Guidelines for the descriptive presentation and statistical analysis of contact allergy data.

The present guidelines aim to support clinical researchers in adequately presenting data on contact allergy, and to use statistical tests appropriate for their data. A description of the mode of selection of patients, and of their relevant demographic details, is an essential prerequisite for the correct interpretation of study results. Proportions and rates, if regarded as estimate of these parameters of a target populations, should normally be supplemented with confidence intervals to address precision. Concordance, i.e., agreement between two ratings in a dependent sample, must be quantified with a chance-corrected measure such as Cohen's kappa coefficient. If the diagnostic quality of an outcome is being assessed, standard measures like sensitivity and specificity, as well as the prevalence-dependent positive and negative predictive values should be calculated. Often, contact allergy to a certain substance depends on several factors. In this situation, depending on the research question, techniques like stratification, standardization or multifactorial analysis should be employed. With increasing complexity of statistical description and analysis, consulting with a biostatistician is often mandatory.