Ascertainment of Minimal Clinically Important Differences in the Diabetes Distress Scale–17: A Secondary Analysis of a Randomized Clinical Trial

IMPORTANCE: The Diabetes Distress Scale–17 (DDS-17) is a common measure of diabetes distress. Despite its popularity, there are no agreed-on minimal clinically important difference (MCID) values for the DDS-17. OBJECTIVE: To establish a distribution-based metric for MCID in the DDS-17 and its 4 subscale scores (interpersonal distress, physician distress, regimen distress, and emotional distress). DESIGN, SETTING, AND PARTICIPANTS: This secondary analysis of a randomized clinical trial used baseline and postintervention data from a hybrid (implementation-effectiveness) trial evaluating Empowering Patients in Chronic Care (EPICC) vs an enhanced form of usual care (EUC). Participants included adults with uncontrolled type 2 diabetes (glycated hemoglobin A(1c) [HbA(1c)] level >8.0%) who received primary care during the prior year in participating Department of Veterans Affairs clinics across Illinois, Indiana, and Texas. Data collection was completed in November 2018, and data analysis was completed in June 2023. INTERVENTIONS: Participants in EPICC attended 6 group sessions led by health care professionals based on collaborative goal-setting theory. EUC included diabetes education. MAIN OUTCOMES AND MEASURES: The main outcome was distribution-based MCID values for the total DDS-17 and 4 DDS-17 subscales, calculated using the standard error of measurement. Baseline to postintervention changes in DDS-17 and its 4 subscale scores were grouped into 3 categories: improved, no change, and worsened. Multilevel logistic and linear regression models examined associations between treatment group and MCID change categories and whether improvement in HbA(1c) varied in association with MCID category. RESULTS: A total of 248 individuals with complete DDS-17 data were included (mean [SD] age, 67.4 [8.3] years; 235 [94.76%] men), with 123 participants in the EPICC group and 125 participants in the EUC group. The MCID value for DDS-17 was 0.25 and MCID values for the 4 distress subscales were 0.38 for emotional and interpersonal distress and 0.39 for physician and regimen distress. Compared with EUC, more EPICC participants were in the MCID improvement category on DDS-17 (63 participants [51.22%] vs 40 participants [32.00%]; P = .003) and fewer EPICC participants were in the worsened category (20 participants [16.26%] vs 39 participants [31.20%]; P = .008). There was no direct association of DDS-17 MCID improvement (β = −0.25; 95% CI, −0.59 to 0.10; P = .17) or worsening (β = 0.18; 95% CI, −0.22 to 0.59; P = .38) with HbA(1c) levels among all participants. CONCLUSIONS AND RELEVANCE: In this secondary analysis of data from a randomized clinical trial, an MCID improvement or worsening of more than 0.25 on the DDS-17 was quantitatively significant and patients in the EPICC group were more likely to experience improvement than those in the EUC group. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT01876485