To the Editor:
A recent paper has shown that on the day of a major football match in the summer of 1996, cardiovascular mortality in the Netherlands was increased in men but not in women. 1 The interpretation given was that stress induced by watching a close game (which was decided eventually by penalty kicks) could be responsible for this increase. We have conducted time series analyses of the association between air pollution and mortality in the Netherlands for the period 1986–1994. 2–4 In this dataset, we tested the hypothesis, generated by Witte et al., 1 that football matches played by the Dutch national team in major tournaments lead to increased cardiovascular deaths.
Daily deaths were obtained from the Central Bureau of Statistics for the period 1986–1994. Daily nonaccidental mortality counts for the entire Netherlands were analyzed as total mortality, cardiovascular (International Classification of Diseases, 9th revision, 390–448), and mortality attributable to acute infarction and stroke as in the Witte et al. paper. 1 Median daily deaths were 328, 140, and 70, respectively. Separate mortality counts for men and women were not available. We focused on five games from all European and World championship matches played by the Dutch national team (see Table 1). The final game of the team in the tournament was selected, with the exception of the 1988 semifinal against archrival Germany. We selected a different analysis approach than did Witte et al., because we wanted to adjust for factors such as ambient temperature and air pollution. The relation between daily mortality and “events” was modeled with Poisson regression with generalized additive models (LOESS smoothing), using the data from all summer seasons. All event-mortality associations were adjusted for long-term and seasonal trend, influenza morbidity counts (based on a general practice sentinel system), ambient temperature (including separate lags for high and low temperature days), ambient humidity, black smoke air pollution, and indicators for day of week and holidays. 2–4
Because the football matches were played in the evening or late afternoon, we analyzed mortality of the same day as the match (lag 0) and mortality on the day after the match (lag 1) separately.
The results (Table 1) showed little or no association between these football matches and cardiovascular deaths. An analysis using the same procedures as Witte et al. 1 produced similar results with lag 0 relative risks of 1.08, 0.93, 0.78, 1.23, and 0.90 for the five games, respectively.
We found no evidence of increased total or cardiovascular mortality associated with five major football games played between 1988 and 1994 by the Dutch national team. Although, in this dataset, we were unable to analyze data separately for men and women, the power of our study was sufficient to detect a 10% increase in mortality attributable to acute ischemic attacks and stroke. The Witte et al. 1 study found a 50% increase in men, and a nonsubstantial 11% increase in women. That our analysis using the same methodology failed to show increased mortality after five major matches suggests that the original finding may have been a chance finding, or that the 1996 game against France featured peculiarities that were not shared by the games we analyzed. It should be noted that in our data, the highest odds ratio for lag 0 (1.16) was associated with a match lost to penalty kicks, as was the 1996 game against France; further inquiries should perhaps focus on such matches. However, in general, major football matches do not seem to lead to increased total or cardiovascular deaths in the population.
1. Witte DR, Bots ML, Hoes AW, Grobbee DE. Cardiovascular mortality in Dutch men during European football championship: longitudinal population study. BMJ
2. Hoek G, Brunekreef B, Verhoeff A, van Wijnen J, Fischer P. Daily mortality and Air pollution in the Netherlands. J Air Waste Manag Assoc 2000; 50: 1380–1389.
3. Brunekreef B, Hoek G, Fischer P, Spieksma FT. Relation between airborne pollen concentrations and daily cardiovascular and respiratory-disease mortality. Lancet 2000; 355: 1517–1518.
4. Hoek G, Brunekreef B, Fischer P, van Wijnen JH. The association between air pollution and heart failure, arrhythmia, embolism and thrombosis and other cardiovascular causes of death in a time series study. Epidemiology 2001; 12: 355–357.