Tests for Equivalence Based on Odds Ratio for Matched-Pair Design

Liu, Jen‐pei; Fan, Hsin Yi; Chia, Mi

doi:10.1080/10543400500265561

Cited by 25 publications

(22 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al . 10 had previously also developed equivalence tests specifically for the odds ratio index. In fact, the five-region methods in this paper work for all indices, as long as a demarcation that respects the partitioning principle can be done.…”

Section: Discussionmentioning

confidence: 99%

A Five-Region Hypothesis Test for Exposure-Disease Associations

Shih

Lee

2017

Sci Rep

View full text Add to dashboard Cite

Characterizing exposure-disease associations is a central issue in epidemiology, one which epidemiologists often approach by adopting the index of the odds ratio and presenting its point estimate, p-value and confidence interval. In this study, the parameter space of the odds ratio is partitioned into five mutually exclusive regions corresponding to ‘strong protective factor’, ‘weak protective factor’, ‘no association’, ‘weak risk factor’, and ‘strong risk factor’, respectively. The authors presented a suite of statistical methods tailored to such a five-region demarcation, including methods for hypothesis testing, confidence interval estimation and calculation of the sample size needed to obtain the desired level of statistical power. The authors show that the five-region methods can efficiently and informatively describe a putative exposure-disease association, including its presence or absence, as well as its direction and strength (if any association exists). Three published results were re-analyzed to demonstrate the methods. R code is provided for convenience as well. The five-region methods are recommended for routine use during the analysis of epidemiologic data.

show abstract

Section: Discussionmentioning

confidence: 99%

A Five-Region Hypothesis Test for Exposure-Disease Associations

Shih

Lee

2017

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This is the new concept brought by the paper. However, the statistical methodology for this new concept is already available, for example in Liu et al (2005). Neverthless, the philosophy for evaluation of safety behind the two approaches will continue to debate for years to come.…”

Section: Commentarymentioning

confidence: 99%

Commentary on “Accounting for the Interim Safety Monitoring of an Adverse Event Upon Termination of a Clinical Trial”

Liu

2008

Journal of Biopharmaceutical Statistics

Self Cite

View full text Add to dashboard Cite

After more than 100 deaths caused by the Elixir Sulfanilamide disaster, in 1938, the U.S. Congress passed the Federal Food, Drug, and Cosmetic Act (FD&C Act) which for the first time in U.S. history required pharmaceutical companies to submit full reports of investigations regarding the safety of new drugs. Not until 1962, after the passage of the Leaver-Harris Amendment of the FD&C Act, was the U.S. Food and Drug Administration (FDA) authorized to require evidence of efficacy for approval of new drugs. Consequently, currently, for approval of a new drug the FDA requires that adequate and well-controlled clinical trials be conducted in humans to demonstrate substantial evidence of effectiveness and safety. In the evolution of regulations on approval of new drugs, therefore, safety of the drug was the primary issue and came before efficacy.However, this may not be totally true for pharmaceutical companies when developing new drugs and for regulatory agencies when approving drugs, at least in design, conduct, and analysis of adequate and well-controlled clinical trials. The current paradigm for approval of a new drug is two-fold. First is to prove that the drug is efficacious. Second is to verify whether there is any excessive safety risk. Most adequate and well-controlled trials are sufficiently powered to evaluate efficacy either by superiority or by noninferiority. However, most so-called "adequate and well-controlled trials" are not powered to detect excessive risk by the following hypothesis:where P T and P C are the proportions of patients with certain adverse events, for test and control groups, respectively. In (1), we use P T ≤ P C simply to illustrate the concept; other parameters for evaluation of risk, such relative risk, odd ratio, or hazard ratios, can also be used.

show abstract

“…In order to solve this problem, the odds ratio can be used as a measure of risk association for both prospective and retrospective studies. Lui [6] suggested using the odds ratio for assessment of equivalence.…”

Section: Introductionmentioning

confidence: 99%

Confidence intervals for odds ratio in matching pairs under inverse sampling

Jiang

2013

2013 6th International Conference on Biomedical Engineering and Informatics

View full text Add to dashboard Cite

Odds ratio is often used in matched pair design to assess the equivalence of two treatments for both prospective and retrospective clinical trials. In this context, we propose five confidence intervals estimator for the odds ratio in matching pairs under inverse sample. The performance of the proposed methods are evaluated with respect to their coverage probability and expected widths. Our empirical results show that the confidence interval based on the score statistic produce similar coverage probabilities and smaller confidence interval widths.

show abstract

Tests for Equivalence Based on Odds Ratio for Matched-Pair Design

Cited by 25 publications

References 6 publications

A Five-Region Hypothesis Test for Exposure-Disease Associations

A Five-Region Hypothesis Test for Exposure-Disease Associations

Commentary on “Accounting for the Interim Safety Monitoring of an Adverse Event Upon Termination of a Clinical Trial”

Confidence intervals for odds ratio in matching pairs under inverse sampling

Contact Info

Product

Resources

About