RESUMO
BACKGROUND AND AIMS: The impact of the COVID-19 pandemic on gambling participation and levels of gambling harm across populations during the pandemic is now addressed in a well-established body of empirical literature. This study aimed to measure the longer-term implications of COVID-19 on gambling participation and levels of gambling harm. DESIGN: Population-based cohort study using group-based trajectory modelling. SETTING: Australia, using gambling participation, problem gambling risk, sociodemographic and psychosocial data from 2019 (pre COVID-19), 2020, 2021 (during COVID-19) and 2023 (post COVID-19). PARTICIPANTS: A population representative survey of Australian adults, including four waves collected in April 2019 (n = 2054), November 2020 (n = 3029), October 2021 (n = 3474) and January 2023 (n = 3370), with a subset (n = 3160) of the sample having longitudinal data available. MEASUREMENTS: Participants were asked which gambling activities they participated in over the past 12 months for money. Problem gambling risk was measured by the nine-item Problem Gambling Severity Index (PGSI). FINDINGS: There was an overall reduction in gambling participation during COVID-19 and return to pre-pandemic levels for most gambling activities by 2023. The longitudinal analysis yielded four trajectories of gambling participation from 2019 to 2023, including individuals who (1) never gambled (25.0% of the longitudinal sample; n = 789); (2) engaged in non-problematic gambling (59.8%; n = 1888); (3) ceased gambling during COVID-19 and started again post pandemic (10.7%; n = 337); and (4) engaged in high risk gambling (4.6%; n = 146), with particular demographic and psychosocial profiles and patterns of participation in specific gambling activities related to these trajectories. CONCLUSIONS: Although overall gambling participation rates decreased at the population level in Australia during COVID-19, by 2023 participation in gambling appeared to have nearly returned to pre-pandemic levels. Patterns of gambling behavior before, during and after the pandemic appear to be heterogeneous.