II. Helmet Effectiveness Studies Based on Variants of Correlation Analysis Which Partially or Fully Contradict the Effectiveness of Helmets
1. Luna, G. K., Copass, M. K., Oreskovich, M. R., and Carrico, C. J. (1981). "The Role of Helmets in Reducing Head Injuries from Motorcycle Accidents: A Political or Medical Issue?" The Western Journal of Medicine, 135:89-92.
This study, discussed above, finds no statistically significant difference in the death rate and total severity of all injuries sustained by helmeted and non helmeted fatal accident victims. Once again , the statistical methodology used to derive these results is subject to the same criticisms discussed above. This result is substantiated (for accident victims and state fatality rates in Goldstein (1985 , 1986 , 1988). The statistical methodology employed in these latter studies is regression analysis and given the careful specification of the regression equation does not suffer from any type of statistical bias.
2. Hart, D. N. J., Cotter, P. W., and MacBeth, W. A. A. G. (1975). "Christchurch Traffic Trauma Survey: Part 2, Victims and Statistics," The New Zealand Medical Journal, 81:542, June.
This accident victim study (New Zealand) finds that: (1) non helmeted riders are over represented in motorcycle accidents; and (2) the use of crash helmets did not significantly reduce the severity and the number of head injuries -- the slight reduction in the number of head injuries for helmeted riders was not statistically significant. The methodology of this study is subject to the same criticisms discussed above. In addition, the result concerning no significant reduction in the severity of head injuries was not formally reported and thus is suspect. No other study tends to support the result that helmets do not prevent head injuries. The over representation result lends further support to the lack of controls for the more risky behavior of non helmeted found in all studies that rely on correlation analysis.
3. New York State, Department of Motor Vehicles. (1969). An Evaluation of Motor Vehicle Accidents Involving Motorcycles - Severity, Characteristics, Effects of Safety Regulation, Research Report No. 1969-12.
This before-after study finds: (1) a 39% decrease in the total number of motorcycle accidents; (2) a 40% reduction in the number of fatalities; and (3) a reduction in the number of fatal head injuries from 75.4% to 45.9% of all fatalities; and (4) an increase in the number of fatal neck injuries from 5.8% to 37.8% of all fatalities as a result of the helmet law enacted in 1967. Results (3) and (4) support the head-neck injury tradeoff result found in the regression analysis used by Goldstein (1986, 1988). While the results of the NYS Study are subject to the same criticisms of correlation studies advanced above, the neck injury result in Goldstein is based on a sound statistical specification that avoids potential sources of statistical bias.
4. Mackenzie , A. R. "Accident and Fatality Rate Data" unpublished.
Dr. MacKenzie uses standard fatality, accident, and registration data and shows that the average accident and fatality rates per million registrations is higher in helmet states than in repeal (non helmet) states for 1985 and for the nine years ending in 1985. He concludes that states without helmet laws are safer. These simple rate comparisons are subject to the criticisms of correlation analysis advanced above.
5. Dare, C. E., Owens, J. C., and Krane, S. (1979). "Effect of Motorcycle Safety Helmet Use on Injury Location and Severity: Before-and-After Helmet Law Repeal in Colorado."
As discussed above, this study finds that the rate of the most severe neck injuries increased slightly for helmeted riders. The statistical methodology employed is subject to the same criticisms advanced above.