Measuring desire for control of health care processes.

We conducted three known-groups studies to obtain discriminant validity information among potential self-report measures of the construct desire for control over health care processes. In the first study we looked at types of preparation for childbirth; in Studies 2 and 3 we investigated choice of a place in which to die and signing a Living Will. With none of the measures could we adequately distinguish those who had signed a Living Will (or intended to) from those who did not intend to sign one. The best discriminator of a choice of a place to die (hospital vs. home or hospice) and type of preparation for childbirth (Lamaze vs. other types of classes vs. no classes) was the Information subscale of Krantz's Health Opinion Survey (KHOS; Krantz, Baum, & Wideman, 1980). With the Behavioral Involvement subscale of the KHOS and our newly designed situation-specific measure of desire for control, we could only inconsistently discriminate among the groups. The generalized measure, Burger and Cooper's (1979) Desire for Control Scale, did not help us to discriminate among these groups. Correlations of these measures with demographic variables are provided and discussed.