Clinical studies are performed for a variety of reasons — preliminary evaluation of safety (first in human, dose escalation, etc.), assessments of efficacy (comparisons of dosage levels, comparisons vs. gold standard, etc.), or performance/user preference comparisons versus alternative products or treatments. Regardless of the study objectives, statistical expertise is critical to running a successful trial that will yield valid results.

At PMI, our statisticians and clinical programmers have experience with trial design and data analysis across a broad range of medical specialties and therapeutic areas. They have been involved in various types of trials, including preliminary, early-phase studies, pivotal studies for regulatory submissions, and post-approval/marketing studies. PMI provides statistical and clinical programming support for each stage of the clinical study process — before, during, and post-study.

Pre-Study
Careful planning in the design phase will not only help provide that the data collected can be used to evaluate the study hypotheses; it can also increase efficiency and decrease costs. During the pre-study stage, we provide:

• Input on clinical study design: Providing a statistical perspective as to the appropriate models for study design and objective study endpoints

• Sample size and power analysis: Determining the number of clinical study subjects necessary to appropriately evaluate study hypotheses

• Protocol development: Collaborating with the clinical team and sponsors to develop the statistical sections of the protocol and to determine that study endpoints are appropriately collected for the analysis of study objectives

• Randomization schedule development: Minimizing the likelihood of study enrollment bias

• CRF Review: Confirming that data is captured in an appropriate manner in light of planned safety and efficacy analyses. Careful review by Biostatistics and Clinical Programming will also provide that data is captured efficiently so that a quick turnaround from data entry to analysis results, with a minimum of queries and data clarifications, is possible.

During the Study
Once a trial has begun, the Biostatistics and Clinical Programming Team can provide a variety of services so that valuable inferences can be drawn from the data at the conclusion of the study. To assist in the analysis of the data, we offer:

• Statistical methodology consultation: Determining the appropriate statistical tests, providing input on the types of tables and graphs best suited for summarizing the study endpoints, re-evaluating study assumptions, and making appropriate modifications to the analysis as necessary

• Statistical analysis plan (SAP) development: Providing pre-specification and documentation of the planned analyses in a comprehensive SAP in order to maintain the validity of statistical results. The SAP typically includes operational definitions of endpoints to be analyzed, rules for data handling and for inclusion into the various analysis populations (intent-to-treat, per protocol, safety, etc.), description of statistical methodology, identification and presentation of shells for planned statistical tables, figures and listings.

• Analysis dataset specifications: Developing specifications for the analysis datasets to document the algorithm used in creating analysis variable

• Interim statistical analyses: Performing pre-specified interim analyses for regulatory bodies, safety monitoring boards or as otherwise requested by the sponsor

We also work with the Data Management team to conduct a careful, ongoing review of data as it becomes available to create a smooth transition to database lock and the generation of tables, listings and graphs. To assist in the review of incoming data, we offer.

• Customized data listings: Enabling independent parties to conduct blinded or un-blinded safety reviews or facilitating the Sponsor’s data cleaning processes

Post-Study
After the clinical data has been entered and cleaned, the Biostatistics and Clinical Programming Team work together to determine whether or not the study’s objectives have been met. At the conclusion of the study, we provide the following:

• Final statistical tables, listings and figures: Generating tables, listings and figures from the locked database in a format that can be easily incorporated into a final report or presentation

• Review and interpretation of statistical results: Interpreting statistical results to facilitate the understanding of the data analysis and provide guidance to medical writers on the implications of the statistical results

• Ad-hoc exploratory analyses: Providing additional analysis of the data as indicated by the initial study results

Additional Support
In addition to the services we provide for formal clinical studies, PMI’s Biostatistics and Clinical Programming Team offer assistance for:

• Scientific presentations/journal articles: Providing statistical consultation and clinical programming support for scientific abstracts, presentations or articles submitted to conferences or publications

• QC of tables and analysis datasets: Providing an independent accuracy check of Program Sponsor’s tables and analysis datasets based on the SAP and analysis dataset specifications

Quality Control
PMI’s quality control procedures for statistical output involve a combination of independent programming and internal review to verify accuracy and completeness of tables, figures, and listings.