Abstract
Introduction:
Unintentional injuries and assaults cause nearly 28 million emergency department visits each year and can incur lasting physical and psychological damage. This study examined whether unintentional injuries and violence increase mental health disorders and spending.
Methods:
In 2025, authors analyzed 2022 U.S. MarketScan employer-sponsored insurance enrollees (N=19.1 million). Enrollees with emergency department visits for unintentional (n=105,389) or assault (n=1,986) injuries were matched to controls (by age, sex as reported in the data source, residence region, health plan type, and comorbidity count). All subjects were observed 6 months before and 12 months after the emergency department visit date. New mental health disorder diagnosis (anxiety, bipolar disorder, major depression, other mood disorders, attention-deficit/hyperactivity disorder, post-traumatic stress disorder, sleep disorders), new psychotropic drug prescription, total injury-attributable medical spending, and mental health–specific medical spending were assessed. Costs were assessed as 2022 U.S. dollars.
Results:
People with emergency department–treated unintentional injuries were 34%, and those with assault injuries were 98% more likely than controls to have a new mental health diagnosis in the year after the emergency department visit. They were also 50% and 71% more likely, respectively, to have a new psychotropic drug prescription. Excess spending on mental health services and prescriptions during the year after the injury averaged $302 for unintentional injuries and $711 for assaults.
Conclusions:
Unintentional injuries and assaults render affected individuals more vulnerable to mental health disorders and increase mental healthcare spending. It is critical to ensure efficient access to mental health services and consider mental health costs as a part of the economic burden of injuries and violence.
INTRODUCTION
Unintentional injuries such as falls and motor vehicle crashes are the third leading all-ages cause of death, and homicide is a top 5 leading cause of death for people aged 1–54 years in the U.S.1 Nonfatal unintentional injuries and assaults caused over 27.6 million emergency department (ED) visits in 2023.1 Injuries and violence can cause lasting physical and psychological damage as well as high personal and societal costs.2–5 Research has increased the understanding of risk factors for mental health disorders in patients with traumatic brain injury as well as other selected injury diagnoses and affected subpopulations—for example, physical assault of children and workers with occupational injuries.6,7 Administrative claims data offer an opportunity to comprehensively quantify the impact of injuries by intent (e.g., unintentional, assault) on the incidence of mental health disorders and associated medical spending among large population samples. These are key issues for public health decision making that aims to develop and implement cost-effective prevention strategies addressing broad social and environmental factors leading to injuries and violence. This study aimed to measure the incidence of new mental health disorder diagnoses, psychotropic drug prescriptions, and medical spending associated with unintentional and assault injuries in the U.S.
METHODS
Authors examined a large convenience sample of U.S. employer-sponsored health insurance enrollees with ED visits for unintentional or assault injuries in the 2022 MarketScan commercial databases, which report enrollee healthcare services and expenditures from ~350 health insurance payers. Costs were assessed as 2022 U. S. dollars. This retrospective matched cohort analysis followed the STROBE and Consolidated Health Economic Evaluation Reporting Standards reporting guidelines, used deidentified data, and did not require institutional review (Common Rule 45 CFR§46).
The analysis applied previously reported methods to estimate attributable medical spending using claims data.8 Exposure was an injury-related ED visit (index visit)—the first ED visit of the calendar year with primary diagnosis (identified by the facility charge record) of unintentional or assault injury using ICD-10-CM codes.9,10 Intent was classified by first-listed external cause of injury codes. Control enrollees were exact matched 1:5 on the basis of age, sex (male/female), residence region, health plan type (e.g., preferred provider organization), number of comorbidities (0, 1, or ≥2 documented within 6 months preceding the index visit using Elixhauser Comorbidity Software Refined for ICD-10-CM V2024.1), and same continuous enrollment (6 months before and 12 months after the index visit month). Enrollees with missing data for matching characteristics, capitated insurance, another injury diagnosis, or mental health disorder diagnosis or those dispensed psychotropic drugs within 6 months preceding the index visit were excluded.
Outcome measures were new mental health disorder diagnosis (yes/no) or psychotropic drug prescription (yes/no), total injury-attributable healthcare spending, and mental health–specific spending during 12 months after the ED injury visit. Mental health disorder diagnoses were identified by ICD-10-CM codes in inpatient or outpatient settings: anxiety (F40, F41, F42), major depression (F32, F33), bipolar disorder (F30), other mood disorders (F34, F39), attention-deficit/ hyperactivity disorder (F90), post-traumatic stress disorder (F43), and sleep disorders (G47, F51). Psychotropic drugs were identified by National Drug Codes for antidepressants, antipsychotics, anxiolytics, mood stabilizers, hypnotics, and stimulants (Appendix Table 1, available online).11 Mental health spending comprised the total cost of inpatient and outpatient care visits when the assessed diagnosis codes were documented plus outpatient pharmaceutical spending on the assessed psychotropic drugs.
