Under-reported drinking major public health concern
medwireNews: Researchers say that current under-reporting of alcohol consumption has major implications for public health and policy, as it leads to underestimates of harmful drinking and inaccurate government guidance.
In England, alcohol consumption reported in social surveys fails to account for around 40% of alcohol purchased according to revenue and customs records, say Sadie Boniface and Nicola Shelton, from University College London, UK. They adjusted survey data to account for this discrepancy assuming that all drinkers under-reported their alcohol consumption equally.
Data came from two surveys in 2008 including reports from 12,490 adults on weekly alcohol consumption, and from 9608 adults on their heaviest drinking day in the prior week.
After adjusting for under-reporting, the prevalence of exceeding weekly guidelines of 21 U in men and 14 U in women rose from 29% to 44% in men and from 20% to 31% in women. Exceeding daily limits (4 U in men; 3 U in women) rose by 19 percentage points in men to 75% and by 26 percentage points in women to 80%, and the prevalence of binge drinking (>8U in 1 day for men; >6U in 1 day for women) rose from 32% to 52% in men, and from 28% to 56% in women.
In the revised scenario, women became significantly more likely than men to exceed daily limits (odds ratio [OR]=1.37), while in the original survey they were less likely to (OR=0.89). And, while the original survey indicated that people living in the most deprived quintile were more likely to exceed daily limits than those in the least deprived quintile (OR=1.20), this finding lost significance in the revised scenario.
The scenario also brought women's binge drinking on a par with men, and showed that people in the two highest income quintiles were significantly more likely to binge drink than those in the lowest quintile (OR=1.31), which was not shown in the original survey.
Additionally, while the original survey results showed people living in the least deprived areas were most likely to binge drink, the revised analysis turned this on its head, with people in the most deprived quintile having a 24% greater odds for binge drinking than those in the least deprived quintile.
This contradictory finding could be an artifact, the researchers suggest, if the accuracy of consumption reporting varies by income or deprivation, or could be explained by the relatively high proportions of heavy drinkers living in deprived city centers.
The authors say their findings prompt the question of whether policymakers should use self-reported consumption data to inform guidelines.
"What's needed now is a detailed understanding of whether some people under-report their consumption more than others: to what extent does this vary between men and women for example, by how much someone drinks, or by what types of drink they prefer," said Boniface in a press statement.
"Little is known on this at present, but this could reveal groups who under-estimate their alcohol consumption substantially, illuminating areas where targeted alcohol education incentives should be developed."
medwireNews (www.medwirenews.com) is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2013
By Kirsty Oswald, medwireNews Reporter