (ADLs) among older adults.

Using the standard gamble (SG) method, 371 medical outpatients (mean age = 74, sd=6.1, range 65-92; 65% female, 35% non-white) gave utility ratings for hypothetical health states of dependency in each of 6 ADLs, a combination of all 6 ADLs, Walking, a combination of 3 ADLs while living in a Nursing Home, and the same combination in their Own Home. Basic understanding of the SG method was measured according to frequency of inconsistent responses in which individuals give a lower utility rating for one health state than for the combination of all 6 and frequency of invariant responses in which individuals give the same rating for each of the 6 ADL dependencies (Med Decis Making. 2005 Mar-Apr;25(2):158-67). Participants also reported their highest level of education.

20(5.4%) participants completed 11th grade or less, 43(11.6%) completed high school, 115(31%) completed some college, 77(20.8%) received a four year degree, and 116(31.3%) received a graduate degree. One-way ANOVAs revealed no difference between varying levels of education for comprehension of the SG task based on frequency of inconsistent responses. For invariant responses, high school graduates(35%) were more invariant than 4 year college graduates(21%), p=.024. There were no significant differences for invariance among other levels of education.

Mean utilities ranged from .55(Eating) to .67(Dressing) for 11th grade or less, .61(Walking) to .73(Bathing) for high school, .63(Eating) to .74(Dressing) for some college, .72(Toileting) to .81(Dressing) for 4 year degree, and .82(Bathing) to .72(Continence) for graduate degree. There was a significant positive correlation between education level and utilities for bathing, r(371)=.152, p&lt.01, dressing, r(371)=.133, p&lt.05, eating, r(371)=.149, p&lt.01, toileting, r(371)=.174, p&lt.01, transferring, r(371)=.179, p&lt.01, and walking, r(368)=.170, p&lt.01. Education level was also positively correlated with utilities for Nursing Home, r(368)=.141, p&lt.01 and Own Home, r(368)=.118, p&lt.05.

Standard Gamble utilities of ADLs increased as education level increased suggesting that individuals with less educational background are willing to take greater risks to avoid health states of dependency. This relationship was not explained by any apparent variation in understanding of the task. Policy analyses using utilities should include ratings from a broad sample across the educational spectrum.