Family relationships are the subjective experience of feeling involved, important and accepted in one's family. A new patient-reported outcome (PRO) domain has been developed as part of the NIH pediatric PROMIS (patient reported outcome measurement information system) which measures for the first time the child's perspective of family relationships. Previous research has demonstrated the impact of cancer on the family from the parent's perspective. This project sought to determine the impact of cancer and cancer therapy on family relationships in children with cancer and their siblings, both on and off therapy. Our primary hypotheses were that 1) Siblings have worse family relationships compared to their brother or sister with cancer and 2) Children receiving active cancer therapy have worse family relationships compared to childhood cancer survivors. Our secondary hypothesis was that children with poor family relationships would have impairment in depression, anxiety, and social relationship domains.


We conducted a cross sectional study of children with cancer and their siblings aged 8-17 years old. The children belonged to one of four groups: 1) On-therapy patients (children receiving cancer directed therapy), 2) Siblings of on-therapy patients, 3) Off-therapy patients (children who completed cancer therapy at least 6 months ago), 4) Siblings of off-therapy patients. All subjects completed the self-reported PROMIS family relationships domain and PROMIS domains of depression, anxiety, and social relationships. The family relationships domain asks questions such as: "my family and I have fun together" and "my parents listen to me". PROMIS is a validated PRO tool that uses a mean T-score of 50 and standard deviation of 10. Recently, the minimally important difference (MID) for PROMIS was estimated to be 2-3 points on the T-score scale. The Mann-Whitney test was used to test the difference between therapy groups while the Wilcoxon signed-rank test was used to test the difference between patients and siblings. Social relationships, anxiety and depression PROMIS domains were correlated to the family relationships domain using Pearson's correlations. P-values <0.05 were considered as significant.


One-hundred and ninety children completed the assessments. Forty-eight were on-therapy patients and 62 were off-therapy patients. Paired data included 25 pairs of on-therapy patients and siblings and 31 pairs of off-therapy patients and siblings. The median and range of age was 12[8-17] years for patients and 13[8-17] for siblings. Fifty-four percent of the children enrolled were male and 65% of patients had leukemia or lymphoma.

Off-therapy siblings had lower family relationships T-score than off-therapy patients (p=0.004, median (IQR) of 44.1 (41.1, 48.6) for siblings vs. 47.4 (43.1, 53.2) for patients). Off-therapy patients had higher family relationships T-scores than on-therapy patients (p=0.042, median (IQR) of 48.6 (43.1, 53.2) for off-therapy patients vs. 46.3 (42.1, 48.6) for on-therapy patients). No significant difference was found in family relationships T-scores between on-therapy patients vs. their siblings (p=0.22) or between the two groups of siblings (p=0.58).

The correlations between family relationships T-scores and depression or anxiety scores were significantly different from zero for off-therapy patients and their siblings (p<0.001 for patients and p<0.005 for siblings). Impaired family relationships were related to increased depression and anxiety. Also, the correlation between family relationships T-scores and social relationships scores were significantly different from zero for off-therapy patients (p=0.031). Worse family relationships were related to worse social relationships.


Off-therapy patients report better family relationships than their siblings and on-therapy patients. Also worse family relationships scores often correlate with increased levels of anxiety and depression. This study provides important information for pediatric oncology staff to help identify at-risk families in need of support.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution