Official figures on rates of problem gambling in Britain – which have been put as low as 0.3% – can “mask” the scale of harms by including people who never bet, a report published by the industry regulator has warned.

Various surveys suggest a problem gambling rate in Britain in a range of between 0.3% to 2.5%. But in the new report that figure rose sharply when non-gamblers were excluded – and was even higher among those who gambled on online casinos.

The report produced by the National Centre for Social Research and the University of Glasgow found that 2.5% of British adults may have struggled with problem gambling, eight times higher than previously thought. But the estimated rate of problem gambling was 4.1% for those who had gambled during the last year.

The Gambling Commission has warned that the latest figures use a different methodology from ­previous work and gambling harms may be overstated, or that prior surveys may have underestimated gambling harms.

The survey reported that 24.5% of respondents who had played online slots in the last year appear to show signs of problem gambling, nearly six times higher than the rate for all those who had gambled in the same period.

The industry has used the population-wide figures to try to fight off proposals for stricter gambling regulation. But the rates of problem gambling are significantly higher when non-gamblers are excluded and riskier forms of gambling are examined.

A recent international study published in the Lancet has found that globally about one in six ­people (15.8%) who play online casino games or slots experience gambling disorder.

An Australian government report warned as far back as 2010 of the risk of using population-wide statistics for assessing the harms caused by gambling.

It said it could be “misleading” for policy purposes given that many people do not gamble or do not engage in activities with higher rates of ­problem gambling.

The Gambling Survey for Great Britain published on Thursday by the Gambling Commission stated: “Focus on population prevalence rates masks the strengths of associations between gambling and [problem gambling severity index] scores because they include people who do not gamble or gamble very infrequently.”

Will Prochaska, of the Coalition Against Gambling Ads, said there needed to be an overhaul of legislation to ensure gambling was “permitted but not promoted”. He said a review of gambling legislation by the last government appeared to have been based on figures which may have significantly understated gambling harms. He said: “We need a new gambling act which puts the protection of the public at its heart.”

Don Foster, the Liberal Democrat peer and chair of the Peers for Gambling Reform group, said the ­figures in the new survey were a “wake-up call” for urgent action.

He said ministers should implement the previous government’s plan for a £100m-a-year levy on gambling firms to deliver research, prevention and treatment to reduce gambling-related harms.

He said: “We also need to ban direct marketing and any form of sponsorship in sport. Gambling should be treated as a public health issue in the same way as drugs, alcohol and tobacco.”

Premier League clubs were last week accused of greed after it emerged that 11 clubs will have a front-of-shirt gambling sponsor for the coming season. A ban for front-of-shirt sponsorship will be introduced for the 2026-27 season.

A spokesperson for the Betting & Gaming Council said it was concerned there was a “significant risk” that the latest findings in the Gambling Survey for Great Britain overstated gambling harms.

The spokesperson said: “Our members are committed to raising standards and we welcome any robust study that accurately gauges betting and gaming participation and problem gambling prevalence.”

The Gambling Commission said a new methodology was being used for the survey. It said: “This methodology is new, and it is too early to use it in isolation to make policy or regulatory changes.”