Daily time series for cardiovascular hospital admissions and previous day's air pollution in London, UK.

Abstract

OBJECTIVE: To test for a significant association between air pollution and emergency hospital admissions for circulatory diseases (international classification of diseases-9 390-459) in London, England, that would be consistent with a causal effect of pollution on the previous day. METHODS: Long term concurrent trends, temperature, humidity, day of the week, influenza epidemic of 1989, and cyclical covariations with periodicity > 20 days in daily measures of pollution and admissions for 1987-94 were allowed for. RESULTS: There were 373556 admissions. No association was found between O3 and circulatory diseases. Four other pollutants were associated with acute myocardial infarction and circulatory diseases combined. P values and attributable cases (95% confidence intervals) for acute myocardial infarction were: black smoke P = 0.003, 2.5% (0.8% to 4.3%); NO2 P = 0.002, 2.7% (0.8% to 4.6%); CO P = 0.001, 2.1% (0.7% to 3.5%); and SO2 P = 0.0006, 1.7% (0.7% to 2.6%). There were also associations between black smoke and angina (P = 0.02), NO2 and arrhythmia (P = 0.04), and CO and other circulatory diseases (P = 0.004), but none with heart failure. Acute myocardial infarction was the only diagnosis for which there were significant associations with and without adjustment for cyclical terms. The associations with acute myocardial infarction were significant only in the cool season. CONCLUSION: Population data were consistent with 1 in 50 heart attacks currently presenting at London hospitals being triggered by outdoor air pollution. Further research is now needed to investigate whether background concentrations of black smoke, NO2, CO, and SO2 are a preventable cause of myocardial infarction. These results, if applied to all myocardial infarctions in the United Kingdom, indicate a potential saving of 6000 heart attacks a year.