Age Differences in the Reproducibility of Seasonal Peak Timing for Alcohol-Associated Injury: A Seven-Year Cosinor and Jackknife Analysis of U.S. Emergency Department Surveillance Data
- 1Texas Academy of Mathematics and Science, University of North Texas, Denton, TX, United States of America
Abstract
BackgroundAlcohol-associated injury is a leading cause of emergency department (ED) utilization in the United States and a clinically important driver of preventable morbidity across the adult lifespan. Prior surveillance research has characterized how the rate and severity of alcohol-associated injury vary by patient age, but whether the seasonal timing of injury risk is equally predictable across age groups (a question directly relevant to the timing of clinical screening intensification and public health intervention) has not been formally tested.
MethodsWe conducted a retrospective surveillance analysis of 45,876 alcohol-associated ED visits among adults aged 18 years and older, identified from the National Electronic Injury Surveillance System (NEISS), 2019-2025 (weighted national estimate: 2,092,319 visits), using the structured Alcohol_Involved indicator introduced into NEISS case abstraction in 2019. Patients were stratified by sex and five age groups (18-24, 25-34, 35-49, 50-64, and [≥]65 years). Single-harmonic cosinor (Poisson) regression was used to estimate the seasonal peak day of injury risk (acrophase) for each stratum. To assess reliability, we performed leave-one-year-out jackknife resampling (seven iterations per group), case-resampling bootstrap confidence intervals (1,000 iterations), and likelihood-ratio tests of seasonal-phase interactions.
ResultsPeak injury timing differed significantly across age groups (X^2 [8] = 2356.2, p < .0001). Adults aged 25-64 years showed a highly reproducible early-to-mid-July peak, with jackknife estimates shifting [≤]14 days when any single study year was excluded. Adults aged [≥]65 years showed significant seasonal variation annually (all p < .0001, amplitude comparable to younger groups) but a pooled peak estimate that shifted by up to 100 days across jackknife iterations. Sex-stratified analyses revealed that this instability was driven entirely by females aged [≥]65 years (jackknife range: 332 days, peak consistently in late October through early January) rather than males aged [≥]65 (jackknife range: 31 days, peak consistently in early August). Hospital admission rates increased monotonically with age from 9.0% (18-24 years) to 31.8% ([≥]65 years).
ConclusionsAlcohol-associated injury follows a reproducible, calendar-stable summer seasonal pattern in adults aged 25-64 years. Among adults [≥]65 years, the previously reported temporal instability is concentrated in the female subgroup, whose seasonal injury risk does not converge on a fixed calendar window. These findings suggest that fixed-calendar prevention and screening strategies are well suited to working-age adults and older men, but older women may require a year-round, individually tailored approach.
Related
- Utility of Rt epidemic trend categories for influenza surveillance: alignment with observed emergency department trends across two influenza seasons in the United States
- Signal detection of substance-related emergency department visits during UEFA EURO 2024 in Germany: enhancing public health surveillance for future mass gatherings.
- Emergency Department Utilization After 60-Day Peripheral Nerve Stimulation Versus Usual Care for Refractory Occipital Neuralgia: A Prospective Propensity-Matched Cohort, Difference-in-Differences Analysis
- Antipsychotic-Induced Parkinsonism With Concurrent Benztropine-Associated Anticholinergic Effects in a Community Emergency Department Setting: A Case Report
- Factors associated with discharge and return visits of emergency department patients with a sepsis best practice alert
BibTeX
@misc{ghuman2026differences,
title = {{Age Differences in the Reproducibility of Seasonal Peak Timing for Alcohol-Associated Injury: A Seven-Year Cosinor and Jackknife Analysis of U.S. Emergency Department Surveillance Data}},
author = {D. Ghuman and T. Achar and D. Gambhirrao},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.27.26361527},
url = {https://www.medrxiv.org/content/10.64898/2026.08.27.26361527v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Age Differences in the Reproducibility of Seasonal Peak Timing for Alcohol-Associated Injury: A Seven-Year Cosinor and Jackknife Analysis of U.S. Emergency Department Surveillance Data
AU - D. Ghuman
AU - T. Achar
AU - D. Gambhirrao
PY - 2026
DA - 2026/08/27
PB - medRxiv
DO - 10.64898/2026.08.27.26361527
UR - https://www.medrxiv.org/content/10.64898/2026.08.27.26361527v1
AB - BackgroundAlcohol-associated injury is a leading cause of emergency department (ED) utilization in the United States and a clinically important driver of preventable morbidity across the adult lifespan. Prior surveillance research has characterized how the rate and severity of alcohol-associated injury vary by patient age, but whether the seasonal timing of injury risk is equally predictable across age groups (a question directly relevant to the timing of clinical screening intensification and public health intervention) has not been formally tested. MethodsWe conducted a retrospective surveillance analysis of 45,876 alcohol-associated ED visits among adults aged 18 years and older, identified from the National Electronic Injury Surveillance System (NEISS), 2019-2025 (weighted national estimate: 2,092,319 visits), using the structured Alcohol_Involved indicator introduced into NEISS case abstraction in 2019. Patients were stratified by sex and five age groups (18-24, 25-34, 35-49, 50-64, and [≥]65 years). Single-harmonic cosinor (Poisson) regression was used to estimate the seasonal peak day of injury risk (acrophase) for each stratum. To assess reliability, we performed leave-one-year-out jackknife resampling (seven iterations per group), case-resampling bootstrap confidence intervals (1,000 iterations), and likelihood-ratio tests of seasonal-phase interactions. ResultsPeak injury timing differed significantly across age groups (X^2 [8] = 2356.2, p < .0001). Adults aged 25-64 years showed a highly reproducible early-to-mid-July peak, with jackknife estimates shifting [≤]14 days when any single study year was excluded. Adults aged [≥]65 years showed significant seasonal variation annually (all p < .0001, amplitude comparable to younger groups) but a pooled peak estimate that shifted by up to 100 days across jackknife iterations. Sex-stratified analyses revealed that this instability was driven entirely by females aged [≥]65 years (jackknife range: 332 days, peak consistently in late October through early January) rather than males aged [≥]65 (jackknife range: 31 days, peak consistently in early August). Hospital admission rates increased monotonically with age from 9.0% (18-24 years) to 31.8% ([≥]65 years). ConclusionsAlcohol-associated injury follows a reproducible, calendar-stable summer seasonal pattern in adults aged 25-64 years. Among adults [≥]65 years, the previously reported temporal instability is concentrated in the female subgroup, whose seasonal injury risk does not converge on a fixed calendar window. These findings suggest that fixed-calendar prevention and screening strategies are well suited to working-age adults and older men, but older women may require a year-round, individually tailored approach.
ER -