Family therapy with adolescents with diabetes: A litmus test for clinically meaningful change

As a part of a randomized controlled trial, 119 adolescents with type 1 diabetes mellitus (DM1) and their parents were randomly assigned to participate in either 10 sessions of behavioral family systems therapy (BFST), 10 sessions of an education/support group (EDSP), or no specific psychosocial treatment (CONT). Pre-treatment and post-treatment scores on diabetes-related conflict were collected from family members (i.e., adolescents, mothers, and fathers). Computation of effect sizes were used to determine the clinical significance of family therapy on diabetes-related conflict scores by comparing scores from subjects in the current study with normative data collected from a demographically similar sample. Post-treatment data revealed that changes in adolescents' scores on a measure of diabetes-related conflict were minimally lower (-0.25 SD) for all three groups (BFST, EDSP, and CONT) in comparison to the untreated normative sample. Changes in mothers' scores for the BFST group were significantly lower (-1.0 SD) in comparison to the untreated normative sample and compared to mothers' scores for the EDSP (-0.27 SD) and CONT groups (-0.33 SD). Finally, changes in fathers' scores for the BFST group were moderately lower (-0.47 SD) in comparison to the untreated normative sample and compared to fathers' scores for the EDSP (-0.11 SD) and CONT groups (-0.08 SD).