Document resource
Objectives This study aimed to determine an minimally important difference (MID) in 6MWD for pulmonary hypertension associated with interstitial lung disease (PH-ILD) using clinical trial data (INCREASE NCT02630316), using different analytical methods, and structured expert elicitation (SEE).Methods A distributional approach was needed to be used to determine the MID. Distributional approaches commonly involve a fraction of the standard deviation (SD), typically 0.5SD. The 0.5SD can either be defined using the baseline SD or the SD of the change from baseline, also known as the effect size (ES) and standardised response mean (SRM), respectively. The trial comprises of both a treatment (Treprostinil, TRT) and placebo (PLB) arm. The TRT arm was selected for the MID using the SRM, 0.5SD change with a change from baseline at 16 weeks. A second analysis was undertaken focusing on all patients whose 6MWD improved over 16 weeks, irrespective of the study arm. The study also determined clinical expert MCID values through SEE using STEER and SHELF methods.Results In the trial the SD for placebo was thirty-percent greater than for the TRT arm suggesting a greater degree of uncertainty associated with placebo relative to treatment. The MID estimate derived using the first analysis was 25.8 m for TRT. The MID also corresponds to the 10% improvement from baseline. The second approach demonstrated a consistent score of 19 m as indicative of a MID. The SEE methodology provided a MCID of 31 m however, when an expert response was excluded based on task comprehension concerns and lack of PH experience, an MCID of 28.6 m was established.Conclusion This study has established an interim 6MWD MID for PH-ILD that ranges between 19 m to 31 m. Due to the variability of the MID for PH-ILD, this study is a ‘call-to-action’ for more research to be conducted.