Wednesday February 23 2011
The reviewed studies included over 1 million participants
“Alcohol in moderation 'can help prevent heart disease',” reported BBC News. It said that a review of 30 years of research found a 14-25% reduction in heart disease in moderate drinkers compared with people who did not drink alcohol.
The news was based on a review that found that light to moderate alcohol consumption was associated with a reduced risk of cardiovascular outcomes, such as having a heart attack or dying from heart disease or a stroke. However, when people drank more than a moderate amount of alcohol, the relationship with cardiovascular outcomes became more complex, with the risk of a stroke increasing with higher alcohol consumption. The researchers noted that this supports the need for limits on alcohol consumption.
It is important to remember that increased alcohol consumption increases the risk of other outcomes that were not assessed by this review, such as alcohol-related injuries, liver disease and certain cancers. As such, the researchers call for robust studies that assess multiple outcomes at the same time and identify people in whom potential cardiovascular benefits of light to moderate alcohol consumption outweigh potential harms.
Where did the story come from?
The study was carried out by researchers from the University of Calgary in Canada and other research institutions in North America. Funding was provided by the Robert Wood Johnson Foundation, the Substance Abuse and Mental Health Services and the Administration Center for Substance Abuse Treatment. The study was published in the peer-reviewed British Medical Journal.
This story was reported by the Daily Mail, Daily Telegraph, Daily Express and BBC News. The newspapers generally reported the research on cardiovascular outcomes accurately. Some made the important point that although light to moderate alcohol consumption appears to have cardiovascular benefits, the risk of some cardiovascular outcomes increases with higher alcohol consumption.
What kind of research was this?
This was a systematic review and meta-analysis which looked at the effect of alcohol consumption on the risk of cardiovascular disease. It was accompanied by a second systematic review and meta-analysis looking at the effect of alcohol consumption on the biological markers that are linked to a person’s risk of coronary heart disease. The review of cardiovascular outcomes is the focus of this Behind the Headlines article.
A systematic review is the best way to identify and summarise everything that is currently known about a particular question. In any systematic review, the reliability of the pooled study results is affected by the reliability of the studies reviewed. In the current review, for example, the results would be affected by how well the pooled studies had taken into account factors other than alcohol consumption that could affect the results (confounding factors). This review did look at these and other issues relating to the quality and methods of the studies to assess how they might affect the results.
What did the research involve?
The researchers searched scientific literature databases for relevant studies published up to 2009. They also looked for additional relevant studies mentioned in the articles they identified or which had been presented at conferences, and contacted experts in the field to ask for any other studies they were aware of.
They included only prospective cohort studies that looked at the risk of cardiovascular outcomes in adults with different levels of alcohol consumption. The participants in these studies did not have cardiovascular disease at the start of the study. The researchers included both published and unpublished studies.
The researchers’ main comparison was between people who reported drinking any alcohol and those who did not drink alcohol at the start of the study. They also classified the level of alcohol consumption by drinkers in each study, and whether the people who did not drink were lifetime abstainers or former drinkers. The main outcomes the researchers were interested in were death from cardiovascular disease, a first coronary heart disease event or diagnosis (based on fatal or non-fatal heart attack, angina, coronary heart disease or a surgical revascularisation procedure to treat disease of the coronary arteries), death from coronary heart disease, first stroke or death from a stroke.
The researchers used statistical methods to combine the data from these studies to assess the effect of alcohol consumption on cardiovascular outcomes. They also used standard techniques to look for variability between the studies’ results. They also carried out an assessment of whether there was any publication bias, where studies that had not shown a link between alcohol consumption and cardiovascular outcomes had not been published.
What were the basic results?
The researchers identified 84 studies that met their inclusion criteria. These studies included over 1 million adults. Just over half the studies looked at both men and women (44 studies), 34 studies looked at men only, and six studies looked at women only. The studies followed people for an average of 11 years (range 2.5 to 35 years). Most of the studies (68 studies) took into account factors other than alcohol consumption that could affect the results. Another eight studies adjusted for only a small number of demographic factors, and eight did not adjust for any potential confounders.
When pooling the results of the studies, the review found that, compared with people who did not drink, those who drank alcohol had a lower risk of death from cardiovascular disease (such as heart disease or stroke), coronary heart disease (CHD) events or death from CHD. However, whether a person drank alcohol did not appear to have any effect on the risk of having a stroke or dying from a stroke.
The researchers got similar results from analyses that pooled studies that adjusted for more potential confounders and those that pooled studies which carried out less adjustment. They also found similar results when they compared people who drank alcohol with those who had never drunk alcohol (that is, excluding those who used to drink alcohol but had given up).
When the researchers looked at the amount of alcohol people drank, they found that consuming up to one drink a day (2.5-14.9 grams of alcohol, one unit of alcohol being equivalent to 8g) reduced the risk of the cardiovascular outcomes assessed by 14-25% compared to no alcohol intake. The effects of higher amounts of alcohol consumption varied across the outcomes. Higher alcohol intake was still associated with a reduced risk of coronary heart disease death, but was associated with either no benefit or an increased risk of other outcomes. For example, drinking above 60 grams of alcohol a day was associated with an increased risk of having a stroke.
How did the researchers interpret the results?
The researchers concluded that “light to moderate alcohol consumption is associated with a reduced risk of multiple cardiovascular outcomes.”
This thorough systematic review used accepted methods and included a large number of studies, with over 1 million participants. There are some inherent limitations in pooling observational studies. For example, the studies that were included varied in how well they took into account potential confounders, but analyses suggested that this did not have a large effect on the results. In addition, although alcohol consumption was assessed at the start of the included studies, it is likely that many people’s consumption changed over time. The review did not look at how a person’s pattern of drinking (such as binge drinking versus non-binge drinking) affected cardiovascular risk.
Importantly, observational studies such as those pooled in this review cannot by themselves prove causation. Another way of assessing whether an exposure causes an outcome is to determine whether such a causal link would be biologically plausible. To assess this, the researchers also carried out a companion systematic review looking at whether alcohol consumption affected the levels of various biological markers associated with risk of cardiovascular disease.
This review found that moderate alcohol consumption was associated with favourable changes in certain markers that give an indication of a person’s cardiovascular risk, such as increased levels of "good" (HDL) cholesterol in the blood. This provides support for the suggestion that light to moderate alcohol consumption may itself lead to the reduction in cardiovascular risk seen.
However, as the authors note, when people drink more than a moderate amount of alcohol, the relationship with cardiovascular outcomes becomes complex, with the risk of some outcomes, such as a stroke, increasing with higher alcohol consumption. The researchers say that this finding supports the need for limits on consumption.
In addition, alcohol increases the risk of other outcomes that were not assessed by this review, such as alcohol-related injuries, liver disease and certain cancers. As such, the researchers call for robust studies that assess multiple outcomes at the same time, and identify those in whom potential cardiovascular benefits of light to moderate drinking outweigh potential harms.
Overall, this study supports the message from earlier systematic reviews that light to moderate alcohol consumption in adults is associated with improved cardiovascular outcomes. Currently, the UK government recommends a maximum daily intake of 2-3 units for women and 3-4 units for men, with one unit equivalent to 8g of alcohol (about half a pint of weak lager).