The second ALSPAC COVID-19 questionnaire was complete by 2711 young adults, with a mean age of 27.8 years (SD: 0.6). The population who responded were predominantly white (> 96%) and 80% had at least A-level qualifications. Of these respondents, 2632 (97.1%) answered the gambling questions: 771 males (29.3%) and 1861 (70.7%) females.
Previously 4304 ALSPAC participants, with a mean age of 24.9 years (SD: 0.6), had answered the 24-year gambling questionnaire: 1501 males (34.9%) and 2803 (65.1%) females. The female preponderance was more marked in the respondents to the COVID-19 questionnaire than in the 24 year sample (p < 0.001). The bias introduced by this female preponderance in the COVID-19 questionnaire probably resulted in an underestimate of the number of regular gamblers.
Of those individuals that answered the gambling questionnaire at 24 years, (N = 4304), 3872 (90%) were sent the COVID questionnaire but only 2160 (56%) answered the questionnaire. Those that did not answer the COVID questionnaire were more likely to be male, have a lower IQ (measured at age 8), have slightly raised hyperactivity scores (measured at age 16.5), were more likely to have a diagnosis of moderate/severe alcohol use disorder (measured at age 22) and more likely to have smoked 100+ cigarettes in their life time (measured at age 25). There was no influence on whether the questionnaire was answered or not of regular gambling at age 24, employment status at age 20, or if there was a history of maternal gambling.
The frequency of gambling during the COVID-19 lockdown, compared with the frequency previously reported at 24 years, is presented in Table 1. Overall, gambling frequency reduced during lockdown for both males and females (< weekly and > = weekly: p < 0.001 both sexes). The gender difference widened in those who engaged in regular gambling, with males nearly 3 times more likely than females to gamble regularly.
Gambling at 24 years strongly predicted whether people gambled or not during lockdown irrespective of gender (OR (95% CI): 10.70 (6.57, 17.44), n = 2112). Using the sub-sample of individuals who gambled at both times (n = 1255) showed that a larger proportion of males compared to females did not change their frequency of gambling (p < 0.001) whereas a larger proportion of females decreased their gambling frequency during lockdown (p < 0.001). The proportion that increased their gambling frequency did not differ between males and females (p = 0.28; Table 2).
The different gambling activities reported in the 24 year and COVID questionnaires are presented in Table 3, which shows a different pattern for regular (weekly or more frequent) gamblers compared to occasional (less than weekly) gamblers. For regular gamblers, a few activities, including national lottery, private betting and online gambling, increased during COVID-19 lockdown compared to what was reported 3 years previously. The proportion gambling on other activities remained stable. In contrast, for occasional gamblers, most gambling activities decreased in frequency or remained stable, with the exception of online gambling which increased in frequency for both occasional and regular gamblers. Online gambling included online poker, bingo, casino games. Online betting, involving placing bets on any event or sport, including e-sports, overall remained at similar rates to 24 years (although most UK based sporting events were cancelled during the lockdown).
Gambling and Mental Health
Anxiety levels reported by the cohort during lockdown were generally high (Table 4), but there did not appear to be a relationship between anxiety levels and gambling frequency (χ2 = 5.8, p = 0.45).
Around 15% of participants reported depressive symptoms during lockdown (Table 5), but likelihood of depression was not associated with gambling frequency (χ2 = 1.4, p = 0.50).
Wellbeing scores (mean [SD]) during lockdown were similar in non-gamblers (44 [8.5]), occasional gamblers (43 [8.8]), and regular gamblers (44 [8.7]; p = 0.20).
Gambling and Alcohol Use
Although frequency of drinking any alcohol (< weekly or > = weekly) during lockdown did not show an association with gambling frequency (p = 0.38), frequency of heavy drinking did (p < 0.001). Participants drinking more than 6 units on one occasion regularly (weekly or more) were more likely to be male [OR (95% CI): 1.62 (1.31, 2.01)] and 2.38 times (CI: 1.56, 3.63) more likely to gamble weekly during lockdown compared to not gambling at all. The odds ratio for less than weekly gambling was 1.83 (1.30, 2.56), no different to that for weekly gambling (p = 0.31). The association between heavy alcohol intake and less than weekly gambling was only significant in males [2.67 (1.66, 4.30); females: 1.04 (0.61, 1.78)], whereas the association with weekly or more frequent gambling was evident in both sexes [males: 2.03 (1.15, 3.58); females: 2.42 (1.28, 4.61)].
Gambling and Personal Financial Situation
Those that struggled financially before COVID were more likely to answer yes to any gambling during lockdown (< weekly gambling and > = weekly gambling combined due to small numbers). Odds ratio for those reporting “just getting by” pre-COVID and any gambling was 1.56 (95% CI: 1.07, 2.26) and 2.23 (1.23, 4.06) for those reporting “finding it quite or very difficult”.
No relationship was apparent between gambling frequency and current employment status during lockdown (p = 0.63).