Probabilities for new mental health disorder diagnoses and psychotropic drug prescriptions for injury patients versus controls were estimated by applying standard postestimation calculations to logistic regression model results; marginal effects (model-adjusted percentage point differences between patients and controls for the outcomes of interest) and semielasticities (model-adjusted proportional differences) are reported (details are provided in table notes). Average 1-year total injury-attributable medical spending and mental health–specific spending per injured person were assessed as the marginal effect of an injury indicator in 2-part models of injury and control patients’ total medical spending. Models controlled for all matching characteristics. Subgroup analysis examined children (aged <18 years) separately. Analyses were conducted using Stata 18 (Stata-Corp) and SAS 9.4 (SAS Institute, Cary, NC).
RESULTS
Among 19,136,928 enrollees, 107,375 met study inclusion criteria (Appendix Figure 1, available online). Among those, 105,389 people had unintentional injuries, and 1,986 had assault injuries and were matched to 522,521 and 9,805 control enrollees (groups were balanced for all matching characteristics) (Appendix Table 2, available online). People with unintentional injuries were 34.2% (95% CI=31.4, 37.0) more likely and those with assault injuries were 98.4% (95% CI=82.0, 116.2) more likely to have a new mental health disorder diagnosis in the following year than people without injuries as well as 49.5% (95% CI=44.1, 55.3) and 70.6% (95% CI=55.0, 87.8) more likely to be dispensed a psychotropic drug (Table 1). The largest probability increase was for post-traumatic stress disorder (+53.0% [95% CI=42.8, 64.0] and +354.9% [95% CI=286.9, 434.9] for unintentional and assault injuries, respectively), but large increased relative probabilities were observed for all assessed mental health disorder diagnoses. Each person with an unintentional injury had on average $6,201 (95% CI=5,985; 6,416) in excess medical spending, of which at least $302 (95% CI=267, 338) was specific to mental health services and treatment, in the year after the ED visit. Each person with an assault injury had on average $6,602 (95% CI=5,315; 7,889) in excess medical spending, of which at least $711 (95% CI=434, 987) was specific to mental health services and treatment (Table 2).
Table 1.
Incidence of Mental Health Disorder Diagnosis and Psychotropic Drug Prescription in the Year After an Injury
| Unintentional injury (n=627,910) | Assault (n=11,791) | |||||||
|---|---|---|---|---|---|---|---|---|
| Incidence % (95% CI)a | Difference (95% CI) | Incidence % (95% CI)a | Difference (95% CI) | |||||
| Injury | Control | Percentage pointsb | Proportional changec | Injury | Control | Percentage pointsb | Proportional changec | |
| Any diagnosis | 15.6 (15.5, 15.8) | 11.6 (11.5, 11.8) | 4.0 (3.7, 4.3) | +34.2% (31.4, 37.0) | 26.4 (25.5, 27.3) | 13.3 (12.6, 14.0) | 13.1 (11.5, 14.7) | +98.4% (82.0, 116.2) |
| Anxiety | 7.3 (7.2, 7.4) | 5.3 (5.2, 5.4) | 2.0 (1.8, 2.2) | +37.4% (33.6, 41.2) | 13.0 (11.6, 14.5) | 6.9 (6.4, 7.4) | 6.1 (4.3, 8.0) | +88.5% (58.4, 124.3) |
| ADHD | 2.0 (1.9, 2.1) | 1.5 (1.5, 1.6) | 0.5 (0.4, 0.6) | +32.7% (22.8, 43.5) | 3.7 (3.4, 4.1) | 1.6 (1.3, 1.8) | 2.1 (1.6, 2.7) | +133.0% (85.0, 193.3) |
| Depression | 3.7 (3.6, 3.8) | 2.5 (2.4, 2.6) | 1.2 (1.0, 1.3) | +47.7% (40.5, 55.4) | 9.2 (8.8, 9.6) | 3.5 (3.0, 3.9) | 5.7 (5.2, 6.3) | +166.4% (132.8, 205.3) |
| Other mood disorder | 0.5 (0.5, 0.5) | 0.3 (0.3, 0.3) | 0.2 (0.1, 0.2) | +52.0% (44.9, 59.4) | 1.4 (1.3, 1.5) | 0.4 (0.4, 0.5) | 0.9 (0.8, 1.0) | +220.3% (177.9, 269.1) |
| PTSD | 3.1 (3.0, 3.2) | 2.0 (2.0, 2.1) | 1.1 (0.9, 1.3) | +53.0% (42.8, 64.0) | 10.1 (9.9, 10.3) | 2.2 (1.9, 2.5) | 7.9 (7.4, 8.4) | +354.9% (286.9, 434.9) |
| Sleep disorder | 4.1 (4.0, 4.2) | 3.1 (3.0, 3.2) | 1.0 (0.8, 1.2) | +32.4% (26.5, 38.7) | 3.7 (3.4, 4.0) | 2.8 (2.4, 3.3) | 0.9 (0.2, 1.6) | +32.3% (5.9, 65.4) |
| Psychotropic prescription | 13.4 (13.3, 13.5) | 8.9 (8.6, 9.3) | 4.4 (4.1, 4.8) | +49.5% (44.1, 55.3) | 18.3 (17.7, 19.0) | 10.8 (10.1, 11.5) | 7.6 (6.3, 8.9) | +70.6% (55.0, 87.8) |
Data source: 2022 Merative MarketScan commercial insurance databases.
