The challenge of measuring intra-individual change in fatigue during cancer treatment

Abstract

Purpose Evaluate how well three different patient-reported outcomes (PROs) measure individual change. Methods 214 patients (from two sites) initiating first or new chemotherapy for any stage of breast or gastrointestinal cancer participated. The 13-item FACIT–Fatigue scale, a 7-item PROMIS® Fatigue Short-Form (PROMIS 7a), and the PROMIS® Fatigue computer adaptive test (CAT) were administered monthly online for six months. Reliability of measured change was defined, under a population mixed effects model, as the ratio of estimated systematic variance in rate-of-change to the estimated total variance of measured individual differences in rate-of-change. Precision of individual measured change, the standard error of measurement (SEM) of change, was given by the square root of the rate-of-change sampling variance. Linear and quadratic models were examined up to 3 and up to 6 months. Results A linear model for the reliability of measured change showed the following by 6 and by 3 months, respectively: PROMIS CAT (0.363 and 0.342); PROMIS SF (0.408 and 0.533); FACIT (0.459 and 0.473). Quadratic models offered no noteworthy improvement over linear models. Both reliability and precision results demonstrate the need to improve the measurement of intra-individual change. Conclusions These results illustrate the challenge of reliably measuring individual change in fatigue with a level of confidence required for intervention. Optimizing clinically useful measurement of intra-individual differences over time continues to pose a challenge for PROs.

Implementation of Electronic Patient-Reported Outcomes in Routine Cancer Care at an Academic Center: Identifying Opportunities and Challenges

“…41 Even if patients were able to complete the ePRO survey, oncologists also raised concerns about the data timeliness, patient literacy, and overall reliability and validity of PROs, all barriers raised in the literature. 30,42,43 Future efforts should develop processes for patients and educational materials for clinicians to mitigate these concerns.…”

Between-group minimally important change versus individual treatment responders

“…Individual-level variation and change can be estimated using simulation modeling for time series data, but it requires a minimum of 10 observation in the data stream [ 8 ]. Similarly, Moinpour et al [ 9 ] estimated mixed effect models and noted that the PROMIS fatigue computer adaptive test would need 15 total assessments to obtain 0.90 reliability of change. Because of limits on research budgets and concerns about respondent burden, nearly all longitudinal HRQOL studies are limited to a few waves of assessment (e.g., two time points).…”

Can Methods Developed for Interpreting Group-level Patient-reported Outcome Data be Applied to Individual Patient Management?

“…Research requires less measurement reliability because sample sizes provide precision in the estimation of mean PRO levels, and measurement error cancels out across the sample (as it is, by definition, random). In contrast, at the individual level, very precise measures are needed (44). Historically, PRO measures designed for use in managing individual patients in psychology were very long, often containing 100 or more items.…”