Wednesday March 23 2011
The research did not look at 'dirty weekends' or having an affair
Having a ‘dirty weekend’ is bad for your heart, according to the Daily Express. The newspaper claims that amorous encounters increase the risk of a heart attack if you are unfit, while The Daily Telegraph says that the risk is doubled if you only have sex occasionally.
The news is based on an interesting and well-conducted review of several studies to examine how transient increases in physical or sexual activity might affect the risk of a heart attack or sudden heart-related death. Most of the studies included men aged over 60 who had suffered a heart attack. They assessed their levels of physical or sexual activity in the period immediately before the event and compared them with their normal activity levels over the preceding year. Transient increases in either activity were associated with an increased risk of heart attack, although raising each person’s episodic activity level by an hour was estimated to lead to only 2 to 3 additional heart attacks per 10,000 years worth of participant data.
Despite the newspapers’ greater focus on sexual activity, most of the studies included in this review related to physical activity. A noteworthy finding is that people who were used to regular physical activity were at lower risk of a heart attack when they transiently increased their level above the norm. This supports general health advice to carry out some form of exercise on a regular basis.
Where did the story come from?
This US study was carried out by researchers from Tufts Medical Center, Tufts University and Harvard School of Public Health. It was funded by the US National Center for Research Resources. The study was published in the peer-reviewed Journal of American Medical Association.
The study itself reported its findings well, but these have been slightly trivialised by some news sources, particularly the Daily Express and Daily Mail, which have related the study’s findings to the issue of “dirty weekends” and extra-marital affairs, which were not really the subjects of this research.
What kind of research was this?
This was a systematic review and meta-analysis, which aimed to combine the results of all relevant studies that had assessed how the risk of heart attacks and sudden cardiac death were affected by episodic physical and sexual activity.
A systematic review is the best way of identifying all research related to how a particular exposure (e.g. physical exertion) is related to an outcome (e.g. a heart attack). While this type of issue might typically be investigated through cohort studies this systematic review has looked at studies with a ‘case-crossover’ design. Case-crossover studies are normally used to investigate the effect of exposures that are transient or intermittent (such as physical activity), and where the effect of the exposure would be expected to occur immediately or abruptly (such as a heart attack).
In case-crossover studies, each participant is separately assessed as both a case and a control: researchers look at the participant’s exposures or behaviours (such as exercise) during the period immediately preceding an event (such as a heart attack), and also during a separate control or comparison period during which the participant did not experience the event of interest. The time surrounding the event is known as the ‘hazard’ period, which could be of variable time depending on the design of the individual study.
In studies of this design, one potential source of bias is that the person themselves (or a person close to them, such as a relative or partner) has to provide information on their exposure during these time periods. This could therefore give the potential for recall bias if the person’s answer was influenced or distorted by a belief that the exposure (i.e. increased physical or sexual activity) was the cause of their cardiac event.
What did the research involve?
The researchers searched medical databases and reference lists of retrieved papers to identify studies that used a case-crossover design to investigate the association between episodic physical or sexual activity and acute cardiac events (heart attack or sudden cardiac death). They assessed the quality of these identified studies, considering the ways that they had measured or determined the exposure.
The researchers also considered the diagnostic criteria that they used for the outcome of interest; the duration of the ‘hazard’ and ‘control’ time periods; and how they had measured or determined the frequency of exposure during the control periods (i.e. how they assessed normal or ‘habitual’ levels of physical or sexual activity).
The researchers pooled the results of the studies using standard methods to identify the increase in risk associated with episodic activity. They used statistical methods to take into account the differences between the study findings (heterogeneity). They also looked at whether a person’s habitual level of activity influenced the effect of episodic activity.
The researchers also used data from other large population-based studies to calculate the general population’s risk of cardiac events per 100 person-years of follow-up, and their risk of death from these events. They used these estimates and the figures from their meta-analysis to calculate the absolute increase in cardiac events that might be expected to be associated with episodic activity. The researchers looked at the risk of triggering an acute cardiac event with each one hour-per-week increase in sexual or physical activity beyond the person’s habitual level (or per one unit increase, depending on the methods of time assessment used in each study).
