Hyperbaric Oxygen Therapy for Severe Pediatric Traumatic Brain Injury: A Secondary Psychometric and Nonparametric...

This secondary analysis builds upon the results of a previous study conducted at the Centre of Hyperbaric Medicine, Ostrava City Hospital, Czech Republic, from 2019 to 2023. Statistically and clinically significant patient responses to hyperbaric oxygen therapy (HBOT) in children with brain injuries were reported. The current study aims to replicate and extend this original work through a secondary analysis. By applying psychometric (reliability and validity) and non-parametric inferential techniques, this research seeks to elucidate previously unexplored data dimensions. The secondary analysis results indicated a moderate level of internal consistency as measured by Cronbach's alpha for the Glasgow coma scale (GCS) across multiple time points (r = 0.719). Due to the skewed nature of the data, non-parametric analyses were necessary. The Glasgow outcome scale (GOS) exhibited strong (r = 0.893) internal consistency. Regarding concurrent validity, robust positive Spearman's correlations (rs = 0.831, p < 0.001, 95% CI (0.615, 0.931)), (rs = 0.826, p < 0.001, 95% CI (0.605, 0.929)), (rs = 0.645, p = 0.002, CI (0.283, 0.846)), (rs = 0.598, p = 0.004, 95% CI (0.211, 0.823)) were found between pre-test and post-test HBOT GCS and GOS scores, highlighting their congruence. However, no significant Spearman correlation (rs = -0.205, p = 0.374, 95% CI (-0.594, 0.262)) emerged between the differences in post-treatment GCS and GOS scores.  When analyzing gender differences, no statistically significant differences (Mann-Whitney U = 43.5, Wilcoxen W = 109.5, p = 0.426) (Mann-Whitney U = 45.0, Wilcoxen W = 111.0, p = 0.512) were observed in GCS and GOS gender outcomes. Similarly, etiological comparisons between traumatic brain injury (TBI) and hypoxic-ischemic encephalopathy (HIE) groups (Mann-Whitney U = 46.5, Wilcoxen W = 82.5, p = 0.910), (Mann-Whitney U = 42.0, Wilcoxen W = 120.0, p = 0.678) did not yield significant etiology differences. Finally, the repeated measures analysis showed a substantial and notable improvement (Wilcoxen Signed Rank Test: z = -2.956, p = 0.003, r = -0.645) in GCS scores following HBOT, indicating and confirming the treatment's large effect size.  Ultimately, the GCS showed moderate internal consistency across different time points, while the GOS had high internal consistency. Positive Spearman's correlations between pre-test and post-test HBOT GCS and GOS scores supported their concurrent validity. However, no significant relationship was found between post-treatment changes in GCS and GOS. There were no significant gender differences in outcomes or significant differences between TBI and HIE groups. Finally, a repeated measures analysis showed a significant and considerable improvement in GCS scores after HBOT, indicating substantial treatment effectiveness.