This article is included in Quince as a reminder to readers
of the excellent service provided by the National Electronic Library of Health which analyses stories from the newspapers and other media.
(see link below)
Several newspapers on 7 May 2002 reported that drinking tea
is associated with lower mortality after myocardial infarction. The newspaper
articles summarise a reasonably well-conducted research study [8] in which it
was found that self-reported tea consumption the year before acute myocardial
infarction was associated with lower mortality in the following 3-4 years.
The newspaper articles are in general accurate in their
reporting of the study. The Daily Mirror, however, mistakenly reports that 44%
of heavy tea drinkers died at follow-up, when, in fact, only 14% of this group
had died, with only 10% dying from cardiovascular causes.
The research was conducted by Kenneth Mukamal and colleagues
at Harvard University. They aimed to determine whether tea consumers have better
long-term survival after acute myocardial infarction.
A cohort design was used. The study used data from
participants enrolled in the Determinants of Myocardial Infarction Onset Study.
Data were collected from chart reviews and face-to-face interviews with 1900
hospitalised patients with confirmed acute myocardial infarction.
A median of 4 days after myocardial infarction, patients were
asked about their usual frequency of caffeinated tea consumption during the past
year. Based on the distribution of tea consumption within the population,
patients' tea consumption was categorised as none, moderate use (less than 14
cups per week), or heavy use (14 or more cups per week).
In December 1995, a median of 3.8 years after recruitment to
the study, the National Death Index was searched for the deaths of study
participants. Using death certificate information, patient deaths were
attributed to either cardiovascular or non-cardiovascular causes.
To determine whether tea consumers have better long-term
survival after acute myocardial infarction, the number of cardiovascular deaths
in the tea consumption group was examined. The number of cardiovascular deaths
was 141 (14%), 67 (11%) and 26 (10%) for non-tea drinkers, moderate consumers,
and heavy consumers, respectively. The comparison between tea consumption groups
revealed a significant inverse relationship, indicating that a higher level of
tea consumption was associated with lower mortality.
This was a reasonably well-conducted study, in which a large
sample of patients were assembled at a similar point in their disease
progression, groups were comparable on most important known confounding
variables, and where they differed there was adequate statistical adjustment for
the effects of these variables. In addition, data indicates a dose-response
relationship between tea consumption and post-infarction mortality and,
importantly, outcome assessors were unaware of each participant's level of tea
consumption.
However, this study is weakened by a number of limitations.
First, not all potential confounding factors were accounted for in the data
collection. For instance, no dietary information was collected for these
patients
and it
might be the case that tea drinkers eat a healthier diet, or lead healthier
lifestyles more generally, and that this difference accounted for differences in
post-infarction survival.
Second, since consumption of 14 or more cups of tea per week
was relatively uncommon, the study had limited power to define mortality
differences between levels of tea consumption.
Finally, it is difficult to determine to what extent the
primary factor of interest (tea consumption) was ascertained accurately.
Self-report measures, in general, are prone to biased or inaccurate recall of
one's own behaviour, and it is unclear therefore to what extent participants
were assigned to the appropriate group.
The results should be treated with some caution. There are no
systematic reviews on the subject
