Feasibility of Conducting a Palliative Care Randomized Controlled Trial in Children With Advanced Cancer: Assessment of the PediQUEST Study

Abstract

Context: Pediatric palliative care randomized controlled trials (PPC-RCTs) are uncommon.

Objectives: To evaluate the feasibility of conducting a PPC-RCT in pediatric cancer patients.

Methods: This was a cohort study embedded in the Pediatric Quality of Life and Evaluation of Symptoms Technology Study (NCT01838564). This multicenter PPC-RCT evaluated an electronic patient-reported outcomes system. Children aged two years and older, with advanced cancer, and potentially eligible for the study were included. Outcomes included: pre-inclusion attrition (patients not approached, refusals); post-inclusion attrition (drop-out, elimination, death, and intermittent attrition (IA; missing surveys) over nine months of follow-up); child/teenager self-report rates; and, reasons to enroll/participate.

Results: Over five years, of the 339 identified patients, 231 were eligible (in 22, we could not verify eligibility); 84 eligible patients were not approached and 43 declined participation. Patients not approached were more likely to die or have brain tumors. We enrolled 104 patients. Average enrollment rate was one patient per site per month; shortening follow-up from nine to three months (with optional re-enrollment) increased recruitment by 20%. A total of 87 patients completed the study (24 died) and 17 dropped out. Median IA was 41% in the first 20 weeks of follow-up and more than 60% in the eight weeks preceding death. Child/teenager self-report was 94%. Helping others, low burden procedures, incentives, and staff attitude were frequent reasons to enroll/participate.

Conclusion: A PPC-RCT in children with advanced cancer was feasible, post-inclusion retention adequate; many families participated for altruistic reasons. Strategies that may further PPC-RCT feasibility include: increasing target population through large multicenter studies, approaching sicker patients, preventing exclusion of certain patient groups, and improving data collection at end of life.

Keywords: Pediatrics; attrition; end-of-life care; feasibility; palliative care; patient-reported outcomes; pediatric oncology; randomized controlled trial; supportive care.

Copyright © 2015 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

Figures

Fig. 1

Methods used in the Pediatric Quality of Life and Evaluation of Symptoms Technology (PediQUEST) Study – RCT Methods (adapted from Wolfe J et al.). 1The PediQUEST-Survey (PQ-Survey) is an electronic survey that includes the adapted PediQUEST-Memorial Symptom Assessment Scale (PQ-MSAS, which evaluates symptom burden) and the Pediatric Quality of Life Inventory (PedsQL4.0™, which evaluates quality of life).

Fig. 2

The Pediatric Quality of Life and Evaluation of Symptoms Technology (PediQUEST) Study flow diagram at nine months of follow-up. The Figure shows the sampling process and disposition at nine months of follow-up. Indicators used to study pre- and post-inclusion attrition are shown. Level A shows the target population; Level B: subjects missed and identified by study procedures; Level C: ineligible and eligible subjects; Level D: not approached and approached subjects; Level E: refused and enrolled; Level F: disposition at nine months of follow-up.

Fig. 3

Families’ perceptions, values, and reasons to enroll in or decline the PediQUEST Study. Panel A shows perceptions of enrolled (83/104) and non-enrolled (12/43) families regarding study's consent and decision-making processes, as well as values and reasons underlying their participation decision. Non-enrollees responded to an abridged survey version. Given the low non-enrolled response rate, no statistical testing was conducted. Only items with fewer than 10% missing values are graphed. Panel B presents reasons to enroll or not in the PediQUEST Study as reported in open-ended questions (or quantitative items in the five surveys where comments were not provided). Over a third of respondents provided more than one reason. Representative quotes from enrolled and non-enrolled respondents are shown.

Fig. 4

Disposition of patients enrolled in the PediQUEST Study over nine months of follow-up and feasibility of longitudinal assessment of electronic patient-reported outcomes. Line graph (top): number of study subjects available per week after censoring for: death (red line); death, drop-out, and elimination (green line); death, drop-out, elimination, and administrative censoring (black line). Area graph (bottom): number of eligible PQ opportunities at each time point and the number that resulted in: an administered survey (green area), a not administered survey (purple) where reason for no administration was known; or a missed opportunity (light purple), i.e. opportunities that were inadvertently missed by research staff.

Fig. 5

Reasons to stay enrolled in the PediQUEST randomized controlled study. This “word cloud” provides a proportional representation of all open-ended responses to the question “Which was the main factor that kept you in the (PediQUEST) study all the way?” Word size relates to the number of times the word appears. Graph built online using wordle.com application.