Patient-reported outcome measures in systemic lupus erythematosus by a web-based application: A randomized, crossover, agreement study.
Uhrenholt L., Høstgaard S., Pedersen JF., Christensen R., Dreyer L., Leffers HCB., Taylor PC., Strand V., Jacobsen S., Voss A., Gregersen JW., Kristensen S.
OBJECTIVES: Patient-reported outcome measures (PROMs) are evaluated in randomized controlled trials (RCTs) in patients with systemic lupus erythematosus (SLE), but not widely used in clinical practice. However, interest in incorporating PROMs into the management of SLE is increasing as PROMs provide a unique insight into the patient's perception of lupus disease activity. The objective was to assess agreement in PROMs answered using a web app versus an outpatient touchscreen among patients with SLE. METHODS: In a crossover RCT, SLE patients answered the following PROMs in a random order using the web app and the outpatient touchscreen: Systemic Lupus Erythematosus Activity Questionnaire (SLAQ) Global Health, SLAQ Symptom, SLAQ Total, SLAQ Worsening, Pain Visual Analog Scale (VAS), Fatigue VAS, Patient Global Health VAS, Health Assessment Questionnaire Disability Index (HAQ-DI), Patient Acceptable Symptom State (PASS), and an Anchoring Question. Equivalence between the two device types was demonstrated if the 95% confidence interval (95% CI) of the difference in PROM scores was within the prespecified equivalence margin. Agreement between the two device types was assessed using mixed linear models. RESULTS: Thirty-four patients with SLE were included. Equivalence was demonstrated between the two device types for SLAQ Global Health with a difference of -0.21 (95% CI: -0.65 to 0.23). Moreover, equivalence was also found for HAQ-DI, Pain VAS, and Fatigue VAS whereas only comparability within the limits of the Minimal Clinically Important Difference (MCID) was demonstrated for VAS Patient Global Health. Statistical comparability was demonstrated for SLAQ Total, SLAQ Worsening, PASS, and Anchoring Question (no predefined MCID/equivalence margins available). However, a statistically significant difference between device types was observed for the SLAQ Symptom of -0.56 (95% CI: -1.10 to -0.01). The difference was, however, very small when considering the scale range of 0-24; thus, it was not judged to be of clinical relevance. Preference for the web app was very high (91.2%). CONCLUSION: For the first time ever, equivalence and comparability between two electronic device types for various PROMs were demonstrated among patients with SLE. Implementation of the device is expected to improve the management of SLE.