Objective: To investigate how cirrhosis-biased referral to liver clinics can explain the wide variation in progression rates for differently recruited cohorts and, in particular, for liver clinic cohorts compared to community-based studies of the natural history of hepatitis C virus (HCV).
Study design and setting: A simulation was designed to illustrate the sort of referral bias pattern that is capable of converting a 20-year progression rate to cirrhosis of around 5% in the community of HCV-infected individuals into a 20% progression rate for patients who have been selectively referred to a liver clinic.
Results: We show that event-biased recruitment, such as occurs if referral to liver clinics is increasingly likely the closer a patient is to cirrhosis, can produce severely upwardly biased estimates of progression rates, can dampen the influence of "poor prognostic" factors (such as history of excessive alcohol consumption), but overrepresents the proportion of patients in the community of HCV-infected individuals who have poor prognosis.
Conclusion: When attempting to establish the natural history of new diseases with long incubation periods, researchers should be on the look out for potential biases that result from the way patients are referred into clinical cohorts.