Assessing the Probability That a Positive Report is False: An Approach for Molecular Epidemiology Studies

Abstract

Too many reports of associations between genetic variants and common cancer sites and other complex diseases are false positives. A major reason for this unfortunate situation is the strategy of declaring statistical significance based on a P value alone, particularly, any P value below.05. The false positive report probability (FPRP), the probability of no true association between a genetic variant and disease given a statistically significant finding, depends not only on the observed P value but also on both the prior probability that the association between the genetic variant and the disease is real and the statistical power of the test. In this commentary, we show how to assess the FPRP and how to use it to decide whether a finding is deserving of attention or "noteworthy." We show how this approach can lead to improvements in the design, analysis, and interpretation of molecular epidemiology studies. Our proposal can help investigators, editors, and readers of research articles to protect themselves from overinterpreting statistically significant findings that are not likely to signify a true association. An FPRP-based criterion for deciding whether to call a finding noteworthy formalizes the process already used informally by investigators--that is, tempering enthusiasm for remarkable study findings with considerations of plausibility.

Genetic variation in five genes important in telomere biology and risk for breast cancer

“…POT1, TEP1, TERF1, TERF2 and TERT), and could be related to breast cancer risk based on the role suggested for telomere biology in this disease (Baykal et al, 2004;Wacholder et al, 2004;Savage et al, 2005). Although associations with less common SNPs are possible, our data indicate that common variation in these genes is unlikely to substantially affect overall breast cancer risk.…”

Polymorphisms of genes coding for insulin-like growth factor 1 and its major binding proteins, circulating levels of IGF-I and IGFBP-3 and breast cancer risk: results from the EPIC study

“…We estimated the false positive reporting probability (FPRP) for statistically significant observations based on the methods described by Wacholder et al (2004). Prior probability is likely to be influenced by the biological knowledge of the gene, the functional significance of the variants and the available epidemiological evidence.…”

“…The large number of statistical tests we have performed raises the issue of potential false positives. An alternative to applying a Bonferroni's correction, which is generally too conservative because of statistical dependence between tests for multiple SNPs that are in LD, is the use of a Bayesian approach, such as the recently introduced calculation of FPRP (Wacholder et al, 2004). Given the absence of previous functional or epidemiologic data on the IGF1 SNPs we found associated with breast cancer risk, we calculated FPRPs by using a prior probability of true association.…”

RGMA and IL21R show association with experimental inflammation and multiple sclerosis

“…Estimation of the false positive report probability We have estimated the probability that the association between MS and Eae30 and Eae31 is false by estimating the FPRP as suggested by Wacholder et al 63 The FPRP depends on the prior probability of association, the power of the study and the significance threshold:…”