The National Highway Traffic Safety Administration (NHTSA) keeps track of deaths due to airbags; you can find a list of deaths on the NHTSA web site, along with conditions under which these deaths occurred. Each death occurred in a low speed collision, and for each, there is no other possible cause of death. Is it reasonable to assume that airbags can kill people only at low speeds? Isn't it more likely that airbags also kill people at higher speeds, but the death may be attributed to the crash? In this study we compare fatality rates for occupants with airbags available to fatality rates for occupants without airbags, controlling for possible confounding factors such as seatbelt use, impact speed, and direction of impact. What effect does the presence of an airbag have on the probability of death in a crash, under various conditions?
The main difference between our study and the previous studies is the choice of the dataset. We use the NASS CDS database, which is a stratified random sample of crashes nationwide. Previous studies showing beneficial effects of airbags have all used the FARS database, which contains data for all crashes in which a fatality occurred. We can limit our analyses to a subset of the NASS CDS database, choosing only crashes where there was at least one fatality; this should be a random sample of the FARS database. When we perform the analyses on this subsample, we can reproduce the results of previous studies: in accidents in which a fatality occurs, airbags are beneficial. However, for the entire random sample of crashes, they increase rather than decrease the probability of death.
Here is an analogy to help understand this: If you look at people who have cancer, radiation treatment will improve their probability of survival. However, radiation treatment is dangerous and can actually cause cancer. Making everyone in the country have airbags and measuring effectiveness only in the fatality group, is like making everyone have radiation treatment and looking only at the cancer group to check efficacy. Within the cancer group, radiation will be found to be effective, but there will be more deaths on the whole.
This is what seems to be happening with airbags. In a severe accident, airbags can save lives. However, they are inherently dangerous and pose a risk to the occupant. Our analyses show that in lower-speed crashes, the occupant is significantly more likely to die with an airbag than without. This effect is not seen in the analysis using FARS, because this database does not contain information about low-speed crashes without deaths.
Our analysis shows that previous estimates of airbag effectiveness are flawed, because a limited database was used. We have demonstrated this by reproducing their results, using a subsample of the CDS data. The new estimates of effectiveness suggest that airbags are not the lifesaving devices we have believed them to be.
The data used in the new analyses are all available to the public on the NHTSA website. All of the information about how to access the data, as well as the SAS code used to merge and concatenate datasets, can be found below. The dataset used in the study represents front-seat occupants aged 16 and older, in car crashes in years 1997-2002.
The dataset -- SAS transport file
The dataset corrected according to article by Charles M. Farmer -- SAS transport file
The NHTSA web sites
The SAS code used to compile the dataset
Example of the SUDAAN code and output represented in the paper