Developing and Evaluating the Efficacy and Feasibility of Delivering a Psychoeducational Individual Intervention to Women with Advanced Stage Ovarian Cancer: A Two Phase Study

PURPOSE: The purpose of this study was to assess the psychosocial needs of women with ovarian cancer treated at Baylor University Medical Center Sammons Cancer Center, and to design and evaluate the feasibility and efficacy of an individual psychoeducational intervention on reported symptoms of depression, anxiety, and psychosocial adjustment to illness. Secondary aims included identifying the barriers to accessing group support services as well as identifying a mechanism to address those barriers. METHOD: This study was conducted in two phases. Phase I consisted of a psychosocial needs assessment (n = 15) to determine the appropriate elements necessary for inclusion in an individual intervention. Perceived barriers to participation in psychosocial services, including group support, were also assessed. The outcome of the needs assessment combined with an extensive literature review was used to develop the Phase II intervention. In Phase II, participants (n = 30) were recruited from the gynecologic oncology practice at Baylor Sammons Cancer Center. Participants were randomized to intervention (n = 15) vs. control (n=15) groups. Intervention sessions lasted 60 minutes and were conducted weekly over a period of 8 weeks. Sessions consisted of a combination of education and psychosocial support. Growth modeling was used to evaluate variation in within-individual rate of change in outcome measures during the treatment (weeks 1-8) and followup (weeks 16-48) periods of the study. Repeated measures ANOVA was used to evaluate average group differences in outcome measures during the treatment and follow-up periods of the study. RESULTS: No main effects of time, group, age, or education were observed with regard to rate of change of psychosocial adjustment to illness during the treatment period of the study using growth modeling. However, during the follow-up period of the study, time and group were significant predictors of within-individual variation in rate of change in psychosocial adjustment to illness. Similarly, repeated measures ANOVA yielded no main effects of time or group in psychosocial adjustment to illness during the treatment period of the study; however, a main effect of group (p<.05) and an interaction between group and time (p<.05) were noted during the follow-up period of the study with the most significant differences in group means occurring at week 48 (effect size = .70). Although not stable across all models created, time was the only significant predictor of within-individual variation in rate of change in depression symptoms during the treatment and follow-up periods. A main effect of time without consideration of group (p<.01) and an interaction between group and time (p<.01) with regard to depression symptoms was noted in the treatment phase of the study, most significantly at week 8. A main effect of group (p<.05), most significantly at week 48, was observed during the follow-up phase (effect size = .69). Similar to depression scores, withinindividual variation in anxiety symptom scores was significantly predicted by time during the treatment period of the study. Although no main effects of predictors were observed with regard to anxiety symptoms during the follow-up period of the study, the model with the greatest goodness of fit included the predictors education, group, and time. Repeated measures ANOVA suggest a main effect of time (p<.05) with regard to anxiety symptoms during the treatment period of the study but no main effect of group. However, during the follow-up period of the study, a main effect of time (p<.05) and group (p<.05) were noted with regard to differences in the average state anxiety symptom score between groups, with the most significant difference occurring at week 48 (effect size = .43). Cost analyses suggest minimal differences in the number of medical office visits and phone calls between groups and a moderate difference in usage of psychiatric and pain medication between groups. Treatment satisfaction was high across all intervention topics, with participants rating sessions 1 and 8 the most helpful. RESEARCH AND CLINICAL IMPLICATIONS: Concordant with the literature, the results of this study suggest that individual interventions with women with advanced stage ovarian cancer may help improve neurocognitive-related mood symptoms and anxiety during the initial phase of treatment and diagnosis. Decline in depression symptoms in intervention participants was primarily related to decline in neurocognitive complaints. A reduction in neurocognitive mood symptoms and anxiety likely contributed to the improvement in psychosocial adjustment to illness noted in the treatment group during the follow-up period of the study. The impact of psychosocial sequelae on cognitive functioning warrants attention by clinicians. ACKNOWLEDGEMENT OF FUNDING: Baylor University Medical Center Research Foundation