Response Letter of Liu and Zhang

To the Editor:

We appreciate the opportunity to respond to the letter of Liu and Zhang regarding our study: Mild traumatic brain injury in older adults: recovery course and insights on early predictors of outcome.

Although we value a critical discussion in research, several concerns raised by the authors are either already addressed in our article or based on assumptions that do not align with our methodological choices and findings. Please find our reply below.

First, we are aware that the exclusion of patients with psychiatric, neurological, or severe comorbid conditions has an impact on the generalizability of our findings. Our study, however, explicitly aimed to investigate recovery courses in a well-defined cohort of older adults with mTBI to avoid confounding factors that could obscure the specific effects of mTBI in older adults. Specifically, because studies on the recovery course of older adults with mTBI are scarce. The influence of preinjury mental health conditions on recovery is indeed an important topic—one that our own research group has investigated in older patients with mTBI. ¹ It is therefore noteworthy that the authors cite specifically our own work and do not fully consider the broader scope of our research efforts.

Second, the authors suggest to use a machine learning approach, which has superior predictive performance to a regression analyses. Choosing between different statistical methods to predict outcome is based on the aim of a study and the nature of the dataset. In our study, we aimed to predict clinical outcome based on several factors used in clinical practice. While more complex models such as SVM may be suitable for nonlinear relationships, regression analysis offers advantages in terms of interpretability and implementation, making it particularly valuable in clinical settings where understanding of the contribution of individual variables is crucial for guiding care and treatment decisions.

Regarding the observed sex differences in recovery, we agree that it is noteworthy that males were more likely to recover at 6 months post-injury. The study on sex differences referred to by the authors, however, focuses on sport-related concussion, ² which involves a completely different and younger population compared to older patients with mTBI. Therefore, findings are not directly comparable to our study. We briefly have described sex differences in our discussion section and refer to a recent publication of the CENTER-TBI study by Mikolic et al. ³ In addition, while further research into underlying mechanisms is warranted, examining interaction effects between sex and other variables to design a gender-specific intervention as suggested was not the aim of our study.

The remark on the cognitive reserve assessment through the Cognitive Reserve Index Questionnaire (CRIq) is redundant. We have acknowledged that CRIq has limitations in our limitation section. Still, it remains a widely used and validated measure in research on cognitive reserve. ⁴ The suggested use of neuropsychological testing, while valuable, was beyond the feasibility of our study design but was also already recommended in our limitations section of the article.

Finally, the observation that some patients transitioned from complete to incomplete recovery between the two follow-ups was discussed in our article, including possible explanations related to comorbid conditions and life events. In total, only eight patients (5,5% of the total) transitioned from the recovery group. This group was too small for further analysis, but it certainly deserves attention in future research with larger cohorts.

We appreciate the opportunity to clarify these points and stand for the methodological approach and clinical relevance of our study.