Appreciate principles of hypothesis testing.
Understand the different types of error.
Review the interpretation of screening tests.
Epidemiology and statistics remain crucial tools for the clinician, especially in pulmonary and critical care medicine. Hypothesis testing describes the process through which one develops a specific question, articulates it, and then determines whether the available data support or refute it. Hypothesis testing is more complex than selecting a specific statistical test for evaluating data and assessing whether the pattern observed in the data is likely to have occurred by chance or not. By convention, one attempts to restate the research question as a null hypothesis—therefore, if one predicts a difference, the null hypothesis is stated such that we believe there is no difference. Then, the investigator attempts to reject the null hypothesis. In hypothesis testing there are two potential types of error: type I and type II. Type I error indicates the error that occurs when the null hypothesis is rejected and it was in fact true, whereas type II error reflects accepting the null hypothesis when it was in fact false. These are essentially two sides to the same coin. Generally epidemiology reflects the study of the distribution and burdens of disease in a population. Descriptive measures in epidemiology distinguish between rates and ratios such as incidence and prevalence. Epidemiology also encompasses the analysis of systematic error that arises related to study design. Bias confounds the interpretation of data and clouds the assessment of the truthfulness or falseness of the hypothesis understudy. Finally, screening tests and their characteristics reflect a crucial component of clinical epidemiology.