Psychosocial Functioning
All BASC mean T-scores were in the normal range at follow-up (Table 3), although scores were variable ranging from within the normal range to the impaired range across most scales. There was a significant negative correlation between age at baseline and the Behavioral Symptoms Index (r = -0.19, n = 187, p = 0.01) and a positive correlation with age at baseline and the Adaptive Skills Composite (r = 0.29, n = 187, p < 0.001). That is, younger patients at baseline had more behavior problems at follow-up while older patients had more problems with adaptive skills. Histology was significantly related to the Behavioral Symptoms Index mean T-scores (F (5, 181) = 2.41, p < 0.05); the ependymoma group had a higher score on average (M = 51.29, S.D. = 8.78) than the other histology groups, although near the mean for age. No significant relationship was found between the Adaptive Skills Composite mean T-scores and histology (F (5, 181) = 1.22, p = 0.30). No significant relationships (p values ≥ 0.10) were found between the Behavioral Symptoms Index or the Adaptive Skills Composite mean T-scores and the following factors: gender, hydrocephalus at diagnosis, history of surgical resection, tumor location or treatment with chemotherapy. Neither the Behavioral Symptoms Index nor the Adaptive Skills Composite T-scores were significantly correlated with the time interval between baseline and follow-up (p values ≥ 0.33) or with SES (pvalues ≥ 0.39). Notably, radiation field was not significantly related to the mean T-scores for the Behavioral Symptoms Index (F (1, 185) = 0.59, p = 0.45) or the Adaptive Skills Composite (F(1, 185) = 0.28, p = 0.60). No further analyses were conducted of the interaction between radiation field and age at baseline since mean T-scores for these two measures were not significantly different by radiation field and age at baseline was not significantly related to radiation field.
Compared to the normative mean, a statistically significantly elevated mean T-score was found in Withdrawal (t (172) = 3.89, p< 0.001) while mean T-scores were significantly lower (indicating less problems) than expectation in Hyperactivity (t (172) = -3.54, p< 0.01), Aggression (t (169) = -7.72, p < 0.001), Conduct Problems (t (149) = -7.93, p < 0.001), Adaptability (t (170) = -2.52, p < 0.05), Social Skills (t (171) = -4.78, p < 0.001), and the Behavioral Symptoms Index (t (186) = -2.17, p < 0.05). Mean T-scores were at or near the normative mean on the remaining BASC scales (p values ≥ 0.06), including Anxiety, Depression, Somatization, Attention Problems, and the Adaptive Skills Composite.
Rates of impairment for the Behavioral Symptoms Index and Adaptive Skills Composite were similar to expected rates in the general population as were the rates of impairment on the following clinical and adaptive scales: Hyperactivity, Aggression, Conduct Problems, Depression, Anxiety, Atypicality, Attention Problems, Adaptability, Social Skills, Leadership, and Functional Communication. In contrast, rates of impairment in Withdrawal (17.9%), Activities of Daily Living (15.0%), and Somatization (12.5%) scales exceeded expectations. There were no significant differences between rates of impairment and radiation field on any scale (p values ≥ 0.10) or between rates of impairment and age at baseline on any scale (pvalues ≥ 0.15).