Note: Counts include injury and control enrollees. Owing to small enrollee sample size, bipolar disorder is not separately presented.
Regression model–predicted average probability (Stata margins).
Marginal effect associated with injury (Stata margins, dydx[injury]).
Semielasticity associated with injury; exponentiated values produced by Stata margins, eydx(injury) were transformed using 1-exponentiated result.
ADHD, attention-deficit/hyperactivity disorder; PTSD, post-traumatic stress disorder.
Table 2.
Injury-Attributable Medical Spending on Mental Health Services in the Year After an Injury
| Unintentional injury (n=627,910) | Assault (n=11,791) | |||||||
|---|---|---|---|---|---|---|---|---|
| Cost (95% CI)a | Difference (95% CI) | Cost (95% CI)a | Difference (95% CI) | |||||
| Injury | Control | Absoluteb | Proportionalc | Injury | Control | Absoluteb | Proportionalc | |
| Totald | $10,995 (10,786; 11,205) | $4,733 (4,671; 4,795) | $6,201 (5,985; 6,416) | 130.9% (125.7, 136.1) | $11,274 (9,994; 12,554) | $4,585 (4,169; 5,000) | $6,602 (5,315; 7,889) | 143.8% (112.8, 179.3) |
| Mental health specific | $603 (568, 639) | $298 (287, 309) | $302 (267, 338) | 101.4% (88.7, 114.9) | $1,107 (824; 1,391) | $395 (297, 493) | $711 (434, 987) | 179.8% (105.2, 281.5) |
Data source: 2022 Merative MarketScan commercial insurance databases.
Note: Counts include injury and control enrollees.
Two-part regression model–predicted average medical spending (Stata margins).
Marginal effect associated with injury (Stata margins, dydx[injury]).
Semielasticity associated with injury; exponentiated values produced by Stata margins, eydx(injury) were transformed using 1- exponentiated result.
Because all patients with injury had >$0 total 1-year medical spending, the 2-part model can accommodate an injury covariate (i.e., identifying patients with injury) in the generalized linear model part (second part) but not logistic part (first part).
Analysis restricted to children indicated that those with unintentional injuries (n=42,728) were 30.2% (95% CI=26.1, 34.4) more likely and those with assault injuries (n=490) were 157.0% (95% CI=96.4, 236.4) more likely to have a new mental health disorder diagnosis in the following year than people without injuries—including a 16-fold increase in the probability of a mood disorder diagnosis (+1,616.7% [95% CI=565.3; 4,330.1])—as well as 43.5% (95% CI=37.7, 49.5) and 129.1% (95% CI=61.8, 224.5) more likely to be dispensed a psychotropic drug (Appendix Table 3, available online). Mental health–specific spending attributable to injuries in the year after the ED injury visit was $94 for children with unintentional injuries and $712 for children with assault injuries (Appendix Table 4, available online).
DISCUSSION
In this retrospective matched cohort analysis, unintentional and assault injuries were associated with increased probability of new mental health disorder diagnoses, new psychotropic drug prescriptions, and increased mental health–related medical spending. Clinical practice and public health decision making benefit from research evidence highlighting both the physical and psychological impact of injuries and violence. It is critical to ensure efficient access to mental health services and consider mental health costs as a part of the economic burden of injuries and violence.
Limitations
Injury diagnosis (e.g., fracture), severity, and mechanism (e.g., fall) were not addressed nor were public insurance costs nor the uninsured population. This study did not examine the impact of the coronavirus disease 2019 (COVID-19) pandemic nor other characteristics associated with mental health disorder diagnosis after injury beyond available demographic, clinical, and health insurance factors in the data source.12 This study did not investigate self-harm among injured people.13 Costs were assessed during only the first year after an injury ED visit.
CONCLUSIONS
Unintentional injuries and assaults render affected individuals more vulnerable to mental health disorders and increase mental health spending. In providing comprehensive assessment of the impact of injuries by intent on multiple mental health disorder diagnoses as well as associated medical spending, these results can assist decision makers in identifying opportunities for cost-effective public health strategies to reduce the substantial economic burden of injuries and violence in the U.S.
