Methods
In the Princess Máxima Center, approximately 60% of families of children in outpatient cancer care participate in regular psychosocial monitoring and screening with the KLIK patient-reported outcome measure portal.7, 8 Assessment takes place every three to six months, combined with their outpatient appointments, and families are asked to provide informed consent to use their clinical data for research. This study used assessments between 01-01-2020 and 01-06-2020 and considered March 13 as the start of the COVID-19 era with national preventative measures, including social distancing and advice to stay at home. Personal and medical characteristics of children (sex, age, diagnosis group [solid tumor, CNS tumor or hematological cancer], time since diagnosis and treatment status [currently in or after treatment]) were extracted from medical records.
Outcomes
Health-related quality of life (HRQoL) and fatigue of children with cancer were assessed using the age-appropriate Dutch proxy- (age: 2-7 years old) or self-report (age: 8-18 years old) pediatric quality of life inventory (PedsQL) generic and multidimensional fatigue scales. Higher scores (scale: 0-100) indicate higher HRQoL or better functioning (e.g.: better emotional function or less fatigue). The PedsQL generic and fatigue scales have good psychometric properties.9, 10 Cronbach’s alphas in this study ranged from 0.74-0.94.
Using the distress thermometer for parents (DT-P), caregivers self-reported their overall distress regarding physical, emotional, social and practical issues on a 0-10 scaled thermometer, scores of 4 or higher indicating clinical distress.11 In addition, experienced problems were assessed for 6 domains and summed to 2 total problem scores. Psychometric properties of the DT-P score are good.11, 12 Cronbach’s alphas in this study ranged from 0.65-0.91.
Statistical analyses
Participant characteristics and outcomes were described for the pre-COVID-19 and COVID-19 era. Pre-COVID-19 and COVID-19 era scores were compared using mixed effect linear regression analyses, to correct for repeated measurements in a minority of participants and both caregivers of a child filling out the DT-P. The occurrence of clinical distress was analyzed using a logistic generalized estimating equation model with exchangeable structure. To adjust for multiple testing, the level of statistical significance was set at .05 divided by the number of analyses conducted on (sub)scales of a questionnaire (e.g. for PedsQL generic: 0.05/6=0.008). Models were corrected for the personal and medical characteristics and effect modification of treatment status was evaluated.