More than 2.5 million high school students sustain concussions in the United States each year.1 Adolescents spend more than 7 hours daily on screen time (ST).2 Historically, expert opinion recommended ST abstinence to various degrees, worrying that ST would be harmful to concussion recovery.3–5 Last year, we published a randomized clinical trial of concussion patients aged 12 to 25 years old, which found that those instructed to abstain from ST for the first 48 hours recovered 4.5 days sooner than those who were permitted ST.6 Our study demonstrated that early ST limitations improved time to recovery but prompted many questions regarding the dosing, timing, and mechanism of ST limitations.
In this issue of Pediatrics, Cairncross et al shed light on the effect of ST on recovery after concussion.7 Using data from a large, longitudinal cohort study of children (Advancing Concussion Assessment in Pediatrics [A-CAP]), they investigated the association of ST on self- and parent-reported symptom severity in 712 children and adolescents with concussion or orthopedic injury (OI). Applying robust statistical models to control for known predictors of concussion outcomes as well as pre- and postinjury lifestyle behaviors, the authors found that ST had a small but statistically significant nonlinear effect on worsening symptom severity of the concussion compared with the OI group. Interestingly, this effect was only significantly different between the groups for self-reported cognitive symptoms and parent-reported somatic symptoms and not for parent-reported cognitive nor self-reported somatic symptoms. It is notable that ST only accounted for 0.6% to 3.5% of the proportion of their model’s total predictive ability, suggesting ST is significant but not one of the most influential factors in concussion recovery. Perhaps most important, among most of the Health and Behavior Inventory scales used in this study, there was a “Goldilocks effect,” demonstrating that those in the 25th to 50th percentiles had less severe symptoms than those on the higher end of ST use or those who minimally used screens.
Mechanistically, it makes sense that low to moderate ST may be the sweet spot for recovery. In our study, the ST-abstinent group still used a median of 130 minutes of ST over the intervention period (interquartile range, 61–275 minutes), suggesting full abstinence may not be tenable and a small amount may not be harmful. The Goldilocks effect demonstrated in this article may be true for many postacute activities including exercise,8,9 school participation, and sleep, in which children need to strike a balance between avoiding boredom, deconditioning, isolation, and the nocebo effect without overexerting themselves physically or cognitively. The Goldilocks effect was also evident in the OI group, which demonstrated similar and sometimes even more pronounced variability in symptom severity by ST quintile than the concussion group. This potentially dampened the demonstrated effect of ST in concussion using this group as a comparison. It would be interesting to see if the Goldilocks effect is present in healthy children.
There are important differences to note between this study and our RCT. The population in our study was older and thus more likely to own and rely on smartphones compared with children at the younger age range (8 to 16 years old) of this study. We prescribed ST instead of observing self-regulated ST. Our study only examined ST in the first 48 hours, whereas this study looked at ST use retrospectively over the first week. Perhaps there is something unique about ST in the first 48 hours versus just a few days later.
There are limitations to studying ST effect on concussion recovery that must be recognized. Physical activity confounds ST analysis. This study controlled for physical activity reported in the Healthy Lifestyle Behavioral Questionnaire, but it is still unclear what activities substituted for ST and how patients and parents might report activities of mild exertion (eg, chores, a light walk, cooking, gardening, etc). Patients in this study were not prescribed an ST limit. There may be unrecognized confounding among certain demographics, regional ST recommendations, concussive symptoms, or their severities that influenced patients to limit ST. Additionally, the role of ST on sleep duration and quality is an important association that is difficult to control when assessing sleep duration alone. The authors noted that postinjury napping, preinjury ST use, and older age predicted greater symptom severity, all of which relate to hours of sleep and may influence or be influenced by postinjury ST. Many questions remain regarding ST after concussion: Do different types of ST have variable effects? Does the consecutive duration matter more than total ST? Does blue light or brightness matter? Is it ST or being sedentary that is harmful? Does age or concussion subtype make a difference?
Taken together, the study by Cairncross et al supports growing evidence regarding ST limitations after concussion. By using rigorous statistics and evaluating the effect size of ST, the investigators have given clinicians new guidance regarding the relative importance of early ST recommendations after concussion.
COMPANION PAPER: A companion to this article can be found online at www.peditrics.org/cgi/doi/10.1542/peds.2022-056835.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest to disclose.