Johns Hopkins links online sports betting legalization to rises in anxiety and depression care

The legalization of online sports betting was associated with increases in outpatient diagnoses and prescriptions for depression and anxiety among people with employer‑sponsored insurance, according to a Johns Hopkins Bloomberg School of Public Health study published in JAMA Psychiatry on 14 October 2026. The analysis compared medical claims from 2016 to 2023, covering nearly 62 million person‑years of enrollment, and found relative increases of up to 7,20% in the probability of an anxiety diagnosis.

## Key finding on online sports betting legalization

The paper "Online Sports Betting Legalization and Behavioral Health Care Use" estimates that legalization between 2018 and 2022 in 22 states, compared with 13 states that had not legalized by the end of 2023, was associated with higher probabilities of outpatient diagnoses and medicine use for anxiety and depression. For men, the probability of an outpatient depression diagnosis rose by 6,73% relative to the sample mean and antidepressant use increased by 5,44%; for women, the relative increases were 5,47% and 5,00%, respectively. The relative probability of an anxiety diagnosis rose 7,20% for men and 6,50% for women, and anxiety medication use rose 4,99% and 5,17%.

## Absolute changes reported

The study abstract also reports point‑change effects: the absolute probability of receiving an antidepressant rose 0,57 points for men and 1,08 points for women. The probability of an outpatient anxiety diagnosis increased 0,67 points for men and 1,15 points for women. The authors note that for men the antidepressant increase is roughly six additional men per 1,000 insured who fill at least one prescription per year.

## Study design and limits

The author team comprises Benjamin Westermeyer (PhD candidate), Matthew D. Eisenberg and Mark K. Meiselbach, assistant professor of health policy and management at the Bloomberg School. The study used a staggered difference‑in‑differences quasi‑experimental design applied to a national convenience sample of people aged 18 to 64 with employer‑sponsored insurance: 61,951,305 person‑years of enrollment, 52% women, mean age 41.9.

The authors excluded states that legalized before 2018 or in 2023 and measured outcomes at the state‑year level: the share of enrollees with at least one medication fill or at least one visit for depression, anxiety, or substance use disorder. The paper acknowledges two clear limits: results apply only to people with employer insurance, and claims data do not identify who placed bets, so it is not possible to link individual betting to use of health services.

"I hope these results make clinicians and online bettors aware that these apps can cause harms beyond financial risk," said Mark K. Meiselbach.

"The findings support concern that there are broad mental‑health effects from online gambling legalization," said Benjamin Westermeyer.

## What the study did not show and international comparisons

No changes were detected in treatment for substance use disorder nor associations with emergency visits or hospital admissions. This contrasts with a University of Toronto study on Ontario, where emergency visits for gambling disorder doubled after the province opened its online market. The two papers, the authors explain, capture different severity levels: the Johns Hopkins study records outpatient care and prescriptions, not only acute episodes coded as gambling disorder.

## Implications for operators, regulators and payers

For operators and regulators, the result suggests many effects may appear outside usual industry harm metrics (help‑line calls, self‑exclusions, prevalence surveys). Depression care managed in primary care and prescription fills are not necessarily attributed to gambling by patients or clinicians, which makes detection from industry internal indicators difficult. The study also places a cost on employer health plans: the additional mental‑health burden can affect budgets already committed to gambling‑harm programs. Considerations for platform product teams and [payments](/payments) managers include how to surface clinical signals that claims data reveal.

The paper arrives as states report partial use of earmarked funds: in fiscal 2025 $145 million of the $188 million reserved for problem‑gambling services were spent.

## Financial and clinical evidence converge, with caveats

The paper adds a clinical component to predominantly financial population evidence. Prior research cited includes a UCLA Anderson study that found a 25–30% increase in the probability of personal bankruptcy in states with online sports betting and an approximate 2‑point drop in credit scores; and a Journal of Financial Economics paper that found reduced savings and increased credit‑card debt among financially constrained households. Literature is not uniform: a 2024 Economics Letters article found mental‑health improvements for men aged 18–24 after legalization and mixed effects by age and marital status.

## Open questions

The authors do not claim individual‑level causation; the population design estimates state‑level associations. The lack of identification of who bet prevents confirming betting as the causal channel. Also, the sample excludes uninsured people and Medicaid beneficiaries, populations that, because of economic constraints, might experience different or more intense effects.

For primary reading and documentation, the study is published in JAMA Psychiatry (DOI 10.1001/jamapsychiatry.2026.3175), and the Johns Hopkins Bloomberg School of Public Health issued a statement with methodological details and quotes from the authors. This topic is relevant both to those following regulation and to teams managing risk in the b2b side of platforms and payments.