The National Head and Spinal Cord Injury Survey: assessment of some uncertainties affecting the findings.

The estimates from the National Head and Spinal Cord Injury Survey, like estimates from other studies, are subject to many uncertainties and must be viewed as conditional. This article provides a list of conditions that affect estimates from studies. For this survey, two structural limitations are discussed. The first is the exclusion from the survey of persons who were fatally injured and did not receive hospital inpatient care (for instance, individuals who were dead on arrival at the hospital). The second is the exclusion from the survey of persons who were injured before 1970. Also, uncertainties in the data of the survey due to nonresponse and variability between data collectors are discussed.