Journal of Patient-Reported Outcomes (Nov 2023)

Psychometric validation of the Pyruvate Kinase Deficiency Diary and Pyruvate Kinase Deficiency Impact Assessment in adults in the phase 3 ACTIVATE trial

  • David A. Andrae,
  • Rachael F. Grace,
  • Adrian Jewett,
  • Brandon Foster,
  • Robert J. Klaassen,
  • Sam Salek,
  • Junlong Li,
  • Feng Tai,
  • Audra N. Boscoe,
  • Erin Zagadailov

DOI
https://doi.org/10.1186/s41687-023-00650-3
Journal volume & issue
Vol. 7, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background Pyruvate kinase (PK) deficiency is a rare hereditary disorder characterized by chronic hemolytic anemia and serious sequalae which negatively affect patient quality of life. This study aimed to psychometrically validate the first disease-specific patient-reported outcome (PRO) instruments: the 7-item PK Deficiency Diary (PKDD) and 12-item PK Deficiency Impact Assessment (PKDIA), designed to assess signs, symptoms, and impacts of PK deficiency in patients enrolled in the ACTIVATE global phase 3 study of mitapivat versus placebo (NCT03548220). Methods All validation analyses for the PKDD and PKDIA were performed on blinded data, with analyses on item integrity, scoring, reliability, and validity conducted on data from screening and baseline. Completion rates and baseline response distributions were characterized using descriptive statistics. Item response modelling was used to inform a weighted scoring system. Reliability was assessed by internal consistency and test–retest reliability; and validity by convergent and known-groups analyses. Results Of the 80 adults enrolled, baseline data were available for 77 (96.3%) and 78 (97.5%) patients for the PKDD and PKDIA, respectively. Item responses skewed right, indicating that mean values exceeded median values, especially for items utilizing a 0–10 numeric scale, which were subsequently recoded to a 0–4 scale; 4 items were removed from the PKDIA due to redundancy or low relevance to the trial population. Both the PKDD and PKDIA demonstrated high internal consistency (McDonald’s coefficient ω = 0.86 and 0.90, respectively), test–retest reliability (intra-class coefficients of 0.94 and 0.87, respectively), and convergent validity with other PROs (linear correlation coefficients [|r|] between 0.30–0.73 and 0.50–0.82, respectively). Conclusions The findings provide evidence of validity and reliability for the PKDD and PKDIA, the first disease-specific PRO measures for PK deficiency, and can therefore increase understanding of, and more accurately capture, the wider impact of PK deficiency on health-related quality of life. Trial registration ClinicalTrials.gov, NCT03548220. Registered June 07, 2018; https://www.clinicaltrials.gov/ct2/show/NCT03548220 .

Keywords