Population-based case-control study of risk factors for unintentional mortality from carbon monoxide poisoning in Taiwan.

We carried out a population-based case-control study to identify people in Taiwan who are at increased risk of unintentional mortality from carbon monoxide (CO) poisoning (ICD-9 CM: 986 or E868). The study included all 439 deaths from unintentional CO poisoning registered in Taiwan's National Mortality Registry during 1997-2003, whereas 878 control subjects were randomly selected, with a control/case ratio of 2, from all deceased individuals from other causes during the same period. The annual mortality rate of CO poisoning significantly increased in Taiwan over the 7-yr period from 1.6 to 3.5 per 10(6) person-years. Thirty-six percent (n = 160) of the deaths occurred at home, and 21% (n = 93) were registered as in-hospital mortalities. A multivariate logistic regression analysis indicated that married people had a significantly reduced mortality odds ratio (MOR) of 0.50 (95% CI = 0.30-0.82) compared to single individuals. Additionally, residents of northern Taiwan (which is relatively urban) had a notably higher MOR of CO poisoning (MOR = 3.44, 95% CI = 1.40-8.44) than people residing in eastern Taiwan (which is relatively rural). Moreover, the MOR peaked in cold periods. A daily maximum temperature of < 18.4 degrees C was associated with a 2.15-fold increase in the MOR compared to a daily maximum temperature of > or = 27.1 degrees C. This study demonstrates an alarming increase in the unintentional death rate from CO poisoning in Taiwan between 1997 and 2003. Certain demographic and geographic characteristics were significant predictors for CO poisoning, suggesting a need for preventive strategies targeting these high-risk populations. Precautions should also be taken during periods of low temperatures.