A counseling intervention to improve personal control, affective outcomes, and satisfaction in surrogate decision makers at end-of-life

Abstract

This study investigated whether personal control, situational anxiety, situational depression, and satisfaction with end-of-life (EOL) care might be improved in surrogate decision makers who receive a cognitive-behavioral decision making intervention. The sample consisted of 38 surrogate decision makers for 38 incapacitated patients in three hospitals in a moderate sized city in the. Southeastern United States. The demographic characteristics of sample included: 50% female and 50% male, 50% African-American and 50% White, and 81.6% Protestant. Although the sample was small, the demographic composition of the sample was highly consistent with the population of the community in which the study was conducted. Guided by Bandura's Social Cognitive Theory, it was hypothesized that experimental (DMCI) group subjects would score higher on posttest measures of personal control and satisfaction with EOL care than subjects in the usual care group. It was also hypothesized that experimental group subjects would score lower on posttest measures of depression and anxiety. Pretest and posttest scores on five instruments were compared and analyzed for an experimental group, N=19 which received the DMCI versus a control group, N= 19, that received only usual care. Instruments used to assess the dependent variables of personal control, affective outcomes, 3!1d satisfaction with end-of-life care were: the Pearlin Mastery Scale, the Perceived Personal Control Questionnaire, the Beck Depression Inventory, the Beck Anxiety Inventory, and the FAMCARE Scale. Data were collected over a seven month period from Summer 2004 to Spring 2005. After obtaining informed consent experimental participants completed all instruments (pretest), received three counseling sessions over approximately 14 days and then completed all instruments again (posttest). Control group members after providing informed consent also completed all instruments (pretest), received three attention- usual care contacts, and subsequently completed all instruments again (posttest). Data were analyzed using descriptive statistics, Chi-Square, Mann-Whitney Test, and 2-way ANOV A with one repeated measure. The factors were group assignment (experimental vs. control) and time (pretest vs. posttest) with time as the repeated factor. Chi-square, and Mann-Whitney tests showed the groups were comparable at pretest and that demographically the random assignment to groups was effective. The 2-way ANOV A results showed significant interaction effects in each case. The experimental group scored higher or lower posttest as predicted by the hypotheses compared to the control group.