"Men with a bald pate are at significantly greater risk of heart disease than their less follicly challenged peers," reports The Daily Telegraph. There are similar headlines across much of the media.
The headlines refer to research into an association between baldness and coronary heart disease. The researchers estimate that the risk of developing coronary heart disease over 10 years or more is 32% higher in bald men compared with those with a full head of hair.
Men concerned by these headlines should not lose sleep – or any more hair – over it. There isn’t much men can do about balding, but there are plenty of ways to reduce risk of heart disease. The quote from the British Heart Foundation that appears in most media reports is particularly apt: “it’s more important to pay attention to your waist line than your hairline”.
This report was generally covered appropriately by the media, with most reporting that even the significant associations were modest compared with the increases seen with well established risk factors. However, the Daily Mail headline that “losing your hair before your 50s can almost double the risk of heart attack” is not an accurate reflection of the research. First, the studies examined more than just heart attacks and, second, the increase in risk for men under 55 or 60 ranged from 44% to 84%, which is not a doubling.
The Guardian should be commended for quoting an independent statistician, Professor David Spiegelhalter, who cautions readers about the difficulties of concluding that balding causes heart disease (or vice versa).
This was a meta-analysis of observational studies that estimated the association between male pattern baldness (or androgenetic alopecia) and coronary heart disease (CHD). The authors report that several studies have identified an association between these two factors, and this research sought to combine the results from multiple studies to establish the overall risk.
CHD is the most common cause of death among men in the UK. In CHD the arteries that pump blood to the heart become narrow. This narrowing occurs when fatty material builds up within the artery walls. If the arteries become too narrow, the heart does not receive enough oxygen-rich blood, which causes chest pain known as angina. This narrowing can escalate to the point that the artery is blocked, which can cause a heart attack, where the lack of oxygen-rich blood causes permanent damage to the heart.
As this was a meta-analysis of observational studies, it can only provide information on the association between baldness and CHD, and cannot say why they are associated or if one causes the other.
Studies were excluded from the review if they assessed a type of hair loss other than male pattern baldness. CHD was defined by the researchers as including coronary artery disease, myocardial infarction (heart attack), angina pectoris, cardiomyopathy and other types of ischaemic heart disease.
The researchers extracted and combined data from the identified studies, and came up with a pooled risk estimate. This provided an overall measure for the relative increase in risk of CHD among bald men, compared with men who were not balding. This analysis took into account (controlled for) some known risk factors for CHD, including:
The results of all of the analyses were reported as relative risks. This is unusual, because the results of case-control studies are usually reported as odds ratios. It is difficult to see from the scientific paper how the relative risks for these case-control studies were calculated.
The researchers did a subgroup analysis to assess the association between baldness and CHD among younger men (under 55 or 60 years old). They also conducted separate analysis among the subset of studies that reported on baldness severity (see box) to determine if the risk of CHD changed with the extent of balding.
The researchers identified 850 studies that potentially met their criteria for inclusion in the meta-analysis. After reviewing these studies, six remained that met all inclusion criteria. These six studies were conducted in the United States, Denmark and Croatia, and published between 1993 and 2008. They included a total of 36,990 participants. Three were cohort studies with a mean follow-up time ranging from 11 to 14 years.
Among the three cohort studies there was a significant increase in risk of CHD among men with severe baldness compared with men without baldness (relative risk (RR) 1.32, 95% confidence interval (CI) 1.08 to 1.63). When the analysis was restricted to men younger than 60 years old, a similar risk increase was seen (RR 1.44, 95% CI 1.11 to 1.86).
Among the three case-control studies, the authors report a significant association between baldness and CHD, both overall (RR 1.70, 95% CI 1.05 to 2.74) and among younger participants (1.84, 95% CI 1.30 to 2.62).
Three studies reportedly assessed the severity of baldness, two were case-control studies and one was a cohort study. This analysis found that the association between baldness and CHD varied according to baldness severity, and was only significant among men with mild to severe balding on the crown of the head, with no significant association seen among men with only frontal balding.
The researchers concluded that baldness on the crown of the head was significantly associated with an increased risk of CHD, and that this association increased with increasing baldness severity. However, just having a receding hairline presented no increase in risk. The researchers concluded that the known CHD risk factors should be carefully assessed among men with vertex baldness.
This meta-analysis suggests that there is an association between balding on the crown of the head and coronary heart disease, and that this holds true even among younger men. However, there are several limitations to the study that should be taken into consideration when interpreting this association.
On the one hand, the analysis included a large pooled sample size (with nearly 40,000 participants). However, only six studies in total were included, and each analysis included just three unique studies. The studies differed in the manner in which baldness was assessed, and the types of CHD included in each study. These variations in the way baldness and CHD were measured may make it difficult to compare the results of the different studies.
It’s also worth noting that cohort studies and case-control studies are often subject to bias and confounding, and that these problems remain when the results of several studies are combined.
It should also be noted that these studies can only assess the association between baldness and CHD, and cannot help explain what causes the association, although most of the press used the news to speculate that both may be caused by testosterone, diabetes or ‘chronic inflammation’.
Statistically speaking, this study is also limited because the pooled results for the case-control studies were reported as relative risks and not odds ratios (which is the more appropriate measure). It is not clear how the researchers converted from one measure to the other. Additionally, converting an odds ratio to a relative risk can overestimate risk if the outcome is common (in this review the outcome was CHD, which is relatively common).
Perhaps most importantly, baldness is not something you can do much about, even if the results of this meta-analysis are confirmed. Making lifestyle changes is a more effective way of reducing risk of CHD. These include: