Journal Club Notes Learning Objectives O Review the

Journal Club Notes

Learning Objectives O Review the data tables and statistical methods used in the assigned article O Choose the proper statistical tests based on the distributions of the variable and on the cell sizes in the contingency tables O Interpret unadjusted and adjusted Odds Ratios O Explain how history is a threat to the validity of a study and what can be done to address this

Feldman et al, “Gestational Diabetes Screening: The International Association of the Diabetes and Pregnancy Study Groups Compared with Carpenter-Coustan Screening”, Obstetrics and Gynecology, vol. 127, no. 1, January 2016.

Common Statistical Tests O Compare means of two groups O T-test if normally distributed O Nonparametric tests if not normally distributed O Wilcoxon test, Mann Whitney U test O Compare proportions of two or more groups O Chi-square test is most common O If any of the cells in the contingency table (e. g. , 2 x 2 table) is less than 5 then Fisher Exact test is used

Wilcoxon test Fisher Exact test

Fisher Exact Test Wilcoxo n test

All outcomes are proportions so Fisher exact test used

Bivariate analyses Not conventional Multivariate analyses

Look at the magnitude of these ORs. Are they clinically significant?

Study Design Consideration O One threat to the validity of a study is history O History refers to something else that occurs at the same time as your intervention that may affect the outcome O How do you determine whether your intervention makes a difference? O Importance of a control group O What the is control group in this study?