Epidemiology, which was initially confined to the distribution of diseases in populations (descriptive epidemiology) and the factors responsible for such distribution (analytical epidemiology), also involves the evaluation of diagnostic and therapeutic strategies, as well as prognostic factors. Epidemiology provides elements that form the basis for medical and public health decisions. Information is collected from systematic records, which include large databases, or through specific surveys. Descriptive studies utilise mortality rates, as well as incidence and prevalence figures. The aim of aetiological epidemiology is to demonstrate a causal relationship between exposure and disease. It is necessary to check that studies do not present any major bias and to seek evidence in favour of causality. The strength of a causal relationship is calculated using relative risk or odds ratio. Other measurements of risk include attributable risk and aetiological fraction. Screening and diagnostic strategies are evaluated using the concepts of sensitivity, specificity, and above all, levels of likelihood that help determine the probability of a given patient becoming ill as a result of his or her personal characteristics and of test results. Evaluation of prognostic factors enables scores to be constructed allowing the probability of a given clinical outcome to be calculated as a function of patient characteristics.

### [Notions of epidemiology of value in dermatology].

Ann Dermatol Venereol. 2018;145(2):129-140.

MeSH terms: Bias; Big Data; Causality; Dermatology; Epidemiologic Methods; Health Status Indicators; Humans; Mass Screening; Odds Ratio; Prognosis; Risk Assessment

DOI: 10.1016/j.annder.2017.12.002