Johns Hopkins Study Ties Legal Online Sports Betting to Up to 7.2% Rise in Depression and Anxiety Care

States that legalised online sports betting saw more depression and anxiety prescriptions and outpatient diagnoses among the employer-insured, a JAMA Psychiatry paper finds. It is a population-level association, not proof of harm to bettors, and it adds a clinical strand to evidence that has so far been mostly financial.

• Researchers at the Johns Hopkins Bloomberg School of Public Health found that legalising online sports betting was associated with higher probabilities of outpatient depression diagnoses, 6.73% for men and 5.47% for women relative to the sample averages, in a study published online in JAMA Psychiatry on 14 October 2026
• On the same relative measure, the probability of antidepressant use was 5.44% higher for men and 5.00% for women, of an anxiety diagnosis 7.20% and 6.50%, and of anti-anxiety medication use 4.99% and 5.17%
• The study used employer-sponsored insurance claims from January 2016 to December 2023, nearly 62 million enrollee-years, comparing 22 states that legalised online sports betting between 2018 and 2022 with 13 that had not by the end of 2023
• It found no change in substance use disorder treatment and no association with emergency department or inpatient visits, a contrast with Ontario, where gambling-disorder emergency visits doubled after its online market opened
• The authors say they could not link individual betting to health care use, and the results apply only to people with employer-sponsored insurance

Legalising States Saw More Prescriptions and Outpatient Diagnoses

States that legalised online sports betting saw increases in filled prescriptions for antidepressants and anti-anxiety medications, and in outpatient diagnoses of depression and anxiety, according to "Online Sports Betting Legalization and Behavioral Health Care Use", published online in JAMA Psychiatry on 14 October. Its authors are Benjamin Westermeyer, a PhD candidate, Matthew D. Eisenberg and the senior author Mark K. Meiselbach, an assistant professor of health policy and management at the Bloomberg School.

The paper describes a quasi-experimental staggered difference-in-differences study of claims from a national convenience sample of people aged 18 to 64 with employer-sponsored insurance: 61,951,305 enrollee-years, 52% of them women, with a mean age of 41.9. The 22 states that legalised online sports betting between 2018 and 2022 were compared with 13 that had not by the end of 2023, and states that legalised in 2023 or before 2018 were excluded, according to the university's announcement. Outcomes are measured at state-year level: the share of enrollees with at least one prescription fill, or at least one visit, for depression, anxiety or substance use disorder.

The headline percentages are relative increases against the sample averages. In absolute terms, the abstract reports that the probability of an antidepressant fill rose by 0.57 percentage points for men and 1.08 points for women, and of an outpatient anxiety diagnosis by 0.67 and 1.15 points. For men, the antidepressant estimate is roughly six more in every 1,000 insured men filling at least one prescription in a year. Women's confidence intervals are wider, 0.23 to 1.94 points for antidepressants against 0.26 to 0.89 for men, and the paper says the associations were "more precise and robust for men than women".

There was no detectable change in substance use disorder treatment and no association with emergency or inpatient visits. The authors conclude that legalisation may increase strain on the behavioural health system.

What the Paper Can and Cannot Show

The authors acknowledge two limits, according to the announcement: the results apply only to people with employer-sponsored insurance, and claims data do not record who bets, so individual betting could not be linked to health care use. The abstract does not name the claims database or the estimator. The full text, including funding and conflict of interest statements, is behind the journal's paywall, and iGaming Times has not been able to check them.

Meiselbach said he hoped the results would make doctors and online sports bettors aware "that these online apps can lead to harms beyond financial risks". Westermeyer said the findings "support the concern that there are broad mental health effects of online gambling legalization, though those who are experiencing those effects, and their clinicians, don't necessarily attribute it to gambling".

Legal online sports betting spread from one state to 31 after 2018, the paper notes. The data end in December 2023, before sports contracts on prediction markets spread, a gap GamblingHarm.org, an advocacy outlet, pointed out on 15 October. By 16 October, iGaming Times had found no published response from the American Gaming Association.

The Effect Shows Up in Doctors' Offices, Not Emergency Rooms

More outpatient diagnoses and prescriptions, with no change in emergency or inpatient care, describe mental health care delivered quietly and, on Westermeyer's account, often not recognised as gambling-related. The University of Toronto study of Ontario emergency visits counted acute cases coded as gambling disorder; the two measure different populations at different ends of the severity range, and do not conflict. For operators and regulators, the point is that the industry's usual harm indicators (helpline calls, self-exclusions, screened problem-gambling rates) would not register depression treated by a family doctor. Nor would prevalence surveys such as early UNLV data on Colorado's problem sports bettors, which count only those who screen positive.

A Population Estimate Cannot Say Who Is Being Harmed

If betting is the channel, an effect of about 6% across all insured adults implies a much larger one among the minority who bet, but the data cannot confirm that betting is the channel. Comparing states over the same years absorbs national shocks such as the pandemic, but assumes legalising and other states were on similar paths beforehand; how that was tested is in the paywalled text. The employer-insured sample also leaves out the uninsured and people on Medicaid, and household-finance research finds harm concentrates among the financially constrained, so the estimate may understate rather than overstate the effect. Nor is the literature uniform: a 2024 Economics Letters paper by Hamilton College economists Cody Couture, Jeffrey Cross and Stephen Wu found mental health improved for men aged 18 to 24 after legalisation, declined for married men aged 30 to 34, and did not move for women.

Financial and Clinical Evidence Now Point the Same Way

Much of the population-level evidence on legal sports betting has been financial. A working paper by Brett Hollenbeck, Poet Larsen and Davide Proserpio found the likelihood of personal bankruptcy rose by 25% to 30% in states with online sports betting, with credit scores down about 2%, according to UCLA Anderson's summary, while retail-only legalisation showed small or insignificant effects. A paper by Scott Baker and colleagues in the Journal of Financial Economics found online betting reduced savings and raised credit card debt and overdrafts among financially constrained households. The Johns Hopkins paper uses the same broad design and adds a cost that lands on employer health plans. It arrives as states struggle to deploy the treatment money they already have, spending $145 million of the $188 million set aside for problem gambling services in FY2025.

The study does not prove that online sports betting causes depression or anxiety, and its authors do not claim that it does. It does put an estimated health care cost beside the tax receipts, and that is a harder number for legislators and insurers to set aside than a survey.