Utilization of electronic patient-reported outcome measures in cystic fibrosis research: Application to the GALAXY study

Abstract

Background: The Food and Drug Administration considers patient-reported outcome measures (PROMs) an essential part of clinical research studies for approval of new drugs and new indications for existing drugs. GALAXY evaluated the feasibility of electronic PROMs (ePROMS) to conduct a comprehensive evaluation of gastrointestinal (GI) symptoms in persons with cystic fibrosis (pwCF).

Methods: Three validated GI ePROMs (PAC-SYM, PAGI-SYM and PAC-QOL) were combined with a Stool-Specific questionnaire to make up the GALAXY ePROMs and administered prospectively across 26 CF centers in the United States. The ePROMs were completed at enrollment visit and then electronically at weeks 1, 2 and 4. PwCF at least 2 years and older were eligible for the study. Reminders were only provided by the mobile application during the study window.

Results: There were 402 participants enrolled in GALAXY. Of those, 169 (42%) were under 18 years old and 193 (48%) were female. The proportion of all follow-up weeks with at least 1 ePROM fully completed was 80%, slightly higher in those ≥18 years of age (82.5%) compared to those <18 years of age (76.5%). When assessing the completion for all 4 ePROMs, the percentage was 77.6%, also higher among those ≥18 year of age (81.5% versus 72.2% for < 18 years of age).

Conclusion: Using ePROMs, our study demonstrated that GI symptoms can be feasibly collected with good reproducibility and with minimal involvement of research coordinator time. This mechanism of symptom collection may provide an efficient tool for future CF trials.

Keywords: Electronic PROMs; Patient reported outcomes measures; cystic fibrosis; gastrointestinal manifestations of cystic fibrosis.

Figure 1:. CONSORT Diagram

The CONSORT diagram shows the number screened (N=422), eligible (N=406), and included in the statistical analyses (N=402).

Figure 2:. Summary of ePROMs completion rates by week.

Full completion for one questionnaire is defined as no question left unanswered on that questionnaire. Follow-up completion rates from week 1 through week 4 were derived based on the total number of participants of corresponding subgroups multiplied by 3 weeks. Note that PAC-QOL was only collected at week 0 and week 4.

Figure 3:. Changes in PAC-SYM, PAGI-SYM, and PAC-QOL scores at follow-up times

Changes in scores from in-clinic week 0 visit for PAC-SYM, PAGI-SYM, and PAC-QOL, among those who fully completed these questionnaires at all follow-up weeks, were summarized at each follow-up time points. Mean and 95% confidence intervals were obtained using two-sided one-sample t-tests.