TECHNOLOGY Q&A

Biostatistics

What is meant by Biostatistics?

Biostatistics is the science of leveraging statistical inputs, methods, and models to facilitate decision making in biological experiments, at various points across a bioresearch/study lifecycle (starting from conception to design, conduct, detailed analysis, and reporting). The applicability of biostatistical techniques (SAS being the preferred validated standard while R is also in use) is paramount in public health and medicine. Biostatisticians/biometricians can work for a biopharma/biotech/medical devices or clinical research organization, where they can be involved across multiple stages - starting from the planning and design of a study, which involves activities like determining the sample size, helping with the feasibility assessment of trials and sites, budget planning and forecasting for certain trial related imperatives, finalizing on the protocol elements and schedule of events, coming up with the statistical analysis plan (SAP) while optimizing on the same during study progression, to support both nonclinical and clinical trials, and health economics and outcomes research (HEOR).

During the Trial conduct phase, some of the Clinical Database Management activities are the initial study database design, data and study performance monitoring during the course of the study, etc -In the Clinical Trials Data analysis phase activities are performing the analysis as per SAP, validating study results, interpreting key findings, safety analytics and reporting 'submission ready' artifacts as per regulatory requirements

Over and above this, biostatisticians are also expected to drive process excellence by improvising/automating existing processes and perform cross-functional collaboration with data management and medical writing teams.

How the future is shaping up, from the perspective of a biostatistician/informatician/biometrician?

  • Big Data in Healthcare: Data is increasing exponentially, so is the number of clinical systems feeding data, and hence there is a need of a robust clinical decision support system to be able to synthesize evidence from disparate and multiple new data sources, e.g. leveraging real-world data for evidence generation, integrating data from wearables and sensors, EMRs, etc.
  • Quick real time clinical decision support tools (for data modelling, exploration, and visualization) across the entire clinical development lifecycle to reduce TAT
  • Flexible, adaptable, and innovative design framework that can support the complex clinical trial scenarios across therapy areas like oncology
  • Use of next gen predictive analytics based technologies like AI/ML/NLP for advanced statistical analysis for intelligent and intuitive decision support
  • All of this while ensuring a cost effective solution