Objectives: To describe the epidemiology of carbon monoxide (CO) poisoning in Jerusalem and identify risk factors for such poisoning.

Design: A retrospective descriptive analysis of patients with CO poisoning who presented to the Hadassah hospitals in Jerusalem from 1994 to 2006.

Patients: All patients with suspected CO poisoning were examined and those with confirmed cases [carboxyhemoglobin (COHb) level >5%] were included. Sources of exposure, seasonal variation, and demographic characteristics were analyzed.

Results: There were 292 patients (49% males) with 40 family clusters that accounted for 149 patients (51%); 230 patients (79%) presented during the winter months. All but one had unintentional CO intoxication. The main sources of exposure were faulty gas heaters (n = 135), fire (n = 102), and other residential heating systems (n = 40). The estimated annual incidence of CO poisoning decreased from 6.45 per 100,000 in 1994-2000 to 3.53 per 100,000 in 2001-2006. High-risk intoxication (COHb level >25%) occurred in 84 patients (29%). Factors associated with severe intoxication were male gender, individual patients (compared with those in clusters), and faulty gas heaters (compared with other sources).

Conclusions: Males exposed to CO may have a more severe intoxication. The lower risk in patients presenting in clusters could be explained by the assumption that severe presentation in one patient alerts the others who are less severely affected. The implementation of safer standards for residential heating systems and CO detectors together with the public education may explain the decline in the incidence of CO poisoning.