Clinical Calculator

menu

Bayesian interpretation of rates

This calculator is based on the work of Greenland (2006) (see section "Information-weighted averaging" of the 2006 article).

Examples

1. Childhood cancer and exposure to magnetic fields

The numbers prepopulated in the calculator form replicate the results of Savitz (1988) as calculated by Greenland (2006).

2. Bisoprolol for reducing perioperative adverse events

The DECREASE I (1999) trial reported a surprisingly strong relative risk reduction from using bisoprolol for the prevention of perioperative morbidity. The study was terminated early which may exaggerate the treatment effect (Goodman, 2007).
 
DECREASE I reported:
RR = 0.20 (95% confidence interval: 0.04 to 0.97)
Based on other published relative risks of observed relative risks of 0.67 in the DECREASE IV (2009) and 1.5 in the POISE (2008) trials, you estimate the likely range of the relative risk in a study to be:
0.25 to 4.00
Enter the numbers above into the calculator and you will get a Bayesian analysis of the DECREASE I results:
RR = 0.50 (credibility interval: 0.18 to 1.42)

References

  1. Greenland S. Bayesian perspectives for epidemiological research: I. Foundations and basic methods Int J Epi 2006 (PMID: 16446352)
  2. Savitz et al, Case-control study of childhood cancer and exposure to 60-Hz magnetic fields, Am J Epi 1988 (PMID: 3164167)
  3. Poldermans D, Boersma E, Bax JJ, et al. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group (DECREASE I). N. Engl. J. Med. 1999. (PMID 10588963).
  4. Goodman SN. Stopping at nothing? Some dilemmas of data monitoring in clinical trials. Ann Intern Med. 2007 (PMID: 17577008)