Trial Outcomes & Findings for The PediQUEST Response Intervention Study (NCT NCT03408314)

Participants were recruited from five U.S. pediatric oncology centers between March 2018 and September 2022. Following enrollment, dyads completed a 2-week run-in period to identify responders. Responders were defined as dyads in which both parent and patient completed at least 2 of 3 assigned PediQUEST surveys during the run-in period. Eligible responder dyads were randomized 1:1 to the PediQUEST Response intervention or Usual Cancer Care and followed for 16 weeks.

We enrolled 194 child-parent dyads (we stopped short of the protocol target 196 when reaching randomization target). Forty dyads did not meet randomization criteria (non-responders) and were excluded. We initially planned to randomize 136 dyads; after DSMB review of a planned interim analysis, the target increased to ≥150 dyads based on baseline scores distributions. Enrollment stopped after the randomization target was achieved. A total of 154 dyads (total: 308 participants) were randomized.

We randomized 154 child-parent dyads (80 children and parents in the Usual care arm and 74 child-parent dyads in the intervention arm). Baseline characteristics are reported separately for children and parents (in different rows). Missing data are indicated within the tables.

The primary outcome was child Pediatric Quality of Life Inventory 4.0 (PedsQL) total score, analyzed by respondent: (a) parent-reported for all enrolled children (n=154), and (b) child-reported (if 5 years of age or older) (n=103). PedsQL total scores are calculated as the average of the 23 items that integrate the tool. Scores range from 0-100 (100 is excellent quality of life). A score of \<70 is considered fair/poor quality of life. Intervention effect was measured as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Secondary child PedsQL subscale outcomes included: (a) PedsQL physical (8 items) and (b) psychosocial (15 items) subscale scores (range 0-100, higher is better). Intervention effect was measured as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Secondary child PedsQL subscale outcomes included: (a) PedsQL physical (8 items) and (b) psychosocial (15 items) subscale scores (range 0-100, higher is better). Intervention effect was measured as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Secondary child outcome PediQUEST-Memorial Symptom Assessment Scale (PQ-MSAS) total scores calculated as the average of individual symptom scores (26 for teens and caregivers and 8 for children aged 7-12). Individual symptom scores are calculated as the average of the three symptom subquestions. All symptom scores and total scores range from 0-100, 100 indicates higher symptom burden. Intervention effect calculated as: mean difference between arms of the difference between the 16-week average and baseline scores as reported by (a) the parent for all enrolled children and (b) the patient, if 7 years of age or older. Effect estimated using mixed linear models with child as a random effect. Unadjusted models included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Secondary child outcome PediQUEST-Memorial Symptom Assessment Scale (PQ-MSAS) physical subscale scores calculated as the average of eight physical symptom scores (only for teens and caregivers). Scores range from 0-100, 100 indicates higher symptom burden. Time Frame: Baseline (average of run-in scores, i.e. 2-3 scores) and 16 weeks (measured weekly from point of randomization, the 16-week average was used). Intervention effect was measured as the mean difference between arms of the difference between the 16-week average and baseline scores as reported by (a) the parent for all enrolled children and (b) the patient, if ≥13 years of age. Effect estimated using mixed linear models with child as a random effect. Unadjusted models included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Secondary child outcome PediQUEST-Memorial Symptom Assessment Scale (PQ-MSAS) psychosocial subscale scores calculated as the average of 6 psychological symptom scores (only for teens and caregivers). Scores range from 0-100, 100 indicates higher symptom burden. Intervention effect was measured as the mean difference between arms of the difference between the 16-week average and baseline scores as reported by (a) the parent for all enrolled children and (b) the patient, if ≥13 years of age. Effect estimated using mixed linear models with child as a random effect. Unadjusted models included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Parent anxiety measured using Spielberger's-State Anxiety Inventory-State tool (STAI-S). Scores range from 20 to 80, with higher scores correlating with greater anxiety. Intervention effect estimated as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Parent depression measured using the Center for Epidemiologic Studies Short Depression Scale (CES-D-10) tool. Scores range from 0 to 30 (Score of 10 or higher indicates the presence of significant depressive symptoms). Intervention effect estimated as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Number of complementary therapies used over the past month were reported by parents using an ad hoc checklist of 33 therapies based on an NIH Center for Complementary and Integrative Health list. Parents reported both use (yes/no) of each therapy and reasons for use. Intervention effect estimated as the mean difference between arms of the difference between the 16-week average and baseline number of complementary therapies used estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Parent stress was measured using an adapted version of the Response to Stress Questionnaire-Pain portion. The tool assesses 12 potential stress sources. Scores range from 1-4 (higher means more stress). Symptom-related stress score calculated as the sum of the 12 item scores. Intervention effect estimated as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Coping was measured using selected items from the short version of the Coping Orientation to Problems Experienced Inventory (BRIEF COPE) tool including the instrumental support scale score. Scale score is calculated as the average of the items that measure the strategy. Items scored 1-4 (higher means using that strategy a lot). Intervention effect estimated as the mean difference between arms of the difference between the 16-week average and baseline scores estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects. We report below on instrumental support scores.

Types of complementary therapies used were measured using an ad hoc checklist of complementary therapies as explained for outcome 9. Complementary measures were grouped as physical, psychological, nutritional, combined, and whole-system approaches based on NCCIH framework. Intervention effect estimated as the mean difference between arms of the difference between the 16-week average and baseline types of therapies used estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.

Information on use of psychosocial services was gathered from the medical record and operationalized as "number of encounters documented with psychosocial clinicians." Intervention effect estimated as the mean difference between arms of the 16-week average of encounters with psychosocial clinicians estimated using mixed linear models with child as a random effect. For unadjusted models we included study arm, time (categorical, 0 to 16 weeks) and arm-by-time interaction as fixed effects.