Use of the capture-recapture method to estimate severe traumatic brain injury rates.

Objective: This study assessed the applicability of using three different data sources (hospital discharge data, Traumatic Brain Injury Registry data, and death certificates) to identify high-risk groups for traumatic brain injuries (TBI). Reporting biases were also addressed.

Methods: Linkage of the data sources and log-linear modelling in conjunction with the capture-recapture method was used to estimate the number of missing TBI. Biased reporting to each data source was assessed by using the total number of estimated TBI by age and sex.

Results: An estimated 2% of TBI that occurred in Iowa were not reported to any of the data sources. Overall, women 85 years old and older were more likely to be missed by the combined data sources. Males and those of advanced age were less likely to be reported to the Registry. By using the capture-recapture method, falls among the elderly were found to be a significant public health problem in Iowa.

Conclusion: Despite biased reporting to the three data sources, the capture-recapture method can be used to identify high-risk groups for TBI in Iowa.