Validation of self-reported motor-vehicle crash and related work leave in a multi-purpose prospective cohort
Pons-Villanueva J, Segui-Gomez M.
Validation studies about self-reported motor vehicle crashes and related work leave are scarce. The Seguimiento Universidad de Navarra (SUN) is a multi-purpose cohort study undertaken in Spain. Motor vehicle crash risk factors are assessed within it.
The objective of this study was to validate the incidents of self-reported motor vehicle crashes (MVC) and related work leave through mailing and clinical notes' review. The method included questions in the cohort's questionnaire regarding motor vehicle crash incident and work leave.
We made both a re-test and a criterion validity study for self-reported answers. The results show a moderate kappa Cohen's correlation coefficient for both events. Re-test reliability kappa for MVC was 0.55, for MVC-related work leave it was 0.53. Criterion validity kappa was 0.25 for MVC and 0.25 for MVC-related work leave. These results show a moderate agreement for re-test both for MVC and work MVC-related leave. For criterion validity, the results show a fair agreement.
The magnitude of the agreement is similar to other similar studies and allows the use of these data in epidemiological studies.