A research group conducted a study to compare the levels of creatine kinase-MB (CK-MB) between type 2 diabetes mellitus (T2DM) patients given statin therapy. Participants were divided into 4 groups based on treatment. Groups I, II, and III consisted of T2DM patients who had been given statin therapy (atorvastatin, simvastatin, rosuvastatin, respectively) for at least 6 months. Group IV consisted of T2DM patients who had not been given statin therapy. Which of the following statistical tests is most appropriate to compare the CK-MB levels between Groups I, II, III, and IV in this study?

Question

A research group conducted a study to compare the levels of creatine kinase-MB (CK-MB) between type 2 diabetes mellitus (T2DM) patients given statin therapy.  Participants were divided into 4 groups based on treatment.  Groups I, II, and III consisted of T2DM patients who had been given statin therapy (atorvastatin, simvastatin, rosuvastatin, respectively) for at least 6 months.  Group IV consisted of T2DM patients who had not been given statin therapy.  Which of the following statistical tests is most appropriate to compare the CK-MB levels between Groups I, II, III, and IV in this study?

Quizplus · Accepted Answer

11ec160b_372a_19fe_9249_71edb5afd575_MD0001_00 Variables are broadly classified as qualitative (ie, categorical) or quantitative (eg, continuous) based on their scale of measurement. Qualitative variables (eg, type of treatment, blood type) represent categories or groups, whereas quantitative variables (eg, temperature, glucose levels) represent numerical values, with quantitative variables (eg, temperature) sometimes transformed into qualitative variables (eg, "no fever" for <38 C [100 F]; "fever" for ≥38 C [100 F ]). The scale of measurement of the dependent (eg, outcome) and independent (eg, exposures, risk factors) variables in a study determines the correct statistical test for any given situation. The analysis of variance (ANOVA) test compares the means of ≥3 groups. It requires a categorical independent variable (ie, exposure) that is used to divide the study pool into ≥3 groups and a quantitative dependent variable (ie, outcome) for which an average (eg, mean) can be calculated. In this study: The quantitative dependent variable was the levels of creatine kinase-MB (CK-MB). The categorical independent variable was type of statin therapy that was used to categorize 4 different groups of patients with type 2 diabetes mellitus (T2DM): Group I (atorvastatin), Group II (simvastatin), Group III (rosuvastatin), and Group IV (no statin). An ANOVA test can determine whether there is a statistically significant difference in mean levels of CK-MB among T2DM patients given different statin therapies. A large, statistically significant difference in mean CK-MB levels among groups indicates that different statin therapies are associated with significant changes in CK-MB levels (ie, the null hypothesis is rejected). (Choice B) The chi-square test evaluates the association between 2 categorical variables, as in a study evaluating the association between sex (ie, "male" and "female") and myocardial infarction (ie, presence of myocardial infarction, absence of myocardial infarction). (Choices C and D) The t-tests compare the mean of a quantitative variable between 2 groups. A paired t-test is used when the groups are dependent (eg, comparing mean blood pressure [quantitative variable] in the same group of patients before and after antihypertensive therapy). An independent t-test is used when the groups are independent (eg, comparing mean CK-MB levels [quantitative variable] between a group of patients given statin therapy and another not given statin therapy). (Choice E) A correlation analysis uses the correlation coefficient to describe the linear relationship between 2 quantitative variables, as in a study evaluating the linear relationship between hours of sleep and irritability score. Educational objective: The analysis of variance test compares the means of ≥3 groups. The test requires a categorical independent variable (ie, exposure) that is used to divide the study pool into ≥3 groups and a quantitative dependent variable (ie, outcome) for which an average (eg, mean) can be calculated. __________ References: Statistical methodology: IV. Analysis of variance, analysis of covariance, and multivariate analysis of variance. (http://www.ncbi.nlm.nih.gov/pubmed/9523936)

A Research Group Conducted a Study to Compare the Levels