Cosmos Study

Gambling Disorder Diagnoses Have Risen More Than 60% in States That Legalized Sports Betting

June 26, 2026

Team A:Kersten Bartelt, RNCaleb CoxJoe Deckert, PhD

Team B:Brad Fox, MDSam Sahakian

Key Findings

Among adult patients living in states that have legalized sports betting, the quarterly rate of diagnosed gambling disorder rose from 3.0 per 100,000 patients in Q1 2018 to 4.8 per 100,000 in Q1 2026, around a 61% increase.
In states that have not legalized sports betting, the rate moved in the opposite direction, falling from 3.1 per 100,000 in Q1 2018 to 2.2 per 100,000 in Q1 2026, a 29% decrease.
Of the overall population, adults aged 30 to 49 had the highest overall quarterly rate of diagnosed gambling disorder throughout the study period, reaching 5.8 per 100,000 by Q1 2026. The largest proportional increase was among adults aged 18 to 29, whose rate more than doubled. Rates rose in both sexes and in every age group, but the male-to-female gap was substantially larger in adults under 50.

Gambling disorder is a recognized behavioral health condition in which a person continues to gamble despite financial, emotional, or social harm. It was reclassified as an addictive disorder in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) in 2013, alongside substance use disorders.¹ In May 2018, the U.S. Supreme Court struck down the federal Professional and Amateur Sports Protection Act in Murphy v. NCAA, clearing the way for individual states to legalize sports betting; by early 2026, sports betting was legal in 39 states and the District of Columbia.² National survey and internet-search data point to rising public interest in gambling addiction help-seeking since 2018,³ but population-level estimates of diagnosed gambling disorder in U.S. clinical care have been limited.

We studied more than 197 million U.S. adults aged 18 and older with at least one qualifying visit between January 2018 and March 2026. Results were stratified by state legalization status (patients residing in states that have legalized sports betting versus states that have not), and separately by age group and sex.

Diagnosis trajectories diverged sharply by state legalization status. In Q1 2018, when only Nevada had legalized sports betting, rates were similar across the two groups: 3.0 per 100,000 in states that had or would later legalize sports betting and 3.1 per 100,000 in states that have not. By Q1 2026, the rate in states that have legalized sports betting had risen to 4.8 per 100,000, while the rate in states that haven’t legalized sports betting had fallen to 2.2 per 100,000, as shown in Figure 1. Notably, the rise in states that have legalized sports betting appeared roughly in parallel across those states rather than staggered by each state’s specific legalization year. One possible explanation is the parallel growth of online prediction markets such as Polymarket and Kalshi, which are not regulated as gambling at the state level and are accessible regardless of where a user lives. Because legalization is staggered and gambling disorder is widely under-recognized in clinical care, these trends cannot be attributed to any single legal event and likely understate true population prevalence. Even so, the directional contrast aligns with national survey and search-trend evidence of rising gambling-related help-seeking concentrated in states that have legalized sports betting.³

Figure 1

Quarterly Rate of Diagnosed Gambling Disorder by State Sports Betting Legalization Status

Figure 1. The quarterly rate per 100,000 adult patients with a diagnosed gambling disorder by whether the patient resides in a state that has ever legalized sports betting.

Diagnosis rates varied substantially by age and sex. Adults aged 30–49 had the highest overall rates throughout the study period, rising from 4.1 to 5.8 per 100,000 between Q1 2018 and Q1 2026, with adults aged 50 to 64 close behind (4.1 to 4.9). The most striking trajectory, however, was among the youngest adults: rates in those aged 18–29 more than doubled. Rates in adults 65 and older were lower throughout but also trended upward. Across every age band, rates in men were consistently higher than in women, and rates rose in both sexes over time, as shown in Figure 2, with the male-to-female gap widest in adults under 50. These patterns are directionally consistent with prior survey-based work identifying men and younger-to-middle-aged adults as the groups most likely to participate in sports betting and to screen positive for problem gambling behaviors.⁴

Figure 2

Quarterly Rate of Diagnosed Gambling Disorder by Age Group and Evaluated Sex

Figure 2. The quarterly rate per 100,000 adult patients with a diagnosed gambling disorder by age group and evaluated sex.