What were the basic results?
Fourteen studies met the eligibility criteria. Ten studies provided data on episodic physical activity, three on episodic sexual activity, and one study looked at both episodic exposures. Seven of the studies of physical activity, and all four of the sexual activity studies (including the study that looked at both physical and sexual activity) had enrolled people to the study on the basis that they had suffered a heart attack. In the majority of studies, people were aged over 60 years and most were male. In most studies, the control period of time, during which habitual activity was estimated, was the year preceding the cardiac event. A few other studies used briefer time periods, e.g. assessing activity in the 24 hours prior to the event, and comparing this with activity that occurred 24-48 hours prior to the event.
The combined results of the seven studies assessing the effect of episodic physical activity upon heart attack (including 5,503 people) found that episodic activity more than trebled the risk (RR 3.45, 95% CI 2.33 to 5.13). The four studies assessing the effect of episodic sexual activity upon heart attack (including 2,960 people) found that episodic sexual activity more than doubled the risk of heart attack (RR 2.70, 95% CI 1.48 to 4.91). Subgroup analysis found that those with higher habitual activity levels were less susceptible to the effects of an episodic increase in activity compared to those with lower habitual activity levels, in whom an episodic increase had greater association with heart attack.
Overall, because exposure to episodic physical or sexual activity is relatively infrequent in terms of a person’s total lifetime, and its effect on risk of a cardiac event is correspondingly transient, the researchers calculated that a one hour-per-week increase in an individual’s physical or sexual activity above their habitual level would only account for an extra 2 to 3 heart attacks per 10,000 person-years of follow-up.
How did the researchers interpret the results?
The researchers concluded that “acute cardiac events were significantly associated with episodic physical and sexual activity” but that the strength of this association was reduced with increases in the level of habitual activity that the person was used to. In plain terms, this means that if a person normally does little physical activity, a sudden increase in their activity levels is more likely to increase their risk of a heart attack than it would in a person who is physically active on a regular basis.
This is an interesting and well-conducted study which has used reliable methods to try and investigate how transient increases in physical or sexual activity beyond a person’s habitual level may be associated with risk of a heart attack or sudden cardiac death. It has been slightly trivialised by some newspapers, particularly the Daily Express, which has related all its findings to the issue of ‘dirty weekends’ which was not really the subject of this research.
There are a few issues to consider about this study and its implications:
The researchers acknowledge that there are potential limitations to case-crossover studies. They state that, to their knowledge, there are no standard criteria to assess the validity of this type of study.
Studies using this design have strengths in that they remove the effect of confounding from other medical and lifestyle factors by using a person as their own control. That said, there is still the possibility of confounding from factors that could vary over time within the individual (e.g. smoking or emotional stress). They also have the potential for bias through the methods used for case selection, how the researchers determine the hazard and control periods, and through relying on participants (or their relatives) to self-report their levels of exposure during these time periods.
The researchers were aware of these potential sources of bias. They tried to account for them by thoroughly assessing each study for quality, and looking at things such as whether the individual studies clearly defined their outcomes and hazard and control periods.
Risk of heart attack and cardiac death
Importantly, as the researchers also acknowledge, although they found an association between an episodic increase in physical or sexual activity and risk of heart attack. Both of these exposures are relatively infrequent (compared, for example, to an exposure that remains with the person constantly, such as blood pressure or diabetes).
The effects of these exposures have on the risk of a heart attack or other acute cardiac events are therefore also correspondingly transient. The researchers estimated that the exposures’ overall impact on an individual’s absolute risk of a cardiac event was found to be small, and would be expected to account for only 2 to 3 extra heart attacks in 10,000 person-years of follow-up.
The studies mainly included men who were in their 60s; the results may not apply to younger individuals or to women.
Despite the newspapers’ greater focus on the sexual aspects of this study, most of the studies included in the review were actually related to transient increases in physical activity above a person’s habitual level. A noteworthy finding is that people who were used to carrying out regular physical activity were at lower risk of a heart attack when they transiently increased their level above the norm, compared to the effects of a sudden burst of activity in a person who is not normally very active. The findings again support general health advice to carry out some form of exercise on a regular basis.