Improving standards for case-control studies

Between 2007 and 2010, six case-control studies were conducted within the European research-project DRUID to estimate the relative risk of serious injury for psychoactive substance use. Guidelines and study protocols were prepared for the DRUID-case-control studies according to the ICADTS guidelines for epidemiological studies to stimulate that the outcomes of these case-control studies were comparable. Furthermore, equivalent cut-offs were applied to adjust for differences between studies that collected information on recent substance use by means of oral fluid samples and by studies that used blood samples. Despite the high comparability of the study designs the results still showed large variations between the calculated odds ratios. It is assumed that these differences were likely to be caused by several kinds of random and systematic errors. To investigate this assumption an in-depth study on eleven indicators of potential study errors was conducted on the results of the six DRUID case-control studies on injury risk. The most commonly detected types of errors were selection bias and lack of sufficient study power due to small sample size. These differences seem to explain the majority of the variance between the calculated odds ratios. In order to avoid bias and confounding due to errors, future guidelines are recommended to more systematically include an overview of the sources of potential bias and instructions of how to avoid them. Furthermore, a-priori assessments on potential bias could reduce the effect of random and specific errors. By increasing the comparability of study designs and decreasing the potential errors of case-control studies a good estimate of the risk of driving under the influence of psychoactive substances might be available in future.

• Record URL:
  • http://www.icadtsinternational.com/documents/?category=20th_T2013_Brisbane
• Authors:
  • Houwing, S
• Conference:
  • International Conference on Alcohol, Drugs and Traffic Safety (T2013), 20th, 2013, Brisbane, Queensland, Australia
• Publication Date: 2013-8

Language

• English

Media Info

• Pagination: 339-44
• Monograph Title: 20th International Conference on Alcohol, Drugs and Traffic Safety Conference Proceedings, 25-28 August 2013, Brisbane Convention and Exhibition Centre, Brisbane, Australia

Subject/Index Terms

• TRT Terms: Case studies; Drivers; Drugged drivers; Drugs; Drunk driving; Errors; Intoxication; Research; Research projects; Sampling; Statistical sampling
• Uncontrolled Terms: Safe systems (road users)
• Geographic Terms: Europe
• ATRI Terms: Case control study; Drink driving; Drug driving; Error; Improvement; Research method; Sampling
• Subject Areas: Research; I83: Accidents and the Human Factor;

Filing Info

• Accession Number: 01495682
• Record Type: Publication
• Source Agency: ARRB
• Files: ITRD, ATRI
• Created Date: Oct 17 2013 10:12AM