IDIBELL offers expertise in statistics for clinical and basic researchers, for planning experiments or studies, analyze data and assess the interpretation of results.
Activities
a) Service of statistical support for research studies
• Specific statistical support: resolution of doubts, analysis focus or review of the adequacy of already performed analysis.
• Major advising: continued support for projects in their various aspects: design, database development, analysis and interpretation of results.
The activities include both basic research studies such as clinical research, observational and clinical trials. We work together with IDIBELL's UCICEC (Central Clinical Research Unit in Clinical Trials), which gives support and statistical advice regarding clinical trials (http://www.idibell.org/modul/ucicec).
b) Teaching
Training courses and seminars will be offered, both for internal and external IDIBELL researchers.
The courses may have different orientations: level of design and protocol development, design of experiments, methodologies, data management, data analysis and statistical results.
Service Request
The application procedure consists on three steps:
• A mail should be sent to sae@idibell.cat with a short description and contact details of the researcher.
• There will be a preliminary meeting to explain the project and the type of required assistance.
• Depending on specific needs, doubts will be resolved at the same time or requested assistance will be budgeted.
Examples of types of applications:
- Guidance and specific advice.
- Supervision of the analysis as they are developed by the research group.
- Analysis and conclusions review, once they have been made by the research group.
- Designing experiment and study prior to its inception.
- Complete data analysis provided by the research group.
Examples of specific services that may be requested:
- Calculation of the needed sample size for an experiment or study.
- Descriptive statistics of patients' surveys or questionnaires results.
- Analysis of designed experiments.
- Analysis of risk factors or prognosis.
- Analysis of data mining of genomics, proteomics, epigenetics.
- Analysis of microarray data to find biomarkers.
- Analysis to identify genetic polymorphisms of susceptibility or prognosis genes.
- Corrected p-values of statistical tests for multiple comparisons to control FDR (False Discovery Rate).
- Multivariate logistic regression analysis, Poisson, Cox models, principal components, multidimensional scaling, etc.
- Statistical monitoring to send articles for publication.
- Statistical monitoring of research projects.
- Statistical features in clinical trial designs and protocol development. Randomization, data management and analysis of clinical trial results.