 |
Current Trends
Comparison of Observed and Self-Reported
Seat Belt Use Rates -- United States
To measure compliance with seat belt use laws, most states have
estimated belt use by direct observation of vehicle occupants. In
addition, since 1984, several states have recorded seat belt use
data
as part of the Behavioral Risk Factor Surveillance System (BRFSS)
telephone survey (1-3). Previous studies indicate that telephone
surveys usually report higher belt use than do observation surveys
conducted in similar areas at similar times (4,5). A systematic
comparison of self-reported belt use rates in 15 states* from the
1987
BRFSS with observed belt use rates in 1987 in the same states
follows.
The BRFSS telephone surveys used similar designs in each state.
A statistically valid random sample of all adults in each state was
obtained by random digit dialing. Each survey asked the same
questions
and classified the responses into the same five categories. Thus,
the
BRFSS surveys in each state can be considered replications of the
same
survey.
For the observation surveys, some states used probability sampling
techniques to select locations and times. These surveys produced
statistically valid estimates of the actual belt use rates under
the
conditions surveyed.** Other states used locations and times
selected
by judgment. The accuracy of the estimates from these surveys is
unknown.
In the BRFSS self-reported surveys, the number of affirmative
answers
was derived in two ways: as the total number of respondents who
reported "always" using seat belts and as the sum of those who
reported
"always" and "nearly always" using them. The average self-reported
"always" use exceeded observed use by about 8% and ranged from 11%
below observed use to 24% above. The average "always or nearly
always"
self-reported use exceeded observed use by 27%, with a range of 12%
above observed use to 39% above. To further examine the
relationship
between observed and reported seat belt use, simple linear
regressions
were used for each state (Figures 1 and 2). The relation is
described
moderately well by either regression; approximately 54% of the
variation in prevalence of observed use was accounted for by the
prevalence of self-reported use. In the regression line for which
"always" was used as the definition, a 1 percentage point increase
in
self-reported use accounted for a 0.7 percentage point increase in
observed use. When "always" and "nearly always" were used, a 1
percentage point increase in self-reported belt use accounted for a
nearly 1 percentage point increase in observed use. However, these
figures are valid only within the range of the self-reported seat
belt
use data.
Reported by: Office of Driver and Pedestrian Research, National
Highway
Traffic Safety Administration. Div of Nutrition, Center for Health
Promotion and Education, CDC. Editorial Note
Editorial Note: Worldwide experience has demonstrated that seat
belt
use laws can substantially reduce deaths and injuries on highways.
More
than 30 foreign countries, 31 states, and the District of Columbia
now
have laws requiring adult drivers and passengers to use seat belts.
Direct observation surveys of seat belt use, if properly designed
and
conducted, can produce accurate estimates of use. However,
observation
surveys are expensive to conduct and usually observe only shoulder
belt
use. Furthermore, although they can estimate a driver's or
occupant's
age and sex, they cannot gather other information useful in
understanding belt use, such as trip purpose or attitudes about
belt
use laws. Telephone surveys provide the opportunity to collect
these
additional data. They may also be less expensive to design and
conduct
than observation surveys. However, telephone surveys can record
only
the respondents' stated behavior, not their actual behavior.
Some of the divergence in the data analyzed here may be due to the
fact that the self-reported data were collected each month
throughout
1987 and thus estimate average belt use throughout the year. The
observed data were at collected different times in each state.
Furthermore, the self-reported data were drawn from a sample of the
entire state while observed data from some states came from only a
few
sites. The moderate fit of the regression lines means that they are
useful in describing general relations between observed
and self-reported belt use, but they should not be used to predict
observed use in a single state when only one self-reported survey
is
available.
More studies such as these are needed to establish reliably the
relationship between the results from observation surveys and BRFSS
telephone surveys. The results from observation surveys could then
be
used to help interpret the BRFSS responses and translate them into
approximate actual belt use levels. The BRFSS data in turn could be
used to investigate characteristics of belt users and nonusers that
cannot be determined from observation surveys and to provide
information on temporal trends without the expense of observation
surveys. In these ways, the usefulness of both types of surveys
would
be enhanced.
References
CDC. Behavioral risk-factor surveillance--selected states,
1984.
MMWR 1986;35:253-4.
CDC. Behavioral risk-factor surveillance in selected
states--1985.
MMWR 1986;35:441-4.
CDC. Behavioral risk factor surveillance--selected states, 1986.
MMWR 1987;36:252-4.
Hunter WW, Stutts JC, Stewart JR, Rodgman EA. Overrepresentation
of
seat belt nonusers in traffic crashes. Chapel Hill, North Carolina:
University of North Carolina, Highway Research Center, 1988.
Wagenaar AC, Streff FM, Molnar LJ, Businski KL, Schultz RH.
Factors
related to nonuse of seat belts in Michigan. Washington, DC: US
Department of Transportation, 1987; DOT publication no.
(HS)807-217.
*California, Florida, Hawaii, Illinois, Indiana, Maryland,
Minnesota,
Missouri, New Mexico, New York, North Carolina, Ohio, Tennessee,
Utah,
and Washington.
**Most surveys took place during daylight hours and measured belt
use
by the driver and right front seat passenger.
Disclaimer
All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices. **Questions or messages regarding errors in formatting should be addressed to mmwrq@cdc.gov. Page converted: 08/05/98
|
|
