Common statistical methods in orthopaedic clinical studies.

We discuss the statistical representation and management of random error in orthopaedic clinical studies. Descriptive studies (such as case series) collect information about a sample that may be generalized to describe a population. Typically this description is in the form of summary statistics, such as means, proportions, or rates. Error in these variables may be represented by confidence intervals. Correlation and regression are techniques for investigation of the relationship between two or more variables. Descriptive statistics, comparisons of groups, especially hypothesis tests, and assessment of association including correlation and regression are important statistical concepts for clinicians involved in the conduct or appraisal of orthopaedic clinical research.