Unintentional carbon monoxide poisoning cases in Ciudad Juarez, Mexico.

To the Editor: In the last 12 years, governmental reports indicate that 1,381 people living in Ciudad Juarez, Mexico have been intoxicated, and 167 killed by carbon monoxide (CO) poisoning. (1) Studies in the U.S. have identified low temperatures and small heating and cooking appliances as risk factors (2-4) but have not addressed northern Mexico. This study examines 91 incidents of carbon monoxide intoxication from October 2003 to February 2004 in Ciudad Juarez. Carbon monoxide intoxication incidents are associated with colder temperatures on both a monthly average and a daily basis. The three coldest nights, one with a minimum temperature of 13[degrees]F and two with minimum temperatures of 19[degrees]F, accounted for 22% of all the incidents. A neighborhood-level analysis suggests that households in the middle of the income spectrum are disproportionately at risk. Only 2% of households in neighborhoods with CO incidents earn less than the full-time minimum wage, compared with 3% of house-holds in other neighborhoods. At the other end of the income scale, 14% of households in neighborhoods with CO intoxications earn more than five times the minimum wage, while 19% do so in neighborhoods without reported CO incidents.

A comparison of risk factors between the case data and baseline data is shown in Table 1. The case reports are from neighborhoods throughout Ciudad Juarez, while the baseline data are from studies of a subset of three neighborhoods. (5) Thus, differences between the prevalence of risk factors may reflect differences between the three neighborhoods and the rest of Ciudad Juarez that are unrelated to CO poisoning. Nevertheless, to the best of the investigators' knowledge, the cases and baseline data come from similar populations, namely lower-income residents of Ciudad Juarez. Gas heaters were identified as the cause of 59% of the incidents (crude odds ratio compared with all households without gas heaters = 3.3, P < 0.001). The second largest category (18% of incidents) included solid-fuel appliances, such as braseros (improvised wood-burning heaters), drums, or grills. A comparison between gas heaters and solid-fuel heaters found that gas heaters were associated with higher risk than the solid-fuel heaters (odds ratio of 2.3, P = 0.009). This may be because gas heaters are usually unvented, while most of the wood heaters are vented. A second possible explanation of the excess risk associated with gas heaters is that wood heaters will tend to burn down if left overnight, while the gas heaters can burn throughout the night. The most common brands of heaters were El Sol and Solmatic, two companies that recently merged and produce heaters of similar design for the lower end of the market. Having an El Sol heater, compared with all other brands, is associated with an odds ratio of 3.9 (P < 0.001). However, several manufacturers have recently introduced new models with oxygen depletion sensors that are expected to be safer than the older models.

In 32% of the incidents the houses were reported to be hermeticamente cerradas (completely sealed) meaning that the doors and windows were covered with plastic, newspaper, or other material as an insulating measure. This practice appears to be unsafe (odds ratio of 3.2, P = 0.006). Nighttime use of heaters was also associated with a significantly elevated risk of intoxication (odds ratio of 3.9, P = 0.026). Incidents reported as occurring between 10 PM and 9 AM were classified as being associated with "nighttime use" (likely overnight use) of a heater. For baseline data, those households that reported more than 8 hours of heater use per night were classified as being in this "nighttime use" risk category.

A number of interventional strategies merit consideration, including efforts to encourage proper ventilation of homes and the use of carbon monoxide detectors and efforts to discourage the use of heaters while sleeping. While this study found differences in the relative risks for heaters produced by different manufacturers, these differences are likely not applicable to current models as the manufacturers have introduced new designs. Interventions are probably better directed at encouraging the replacement of older heaters that lack an oxygen depletion sensor with newer heaters that do have oxygen depletion sensors, rather than targeting particular manufacturers.

Patrick L. Gurian, PHD

Department of Civil, Architectural, and

Environmental Engineering

Drexel University 

Philadelphia, PA

Teresa Montoya, Veronica Corella-Barud, MPH

Center for Environmental Resource

Management

University of Texas at El Paso 

El Paso, TX

Zuber D. Mulla, PHD

University of Texas Health Science Center at Houston

School of Public Health El Paso Regional

Campus

El Paso, TX

References

1. Calentadores mortales: unmal que se repite. Diario de Juarez. January 4, 2004.

2. Cobb N, Etzel R. Unintentional carbon monoxide related deaths in the United States, 1979 through 1988. JAMA 1991;266:659-663.

3. Liu KS, Paz MK, Flessel P, et al. Unintentional carbon monoxide deaths in Calfornia from residential and other nonvehicular sources. Arch Environ Health 2000;55:375-381.

4. Ralston JD, Hampson, NB. Incidence of severe unintentional carbon monoxide poisoning differs across racial/ethnic categories. Public Health Rep 2000;115:46-51.

5. Graham J, Gurian P, Corella-Barud V, et al. Peri-urbanization and in-home environmental health risks: the side effects of planned and unplanned growth. Int J Hyg Environ Health 2004;207:447-454.

This project was supported by grant A-03-01 from the EPA's Southwest Consortium for Environmental Research and Policy, 5250 Campanile Drive, San Diego. California 92182-1913, Teresa Montoya was supported by a fellowship provided by the US EPA, Washington, DC. Analila Rojo and Jay Graham assisted with the preparation of the baseline databases. The assistance of Hector Apodaca and Cindy Conroy is gratefully acknowledged.

Table. Comparison of cases with baseline data from three neighborhoods

                                                 Number of     Cases
                                                 applicable    with risk
Risk factor            Referent group            cases         factor

Gas heater             No gas heater             91            59%
Gas heater             Solid-fuel heater         70            77%
El Sol heater          Other gas heater          54            48%
Gas stove for heating  No gas stove for heating  91             1%
Heater on at night     Heater off at night       91            16%
Windows/doors sealed   Not sealed                91            32%

                       Number of   Baseline
                       applicable  households  Crude   P
                       baseline    with risk   odds
Risk factor            households  factor      ratio

Gas heater             299          30%         3.3    <0.001
Gas heater             154          59%         2.3     0.009
El Sol heater           63          19%         3.9    <0.001
Gas stove for heating  299           8%         0.12    0.02
Heater on at night      63           5%         3.9     0.026
Windows/doors sealed    63          13%         3.2     0.006