Kahn · Journal of burn care & research : official publication of the American Burn Association 2015 · retrospective registry analysis · n=2314

Line of duty firefighter fatalities: an evolving trend over time.

Cited 27 times in the scientific literature.

Level 4 - case-series / case-control

Retrospective database analysis comparing mortality trends across historical registry cohorts

PubMed 25055007 · doi:10.1097/BCR.0000000000000104 · record verified 2026-08-30

What was done

The authors queried the U.S. Fire Administration fatality database for all on-duty firefighter deaths across two time periods: 1990 to 2000 and 2002 to 2012. The year 2001 was excluded to avoid distortion from the 347 deaths on September 11. Data extracted included age category (exact age was unavailable), duty type, incident type, and cause of death. Proportions were compared between decades using chi-square tests.

What was found

A total of 2,314 deaths were analyzed (1,140 in 1990–2000; 1,174 in 2002–2012). Significant proportional increases occurred for myocardial infarction (43.0% to 46.5%, P = .012), cerebrovascular accident (1.6% to 3.7%, P = .002), fatalities in firefighters older than 40 years (52% to 65%, P = .0001), and deaths occurring during training (7.3% to 11.2%, P = .00001). Significant decreases were observed for asphyxiation (12.1% to 7.9%, P = .003), burns (7.7% to 3.9%, P = .0004), and electrocution (1.8% to 0.5%, P = .004). Deaths from trauma remained unchanged (27.8% vs. 29.6%, P = .12), and female deaths remained constant at 3%.

Why it matters

As protective equipment and safety practices have reduced direct fireground hazards like burns and asphyxiation, cardiovascular disease in an aging workforce and training safety have emerged as the primary targets for line-of-duty mortality prevention.

Limits

The study relies on retrospective administrative registry data without granular clinical or autopsy details. Exact ages, baseline cardiovascular risk profiles, and total active firefighter denominators (precluding calculation of true incidence rates) were not reported in the abstract. The year 2001 was omitted entirely.

Cited by