Services: The HDDC Program in Biostatistics and Study Design
- The Biostatistics and Study Design Program provides HDDC investigators with access to high-level expertise in experimental design, clinical research methods, biostatistics, epidemiology, clinical trials computer programming, data management, quality control, ethics and regulatory requirements for IRBs, the FDA and government. The Program also plays an important role in the education and training of clinical and translational investigators.
- Expertise Provided:
- Study Design: methods for randomization including cluster randomization and randomized recruitment, quasi-experimental alternatives, sequential inference and complex designs for clinical studies such as multi-period crossover trials;
- Sample-size calculations and power projections for standard and non-standard study designs such as unbalanced, multi-category, covariate-adjusted, or multi-factor comparisons;
- Case Report Form/Survey Design including psychometric development of symptom scales or clinical assessment instruments, and designs for establishing validity and measuring inter- and intra-rater reliability;
- Evaluation of new diagnostic techniques in comparison to a “gold standard”, including examination of covariate influences by multiple logistic regression and receiver-operating characteristic (ROC) analysis;
- Data Management Systems Design;
- Biostatistical analysis methods including advanced analytic techniques such as multi-factor and repeated-measures analysis of variance, multiple linear and logistic regression, principal-components analysis, survival (“time-to-event”) analysis, conditional logistic regression for paired binary data.