Supplementary Material
Supplemental materials associated with this article can be found in the online version at https://doi.org/10.1016/j.amepre.2026.108322.
ACKNOLWEDGMENTS
CD had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Footnotes
Disclaimer: The findings and conclusions in this manuscript are those of the authors and do not necessarily reflect the official position of the Centers for Disease Control and Prevention.
Declaration of interest: None.
CREDIT AUTHOR STATEMENT
Christopher Dunphy: Conceptualization, Methodology, Software, Formal analysis, Investigation, Resources, Writing – original draft, Writing – review & editing, Visualization, Project administration. Cora Peterson: Methodology, Investigation, Resources, Writing – original draft, Writing – review & editing, Project administration. Marwa Rawy: Conceptualization, methodology, investigation, resources, writing – review & editing, project administration. Likang Xu: Methodology, Software, Formal analysis, Investigation, Resources, Data curation, Writing – review & editing, Project administration. Kun Zhang: Conceptualization, Methodology, Investigation, Resources, Writing – review & editing, Project administration. Sha Zhu: Methodology, software, validation, formal analysis, investigation, resources, data curation, writing – review & editing, project administration. Curtis Florence: Methodology, Investigation, Resources, Writing – original draft, Writing – review & editing, Project administration, Supervision.
REFERENCES
- 1.Web-Based Injury Statistics Query and Reporting System. National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. https://wisqars.cdc.gov. Accessed February 5, 2026. [Google Scholar]
- 2.van der Sluis CK, Eisma WH, Groothoff JW, ten Duis HJ. Long-term physical, psychological and social consequences of severe injuries. Injury. 1998;29(4):281–285. 10.1016/s0020-1383(97)00199-x. [DOI] [PubMed] [Google Scholar]
- 3.Peterson C, Miller GF, Barnett SBL, Florence C. Economic cost of injury - United States, 2019. MMWR Morb Mortal Wkly Rep. 2021;70 (48):1655–1659. 10.15585/mmwr.mm7048a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Johnson PL, Hemmila MR, Mullens CL, et al. Association of pre- and postinjury mental health with long-term clinical and financial outcomes. J Trauma Acute Care Surg. 2025;98(3):425–434. 10.1097/ta.0000000000004521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Murphy PB, Severance S, Savage S, Obeng-Gyasi S, Timsina LR, Zarzaur BL. Financial toxicity is associated with worse physical and emotional long-term outcomes after traumatic injury. J Trauma Acute Care Surg. 2019;87(5):1189–1196. 10.1097/ta.0000000000002409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Archambault É, Vigod SN, Brown HK, et al. Mental illness following physical assault among children. JAMA Netw Open. 2023;6(8): e2329172. 10.1001/jamanetworkopen.2023.29172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ogonah MGT, Botchway S, Yu R, Schofield PW, Fazel S. An umbrella review of health outcomes following traumatic brain injury. Nat Ment Health. 2025;3(1):83–91. 10.1038/s44220-024-00356-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Peterson C, Xu L, Zhu S, Dunphy C, Florence C. Medical and work loss costs of violence, self-harm, unintentional and traumatic brain injuries per injured person in the USA. Inj Prev. 2025;31(4):333–339. 10.1136/ip-2024-045259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hedegaard H and Johnson RL, An Updated International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) surveillance case definition for injury hospitalizations, Natl Health Stat Report, (125), 2019, 1–8. https://pubmed.ncbi.nlm.nih.gov/31751206/. Accessed May 1, 2026. [PubMed] [Google Scholar]
- 10.Hedegaard H, Johnson RL, Garnett MF and Thomas KE, The International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) external cause-of-injury framework for categorizing mechanism and intent of injury, Natl Health Stat Report, 136, 2019, 1–22. https://pubmed.ncbi.nlm.nih.gov/32510317/. Accesed May 1, 2026. [PubMed] [Google Scholar]
- 11.Wettstein ZS, Vaidyanathan A. Psychotropic medication prescriptions and Large California wildfires. JAMA Netw Open. 2024;7(2):e2356466. 10.1001/jamanetworkopen.2023.56466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Østergaard SD, Mundy J, Hall ASM, Musliner KL. Genetic confounding in the association between traumatic brain injury and mental disorder or suicide. JAMA Psychiatry. 2025;82(6):622–626. 10.1001/jamapsychiatry.2025.0318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yaw LK, Burrell M, Ho KM. Long-term outcomes and determinants of new-onset mental health conditions after trauma. JAMA Netw Open. 2025;8(3):e250349. 10.1001/jamanetworkopen.2025.0349. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.