PROMIS Fatigue Item Bank had Clinical Validity across Diverse Chronic Conditions

Abstract

Objective: To evaluate the comparability and responsiveness of Patient-Reported Outcomes Measurement Information System (PROMIS) fatigue item bank across six chronic conditions.

Study design and setting: Individuals (n = 1,430) with chronic obstructive pulmonary disease (n = 125), chronic heart failure (n = 60), chronic back pain (n = 218), major depressive disorder (n = 196), rheumatoid arthritis (n = 521), and cancer (n = 310) completed assessments from the PROMIS fatigue item bank at baseline and a clinically relevant follow-up. The cancer and arthritis samples were followed in observational studies; the other four groups were enrolled immediately before a planned clinical intervention. All participants completed global ratings of change at follow-up. Linear mixed-effects models and standardized response means were estimated to examine clinical validity and responsiveness to change.

Results: All patient groups reported more fatigue than the general population (range = 0.2-1.29 standard deviation worse). The four clinical groups with pretreatment baseline data experienced significant improvement in fatigue at follow-up (effect size range = 0.25-0.91). Individuals reporting better overall health usually experienced larger fatigue changes than those reporting worse overall health.

Conclusion: The results support the PROMIS fatigue measures's responsiveness to change in six different chronic conditions. In addition, these results support the ability of the PROMIS fatigue measures to compare differences in fatigue across a range of chronic conditions, thereby enabling comparative effectiveness research.

Keywords: Chronic conditions; Fatigue; Item bank; PROMIS; Patient-reported outcomes; Responsiveness.

Figure 1

Change in PROMIS Fatigue T-scores Across Clinical Samples CHF=Chronic heart failure; COPD=Chronic Obstructive Pulmonary Disease; Scores reported are on the T-score metric as referenced to the US general population (mean=50; SD=10).[34] Higher scores reflect more fatigue. NOTE: 1. Cancer (group 1 n=81) : patients reported more (worse) fatigue at follow-up rated by the fatigue-specific global change item. Cancer (group 2 n=84) : patients reported less (better) fatigue at follow-up rated by the fatigue-specific global change item. 2. RA (group 1 n=171) :rhemautoid patients who reported more (worse) fatigue at follow-up rated by the fatigue-specific global change item. RA (group 2 n=48) :rhemautoid patients reported less (better) fatigue at follow-up rated by the fatigue-specific global change item.