Article Text


Carbon monoxide poisoning
To make further progress against carbon monoxide poisoning, focus on motor vehicles
  1. B J Strife1,
  2. L Paulozzi2
  1. 1National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Georgia
  2. 2Division of Violence Prevention, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Georgia
  1. Correspondence to:
 Dr L Paulozzi
 Division of Violence Prevention, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 4770 Buford Highway, NE, Mailstop K60, Atlanta, GA 30341, USA;

Statistics from

Modification of motor vehicles may help prevent deaths from carbon monoxide

Motor vehicle crashes are well publicized in the US, where they kill over 40 000 people each year. It is less well known that motor vehicles can also be lethal while standing still. In most such cases, the motor vehicle may be a convenient way to commit suicide. In other cases the death is wholly unintended. Modification of motor vehicles may help prevent some of these deaths.

The best known measure taken to date to address risks from motor vehicle exhaust was intended to reduce chronic exposure to air pollutants rather than acute poisonings. The US introduced catalytic converters in 1975 to improve ambient air quality by reducing the amount of carbon monoxide (CO) and other pollutants emitted by the average automobile. This change apparently also reduced deaths from acute motor vehicle carbon monoxide (MVCO) poisoning, both intentional and unintentional.1

Catalytic converters, however, do not remove all CO from exhaust. Motor vehicles emit toxic levels of CO when their emission controls malfunction or after cold starts,2 and the engine can be “fooled” into putting out more CO when the vehicle is running in a tightly enclosed space.3 Emission controls therefore have not eliminated the MVCO poisoning problem, even though essentially all American cars now have them. In 1998, the most recent year for which deaths traceable to MVCO are available in the US,4 the US motor vehicle fleet caused 1561 MVCO deaths, which accounted for 72% of the 2160 CO deaths from all sources combined (table 1). MVCO deaths are also a current problem in other countries with similar automobile emission controls. For example, there were 294 MVCO deaths in the UK in 2000,5 534 in Australia in 1998,6 and 276 in Canada in 1999 (Statistics Canada, personal communication, 10 July 2003).

In addition to deaths from MVCO, there are probably many thousands of non-fatal MVCO poisonings. One study has estimated the total number of emergency department visits for CO poisoning in the US annually as 43 000.7 The proportion of these visits caused by motor vehicle exhaust is not known, but it is probably a large share given the contribution of motor vehicles to overall CO mortality. Most people who attempt suicide do not seek care when their attempt is unsuccessful,8 so emergency department visits probably represent an underestimate of at least the self inflicted morbidity associated with MVCO.

Over 90% of people who kill themselves with motor vehicle exhaust are found inside the passenger cabin of a car or truck.9 Similarly, in an unpublished review of 58 non-fatal emergency department visits due to MVCO captured in an emergency department surveillance system in the US, described elsewhere,10 we found that 71% of the victims were poisoned while inside a motor vehicle, and an additional 23% were poisoned in an enclosed space such as a garage housing a running motor vehicle. Very few incidents involved scenarios where the exhaust had been channeled into an enclosed space that did not contain the motor vehicle.

Given these circumstances, putting carbon monoxide detectors like those used in private homes inside cars would seem like a logical precaution.11 However, the literature shows that alcohol consumption is a consistent contributing factor in fatal CO poisoning,7,11 and a detector that sounds an alarm when levels of CO are dangerously high may have limited effect when a person is inebriated. And it will probably have no effect when that person intends to die and has taken precautions against discovery.

A better approach might be a CO detector in motor vehicles that sounds an alarm and then slows the engine to a stop, allowing the accumulated CO to dissipate. Such an alarm should preferentially measure the cabin air rather than the air outside the motor vehicle. It should be difficult to tamper with and inexpensive to install. Such devices were actually recommended by the US Centers for Disease Control in 1990 to prevent adolescent MVCO suicides.12 Technical details for such detectors for automobile passenger compartments were sketched out in a 1991 report from the Carnegie-Mellon Research Institute to the US National Highway Traffic Safety Administration (NHTSA),13 and results of field tests with such devices have been published more recently.14 The Carnegie-Mellon report recommended the development of such devices and calculated that they could be mass produced for less than $12 apiece. If this estimate is correct, the cost of this technology is lower than that of some other required automobile safety features that are designed to prevent smaller numbers of deaths. Tire pressure monitoring systems, for example, cost approximately $48 per vehicle.15 NHTSA has recently required such systems for American automobiles. NHTSA estimated that such systems could prevent approximately 124 deaths annually,15 many fewer than the 1500 MVCO deaths that might be prevented by MVCO detectors.

Depriving suicidal people of the tools they need or prefer to use to end their lives, an approach known as “means reduction”, is not foolproof, because suicidal people may simply substitute an alternative means to kill themselves. As a result, it is argued, the MVCO suicide rate may decline, but the total suicide rate may remain the same. However, survivors of deliberate MVCO poisoning report that they had acted impulsively—two thirds reported spending less than 24 hours planning the act—and chose the method in part because of its availability.16 All such persons may not turn to alternative means once the impulse to kill themselves has passed. This may be why the total suicide rate in Britain declined after the substitution of natural gas for coal gas, which has high concentrations of CO.17 It may also be why the introduction of catalytic converters, at least in Britain, has been given credit for at least some of the decline in total suicide rates for men since that change.18 Similar reductions in total suicide mortality have been correlated with efforts to control the availability of the pesticide paraquat in Samoa.19 So MVCO detectors may in fact prevent some suicides. In any case, the use of such technology may be justifiable if only because it may help prevent the 190 annual unintentional deaths from MVCO in the US.

Making modifications to motor vehicles such as CO monitors will require collaboration with automobile manufacturers. Manufacturers may be willing to work with public health agencies to come up with a low cost solution to this problem. They may in fact be able to suggest solutions other than CO monitors. We therefore urge that government agencies in any countries that experience MVCO fatalities work with domestic automobile manufacturers to address this problem.

Table 1

Deaths due to carbon monoxide (CO) poisoning by source of CO and manner of death, US, 1998


This research was supported in part by Mr Strife’s appointment to the Research Participation Program at the US Centers for Disease Control and Prevention, National Center for Injury Prevention and Control administered by the Oak Ridge Institute for Science and Education through an inter-agency agreement between the US Depart-ment of Energy and CDC.

Modification of motor vehicles may help prevent deaths from carbon monoxide


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.