Residents of States Reporting Most West Nile Virus Cases are Less Likely to Take Precautions Against Mosquitoes
For immediate release: September 15, 2004
BOSTON, MA — A new national study conducted by the Harvard School of Public Health Project on the Public and Biological Security finds that California and Arizona residents are significantly less likely than Americans nationwide to have taken any precautions against mosquito bites. Nearly two-thirds of California (64%) and Arizona (65%) residents report not having taken any precautions against mosquito bites since the beginning of June, as compared with 45% of adults nationally (see links below) who did not. More than half of this year’s human cases of West Nile virus have been reported in California and Arizona, including 17 deaths in those states as of September 14.
In spite of public health campaigns and commercial advertising encouraging the use of DEET-based mosquito repellents, such as the brands Off and Cutter, many Americans do not know what DEET is or have health concerns about using it. Only about half (49%) of Americans—including 41% of California and 47% of Arizona residents—report that they know what DEET is. One in five Americans (20%)—including 16% of Californians and 12% of Arizonans—are concerned that using DEET could be dangerous to their health. The Centers for Disease Control and Prevention (CDC) recommend using mosquito sprays containing DEET.
A majority of Americans report that they have not used a DEET-based mosquito repellent since the beginning of June, when West Nile infections start to become prevalent. The proportion is even larger in California and Arizona, where more than three-fourths of residents report that they have not used DEET-based repellents during that same period.
“This study shows there is much work to be done in educating the public about the importance of protecting themselves from the West Nile virus,” said Robert Blendon, professor of Health Policy and Political Analysis at the Harvard School of Public Health. “If public health authorities cannot convince people in these states to start taking more precautions against mosquito bites, we are likely to see an increase in the number of cases during the remainder of the mosquito season.”
The study’s findings were based on interviews with 1,014 Americans nationwide and 1,143 residents of California, Arizona, and Colorado. So far in 2004, 1,386 human cases of illness from West Nile virus, resulting in 35 deaths nationwide, have been reported (as of September 14, 2004). There have been 392 cases and 12 deaths in California, 345 cases and five deaths in Arizona. Colorado has had the next highest number of reported cases (225 cases) and two deaths. West Nile is a virus that is mainly transmitted by mosquito bites.
Precautions Against Mosquito Bites in California, Arizona, and Colorado
In the three states where the most human cases of illness from West Nile virus have been reported this year, people who see themselves as living in high-mosquito areas in the state are about twice as likely as those who see themselves as living in low-mosquito areas (61% to 31%) to have taken precautions against mosquito bites. But many of this year’s cases have been found in what people typically call “low-mosquito areas,” where significantly fewer people have been taking precautions. Only about one-fourth (24%) of residents of these three states see themselves as living in high-mosquito areas, compared with almost half (48%) of Americans nationwide.
In California and Arizona, only one-fourth or fewer report having taken each of four specific CDC-recommended precautions since June 1. About one in four report having removed standing water from spare tires, gutters, bird baths, kiddie pools, and other places where water collects (24% in California, 26% in Arizona). About one in five report having used a mosquito repellent containing DEET (including the brands Off and Cutter) (19% in California, 20% in Arizona) and having avoided going outdoors at the peak mosquito hours of dawn and dusk (19% in California, 16% in Arizona). One in five Californians (21%) and one in eight Arizonans (12%) report having worn long-sleeve shirts and other protective clothing outdoors to protect against mosquito bites (Table 1).
In 2003, Colorado experienced the largest number of human cases (2947) and deaths (63) from West Nile virus, and this year the state has the third highest number of reported human cases. Compared with the country as a whole, Colorado is about average in the proportion of people taking precautions. About three in five Colorado residents (58%) report having taken any precautions against mosquito bites since the beginning of June. Two in five (40%) report having used DEET during that period. Three in five (60%) say that they know what DEET is, and about one in four (24%) are concerned that using DEET could be dangerous to their health.
For information about preventive measures, go to the CDC website: www.cdc.gov/ncidod/dvbid/westnile/index.htm
Attitudes About Spraying Against Mosquitoes
On the sometimes controversial issue of special spraying against mosquitoes in order to control the spread of the West Nile virus, the study found strong support for the practice. Nationwide, large majorities of those who live in areas where there has been spraying at ground level (90%) or from the air (77%) approve of these practices. Even in areas where there has not been any ground spraying, three-fourths (76%) of Americans said they would favor such spraying to prevent the spread of West Nile virus if it appeared in their area. In areas where there has not been spraying from the air, support for such spraying is lower, but a majority (58%) would still approve of air spraying for this purpose. The levels of public approval for these practices are similar in the three states where the most cases of illness from West Nile virus have been reported.
For the complete survey and graphics:
U.S. Survey: http://www.hsph.harvard.edu/press/releases/blendon/toplinenationalrelease.pdf
State Survey: http://www.hsph.harvard.edu/press/releases/blendon/toplinestatesrelease.pdf
State Map graphic: http://www.hsph.harvard.edu/press/releases/blendon/graphic913.gif
Minorities in the Three High-Incidence States
In the three states with the largest number of reported human cases of illness from West Nile virus, whites (40%), African-Americans (39%), and Latinos (40%) are about equally likely to report having taken precautions against mosquito bites since the beginning of June. However, Latinos (17%) are significantly less likely than whites (26%) to report having used a mosquito repellent containing DEET.
This is the 17th in a series of studies by the Harvard School of Public Health Project on the Public and Biological Security. The project was funded through the federally-supported Center for Public Health Preparedness at HSPH.
The study, which involved two parallel surveys, was designed and analyzed by researchers at the Harvard School of Public Health. The project director is Robert J. Blendon of the Harvard School of Public Health. The research team also includes John M. Benson, Catherine M. DesRoches, and Kathleen Weldon of the Harvard School of Public Health and Melissa J. Herrmann of ICR/International Communications Research.
Fieldwork for each survey was conducted via telephone for the Project by ICR/International Communications Research of Media (PA) between August 25 and August 29, 2004 (national survey) and between August 25 and August 31, 2004 (three- state survey). The national survey was conducted with a nationally representative sample of 1,014 adults age 18 and over. The three-state survey was conducted with a representative sample of adults in California (845), Arizona (145), and Colorado (153), including 203 Latinos and 100 African-Americans. The margin of error for the total samples in each survey is plus or minus 3 percentage points; for residents of California, plus or minus 3.5 percentage points; for residents of Arizona and Colorado, plus or minus 8 percentage points.
Possible sources of nonsampling error include nonresponse bias, as well as question wording and ordering effects. Nonresponse in telephone surveys produces some known biases in survey-derived estimates because participation tends to vary for different subgroups of the population. To compensate for these known biases, sample data are weighted to the most recent Census data available from the Current Population Survey for gender, age, race, education, as well as number of adults and number of telephone lines in the household. Other techniques, including random-digit dialing, replicate subsamples, callbacks staggered over times of day and days of the week, and systematic respondent selection within households, are used to ensure that the sample is representative.
For further information, please contact:
Office of Communications
Harvard School of Public Health
677 Huntington Avenue
Boston, MA 02115