Citation

Cash WS, Moss AJ. J. Saf. Res. 1973; 5(1): 12-25.

Copyright

(Copyright © 1973, U.S. National Safety Council, Publisher Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

A study was carried out in 1967 and 1968 to determine the best length of recall period to use in a survey for estimating the annual incidence of motor vehicle injuries. A special supplement to the Health Interview Survey questionnaire was admininstered to persons in a 3-county area of North Carolina who were known from data on official accident report forms to have been involved, and usually injured, in a motor vehicle accident. Comparison of answers received in the survey with data from the official report forms revealed the accuracy of respondents' answers as to whether or not they had been injured and the length of time since the accident had occurred. These data were analyzed in terms of length of recall period, classification of injury, and type of respondent (proxy or self). The technique used to determine the optimum recall period was the analysis of the relative root mean square error. The results of this study indicated that 3 months was the optimum recall period for the reporting of motor vehicle injuries and selected characteristics of the accident.