Skip to main page content

• HOME
• CURRENT ISSUE
• PAST ISSUES
• MOBILE
• ABOUT
• ACP
• SUBSCRIPTIONS
• E-MAIL ALERTS
• SUBMIT

RSS Feeds

1. Factor V Leiden: a difference in cohort and case-control studies?

   • Jan P Vandenbroucke, MD, PhD
   • Frits R. Rosendaal

   Deptartment of Clinical Epidemiology, Leiden University Medical Center

   In a Danish cohort study, following 9523 individuals from 1976, venous thrombosis risk was increased 3-fold among carriers of factor V Leiden (216 persons developed venous thrombosis; 43 carriers)[1]. This relative risk was substantially lower than the 7-fold risk that we reported from a case-control study of 474 patients and 474 controls (92 and 14 carriers)[2]. The Danish authors ascribe the difference to "ascertainment bias" and a preponderance of risk factors among cases. This is a misapprehension of case-control studies: persons who develop disease always have more risk factors, in whatever study design [3]. As an example the Danish authors cite that in our study 66% of young women with venous thrombosis used oral contraceptives, while they found no users among their patients. This does not mean that something is wrong with case ascertainment in a case-control study, and certainly not that oral contraceptives play no role in venous thrombosis in Danish adults. Rather, this is a consequence of the Danish study base which consisted of an aging cohort of mainly middle aged and older persons. The one-time addition of 500 younger persons is of little help, given the low incidence of venous thrombosis. In contrast, we included cases from age 15 onwards as they occurred in the general population. Thus, a difference in relative risk of factor V Leiden may first of all reflect an age difference of the study base. Moreover, the inclusion of persons with a previous venous thrombosis, of persons with cancer and with a primary diagnosis of pulmonary embolism in the Danish study will all tend to lower the relative risk, as these conditions are less associated with factor V Leiden.

   Both estimates might be correct: ours for a first venous thrombosis of the legs in the general population free of cancer, and the Danish for all venous thromboembolism in a cohort of middle-aged and elderly persons. However, the Danish figures may well suffer from diagnostic misclassification: the study was initially prospective, but case detection was by routine administrative registries. Cases were verified afterwards, and had occurred over 23 years in many different hospitals and medical practices. Since the Danish case ascertainment amounts to a retrospective chart review - spanning two decades - a large degree of diagnostic uncertainty is likely. In contrast, in our case-control study cases were enrolled concurrently over a short period of time and only included when diagnosed by appropriate and recent instrumentation.

   Jan P Vandenbroucke, MD, PhD Frits R Rosendaal, MD, PhD Department of Clinical Epidemiology and Thrombosis and Haemostasis Research Center, Leiden University Medical Center PO Box 9600 2300 RC Leiden The Netherlands

   1. Juul K, Tybjaerg-Hansen A, Schnohr P, Nordestgaard BG. Factor V Leiden and the risk for venous thromboembolism in the adult Danish population. Ann Intern Med 2004;140:330-337.

   2. Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH. High risk of thrombosis in patients homozygous for factor V Leiden (activated protein C resistance). Blood 1995;85:1504-8.

   3. Rosendaal FR. Bridging case-control studies and randomized trials. Curr Control Trials Cardiovasc Med. 2001;2(3):109-110. Epub 2001 May 31.

   Conflict of Interest:

   None declared

   Submit response
   Published April 7, 2004