These data come from Cosmos, a dataset created in collaboration with a community of Epic health systems representing more than 304 million patient records from 2,000 hospitals and more than 47,000 clinics from all 50 U.S. states, Canada, Lebanon, and Saudi Arabia. This study was completed by two teams that worked independently, each composed of a clinician and research scientist. The two teams came to similar conclusions. Graphics by Brian Olson.

References

Petry NM, Blanco C, Auriacombe M, et al. An overview of and rationale for changes proposed for pathological gambling in DSM-5. J Gambl Stud. 2014;30(2):493-502. doi:10.1007/s10899-013-9370-0
Murphy v. National Collegiate Athletic Association, 584 U.S. 453 (2018). U.S. Supreme Court. Decided May 14, 2018. https://www.supremecourt.gov/opinions/17pdf/16-476_dbfi.pdf. Accessed May 19, 2026.
Yeola A, Allen A, Desai N, et al. Growing health concern regarding gambling addiction in the age of sportsbooks. JAMA Intern Med. 2025;185(4):454-456. doi:10.1001/jamainternmed.2024.8193
National survey on gambling attitudes and gambling experiences (NGAGE 3.0). National Council on Problem Gambling. Published 2024. https://www.ncpgambling.org/training/ngage-survey/ngage-3/. Accessed May 19, 2026.

Data Definitions

Study period: 1/1/2018 to 3/31/2026

Study population: inclusion: Adult patients meeting all of the following:
• Have a residence in the U.S.
• Have at least one qualifying encounter during the study period
• Aged 18 or older at the time of the qualifying encounter

Exposure: Calendar time (quarter), Q1 2018 through Q1 2026

Outcome: Quarterly rate of distinct active patients with a gambling-related encounter out of all distinct active patients in that quarter, expressed per 100,000 patients.

Qualifying encounter: An encounter with one of the following types:
• Office Visit
• Telemedicine
• Hospital Outpatient Visit
• Clinical Support
• Social Work
• Inpatient Admission
• Emergency
• Emergency to Inpatient
• E-Visit
• Follow-Up

Active patient: A patient with at least one qualifying encounter during a given quarter. Used as the denominator for the rate measure.

Gambling-related encounter: An encounter with a gambling diagnosis recorded as one of the following diagnosis sources:
• Encounter Diagnosis
• Billing Final Diagnosis
• Billing Procedure-Linked Diagnosis

Gambling diagnosis: Pathological gambling / DSM-5 gambling disorder: ICD-10-CM code F63.0
Gambling and betting: ICD-10-CM code Z72.6

State legalization status: Ever legalized: Patient with a residence in one of the 39 states or the District of Columbia that legalized sports betting during the study period.
Never legalized: patient resides in one of the 11 states that have not legalized sports betting in any form as of the end of the study period (Alabama, Alaska, California, Georgia, Hawaii, Idaho, Minnesota, Oklahoma, South Carolina, Texas, Utah)

Stratifications: State sports betting legalization status (have legalized vs have not legalized)
Age group: 18–29, 30–49, 50–64, 65+
Evaluated sex

Limitations: • Gambling disorder is widely under-recognized as a healthcare issue, and many affected individuals do not present to medical care. The Cosmos rate represents a floor on diagnosed gambling and is not a population prevalence estimate.
• Coding-practice drift: F63.0 may be more recognized and coded in recent years than in 2018 due to DSM-5 diffusion, Mental Health Parity and Addiction Equity Act enforcement, and growing clinician awareness. A real change in prevalence cannot be cleanly distinguished from a coding or recognition change.
• Staggered state-level legalization: sports betting and prediction-market legalization took effect at different times across states, with no single timestamp marking impact onset. The study cannot attribute trends to specific legal events.
• 42 CFR Part 2 reporting variability: some organizations or specific clinicians may not record gambling, substance use, or behavioral health diagnoses in the chart due to their interpretation of 42 CFR Part 2. We assume this pattern is roughly stable over time and does not contaminate trend interpretation but cannot verify directly.
• COVID-19 utilization shock: the pandemic decreased volume of elective visits and shifted encounter mix toward telehealth. The proportion-based denominator absorbs most of the volume effect, but residual differential capture of behavioral health diagnoses cannot be ruled out.
• Telehealth expansion since 2020 may differentially affect behavioral health diagnosis capture; the proportion-based denominator absorbs most but not all of this.
• Cross-state moves are not modeled, so a patient who moved from a have not legalized to a legalized state during the study period contributes only to their most recent state of residence.