The Use of Reference Priors and Bayes Factors in the Analysis of Clinical Trials

Abstract Background Until recently a typical rule that has often been used for the endorsement of new medications by the Food and Drug Administration has been the existence of at least two statistically significant clinical trials favoring the new medication. This rule has consequences for the true positive (endorsement of an effective treatment) and false positive rates (endorsement of an ineffective treatment). Methods In this paper, we compare true positive and false positive rates for different evaluation criteria through simulations that rely on (1) conventional p-values; (2) confidence intervals based on meta-analyses assuming fixed or random effects; and (3) Bayes factors. We varied threshold levels for statistical evidence, thresholds for what constitutes a clinically meaningful treatment effect, and number of trials conducted. Results Our results show that Bayes factors, meta-analytic confidence intervals, and p-values often have similar performance. Bayes factors may perform better when the number of trials conducted is high and when trials have small sample sizes and clinically meaningful effects are not small, particularly in fields where the number of non-zero effects is relatively large. Conclusions Thinking about realistic effect sizes in conjunction with desirable levels of statistical evidence, as well as quantifying statistical evidence with Bayes factors may help improve decision-making in some circumstances.

Download Full-text

True and False Positive Rates for Different Criteria of Evaluating Statistical Evidence from Clinical Trials

10.21203/rs.2.14574/v2 ◽

2019 ◽

Author(s):

Don van Ravenzwaaij ◽

John P A Ioannidis

Keyword(s):

Clinical Trials ◽

Confidence Intervals ◽

False Positive ◽

Evaluation Criteria ◽

Small Sample ◽

Statistical Evidence ◽

Bayes Factors ◽

True Positive ◽

P Values ◽

Meta Analyses

Abstract Background: Until recently a typical rule that has often been used for the endorsement of new medications by the Food and Drug Administration has been the existence of at least two statistically significant clinical trials favoring the new medication. This rule has consequences for the true positive (endorsement of an effective treatment) and false positive rates (endorsement of an ineffective treatment). Methods: In this paper, we compare true positive and false positive rates for different evaluation criteria through simulations that rely on (1) conventional p-values; (2) confidence intervals based on meta-analyses assuming fixed or random effects; and (3) Bayes factors. We varied threshold levels for statistical evidence, thresholds for what constitutes a clinically meaningful treatment effect, and number of trials conducted. Results: Our results show that Bayes factors, meta-analytic confidence intervals, and p-values often have similar performance. Bayes factors may perform better when the number of trials conducted is high and when trials have small sample sizes and clinically meaningful effects are not small, particularly in fields where the number of non-zero effects is relatively large. Conclusions: Thinking about realistic effect sizes in conjunction with desirable levels of statistical evidence, as well as quantifying statistical evidence with Bayes factors may help improve decision-making in some circumstances.

Download Full-text

True and False Positive Rates for Different Criteria of Evaluating Statistical Evidence from Clinical Trials

10.21203/rs.2.14574/v1 ◽

2019 ◽

Author(s):

Don van Ravenzwaaij ◽

John P A Ioannidis

Keyword(s):

Clinical Trials ◽

Confidence Intervals ◽

False Positive ◽

Evaluation Criteria ◽

Small Sample ◽

Statistical Evidence ◽

Bayes Factors ◽

True Positive ◽

P Values ◽

Meta Analyses

Abstract Background: Until recently a typical rule that has often been often used for the endorsement of new medications by the Food and Drug Administration has been the existence of at least two statistically significant clinical trials favoring the new medication. This rule has consequences for the true positive (endorsement of an effective treatment) and false positive (endorsement of an ineffective treatment) rates. Methods: In this paper, we compare true positive and false positive rates for different evaluation criteria through simulations that rely on (1) conventional p -values; (2) confidence intervals based on meta-analyses assuming fixed or random effects; and (3) Bayes factors. We varied threshold levels for statistical evidence, and thresholds for what constitutes a clinically meaningful treatment effect. Results: Our results show that Bayes factors, meta-analytic confidence intervals, and p-values often have similar performance. Bayes factors may perform better when trials have small sample sizes and clinically meaningful effects are not small, particularly in fields where the number of non-zero effects is relatively large. Conclusions: Thinking about realistic effect sizes in conjunction with desirable levels of statistical evidence, as well as quantifying statistical evidence with Bayes factors may help improve decision-making in some circumstances.

Download Full-text

True and False Positive Rates for Different Criteria of Evaluating Statistical Evidence from Clinical Trials

10.31222/osf.io/kcz3y ◽

2018 ◽

Author(s):

Don van Ravenzwaaij ◽

john Ioannidis

Keyword(s):

Clinical Trials ◽

Confidence Intervals ◽

False Positive ◽

Evaluation Criteria ◽

Small Sample ◽

Statistical Evidence ◽

Bayes Factors ◽

True Positive ◽

P Values ◽

Meta Analyses

Background: Until recently a typical rule that has often been used for the endorsement of new medications by the Food and Drug Administration has been the existence of at least two statistically significant clinical trials favoring the new medication. This rule has consequences for the true positive (endorsement of an effective treatment) and false positive rates (endorsement of an ineffective treatment). Methods: In this paper, we compare true positive and false positive rates for different evaluation criteria through simulations that rely on (1) conventional p-values; (2) confidence intervals based on meta-analyses assuming fixed or random effects; and (3) Bayes factors. We varied threshold levels for statistical evidence, thresholds for what constitutes a clinically meaningful treatment effect, and number of trials conducted. Results: Our results show that Bayes factors, meta-analytic confidence intervals, and p-values often have similar performance. Bayes factors may perform better when the number of trials conducted is high and when trials have small sample sizes and clinically meaningful effects are not small, particularly in fields where the number of non-zero effects is relatively large. Conclusions: Thinking about realistic effect sizes in conjunction with desirable levels of statistical evidence, as well as quantifying statistical evidence with Bayes factors may help improve decision-making in some circumstances.

Download Full-text

Localization of Gallium-67 in Peritoneal Cells by Electron Microscopic Autoradiography

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100072460 ◽

1973 ◽

Vol 31 ◽

pp. 404-405

Author(s):

D. C. Swartzendruber ◽

Norma L. Idoyaga-Vargas

Keyword(s):

Clinical Trials ◽

Soft Tissue ◽

Tumor Tissue ◽

Soft Tissue Tumors ◽

Clinical Usefulness ◽

Electron Microscopic ◽

Normal Cells ◽

Electron Microscopic Autoradiography ◽

Peritoneal Cells ◽

Gallium 67

The radionuclide gallium-67 (67Ga) localizes preferentially but not specifically in many human and experimental soft-tissue tumors. Because of this localization, 67Ga is used in clinical trials to detect humar. cancers by external scintiscanning methods. However, the fact that 67Ga does not localize specifically in tumors requires for its eventual clinical usefulness a fuller understanding of the mechanisms that control its deposition in both malignant and normal cells. We have previously reported that 67Ga localizes in lysosomal-like bodies, notably, although not exclusively, in macrophages of the spocytaneous AKR thymoma. Further studies on the uptake of 67Ga by macrophages are needed to determine whether there are factors related to malignancy that might alter the localization of 67Ga in these cells and thus provide clues to discovering the mechanism of 67Ga localization in tumor tissue.

Download Full-text