Simulation studies use computer intensive procedures to assess the performance of a variety of statistical methods in relation to a known truth. Such evaluation cannot be achieved with studies of real data alone. Designing high-quality simulations that reflect the complex situations seen in practice, such as in prognostic factors studies, is not a simple process. Unfortunately, very few published simulation studies provide sufficient details to allow readers to understand fully all the processes required to design a simulation study. When planning a simulation study, it is recommended that a detailed protocol be produced, giving full details of how the study will be performed, analysed and reported. This paper details the important considerations necessary when designing any simulation study, including defining specific objectives of the study, determining the procedures for generating the data sets and the number of simulations to perform. A checklist highlighting the important considerations when designing a simulation study is provided. A small review of the literature identifies the current practices within published simulation studies.

Prognostic models play a crucial role in the clinical decision-making process. Unfortunately, missing covariate data impede the construction of valid and reliable models, potentially introducing bias, if handled inappropriately. The extent of missing covariate data within reported cancer prognostic studies, the current handling and the quality of reporting this missing covariate data are unknown. Therefore, a review was conducted of 100 articles reporting multivariate survival analyses to assess potential prognostic factors, published within seven cancer journals in 2002. Missing covariate data is a common occurrence in studies performing multivariate survival analyses, being apparent in 81 of the 100 articles reviewed. The percentage of eligible cases with complete data was obtainable in 39 articles, and was o90% in 17 of these articles. The methods used to handle incomplete covariates were obtainable in 32 of the 81 articles with known missing data and the most commonly reported approaches were complete case and available case analysis. This review has highlighted deficiencies in the reporting of missing covariate data. Guidelines for presenting prognostic studies with missing covariate data are proposed, which if followed should clarify and standardise the reporting in future articles.