TCS Biostatistics and Statistical Programming Services
Turn clinical data into actionable insights
Industry
Life Sciences
Patient-centered Care Models Taking Center Stage in Life Sciences
Highlights
- Clinical trials are becoming increasingly complex, with many now leveraging genomic biomarkers data to improve treatment efficacy and precision, minimise adverse effects, and support adaptive treatment-arm assignment.
- They throw up a plethora of data, but integrating data from diverse sources, and reconciling and generating meaningful insights are becoming daunting, thus making statistical analyses cumbersome.
- This calls for capabilities for complex statistical modelling, analysis of growing multimodal data, and real-time data surveillance by integrating advanced analytics, automation, and modern data science to help implement adaptive strategies.
On this page
On this pageinpage
Overview
Advanced innovative and personalised medicine has revolutionised clinical research, leading to improvements in drug discovery, study designs, data collection and processing, statistical methodologies, tools and technology, and reporting.
Clinical trials today generate unprecedented volumes of diverse and complex data. Given this data deluge, integrating data from diverse sources, reconciling, and generating meaningful insights are becoming daunting.
Real-time analysis and interpretation of this data is essential to clinical trial outcome, and it demands high expertise.
The changing dynamics of the pharma industry requires advanced statistical and programming skills to implement adaptive strategies and statistical approaches to analyse data and generate meaningful insights. It involves data integration and reconciliation from diverse sources such as case report forms (CRFs), electronic health records (EHRs), electronic medical records (EMRs), wearable devices, and patient registries, prior to data standardisation and analysis.
Technologies such as R, Python, and artificial intelligence (AI), generative AI (GenAI), machine learning (ML) help optimise clinical trial lifecycle and reduce cycle times.
Our offering
TCS offers a comprehensive service offering for clinical trial analysis, right from designing the study and planning the analysis, to data transformation, data analysis, and reporting.
TCS Biostatistics and Statistical Programming Services include protocol development support, sample size calculation, randomisation schedule generation, statistical analysis plan (SAP), mock shell creation, and interpretation of results. Our expertise also extends to complex statistical models such as mixed model for repeated measures (MMRM), logistic model, hazard model, analysis of variance (ANOVA), analysis of covariance (ANCOVA), missing data analysis, time series analysis, cluster analysis, regression analysis, and survival analysis.
The TCS offering also includes regulatory compliant Clinical Data Interchange Standards Consortium (CDISC) deliverables such as study data tabulation model (SDTM), analysis data model (ADaM), and tables, listings and figures (TLFs). TCS standardises and streamlines the statistical analysis and reporting processes such as metadata-driven approach for creating SDTM and ADaM.
We provide statistical and programming support enabling rapid turnaround of data curation and analysis for annual and periodic safety reports such as Data and Safety Monitoring Board (DSMB), Development Safety Update Report (DSUR), and Periodic Benefit Risk Evaluation Report (PBRER). Our approach ensures compliance, transparency, and traceability while addressing complexities like multi-source data integration, data lock alignment, and integrated analyses.
Our Biostatistics and Statistical Programming (BSP) Centre of Excellence (CoE) team delivers a range of specialised services such as statistical monitoring, anticipated adverse event analysis, data monitoring committee (DMC) analysis, e-narratives, e-submissions, and statistical support for chemistry manufacturing and controls (CMC).
Drawing on insights and leveraging our expertise of over 20 years, we have created notable assets, such as:
- TCS ADD™ Metadata Repository platform: This provides framework for standards management, facilitates traceability of clinical trial data, enabling automated SDTM and ADaM transformation.
- SAS programs library: This library hosts template-based SAS programs and SAS utility macros for standard safety and efficacy outputs.
Benefits
Our extensive and deep domain knowledge and technological expertise have led to successful implementation of cutting-edge solutions leveraging AI-ML and GenAI in core areas of BSP.
- Our AI-powered SDTM, ADaM, and TLF creation enables early availability of analysis-ready datasets and outputs, supporting faster interim analyses, decision-making, and regulatory submission.
- Our statistical programming experience extends beyond SAS to the use of R technology in areas such as data transformation and analysis, data visualisation, pharmacokinetics (PK)/ pharmacodynamics (PD) modelling and analysis, and Python in reading medical images.
- Point automation: Implementation of several simple to complex point automations have led to saving of efforts, and reduced costs while delivering high quality clinical data analysis and reporting.
- Transparent metrics: Our ability to deliver customisable and transparent metrics embedded into an objective-oriented governance structure ensures successful partnerships.
- Our ability to manage end-to-end study deliverables, close collaboration, and quick exchange of information between functions offers synergistic benefits such as early identification of potential issues and reduced hand-offs.
- Our expertise facilitates real-time review for data-driven clinical trial insights such as early detection of trends, and discovery of anomalies, nurturing timely and well-informed decision-making.
The TCS advantage
We provide seamless, end to end clinical data services powered by real time insights, scalable expertise, and flexible pricing. Our multi-technology capabilities and continuous upskilling ensure high quality, efficient, and innovative delivery for our customers.
- Our flexible operating models and skilled resource scalability meet the growing needs of our customers.
- Our dynamic pricing models comprise study-based, work package-based, unit-based, and hybrid models, thereby ensuring that our clients reap maximum benefits.
- To upskill our associates in clinical data science, we have created various modules covering advanced computational techniques, machine learning, and big data analytics.
Solving integrity issues
Statistical analysis for a global biopharma organisation in Europe helped identify integrity issues in clinical trial data (risk-based quality monitoring).
TCS data analysts collaborated seamlessly with customer’s critical risk leaders (CRLs) to identify potential anomalies in clinical data.
We used R, R Shiny, and Gitlab tools and applications to explore data patterns for overall site and subject performance. R programs were created to perform various analysis such as cluster analysis, multivariate analysis, and adverse event reports. These analyses enabled identification of anomalous data patterns, detect outliers, unusual correlation, and site under-reporting of adverse events (AEs).
Visualisations of critical data points were created for monitoring major protocol deviations, adverse events and data discrepancies, facilitating early detection of potential deviations at a site based on predefined quality tolerance limits (QTL).
For QTL and statistical monitoring (SM), approximately 100 unique studies were monitored annually to generate compliance and performance reports. During these monitoring activities, breaches were detected in about 50–60% of the runs, indicating that a significant proportion of the executed checks or analyses had flagged deviations from predefined thresholds or statistical norms.
Based on the analysis outcomes and visual displays, we recommended the customer’s study team to enhance clinical site performance and incorporate additional data monitoring measures.
With these enhancements, the study team was better equipped to further investigate the highlighted patterns and make early corrective actions to improve the overall quality of data and increase study compliance.
Automating data anonymisation
We automated data anonymisation for a US-based biopharma.
A metadata-driven, SAS-based approach was developed by TCS to anonymise data.
The requirement was to anonymise huge volumes of study data, within a short period of time. The studies included datasets in CDISC-aligned SDTM as well as legacy data standards.
In just a month, TCS built and implemented an automated, 21 CFR part 11 compliant system by leveraging multiple technologies. This replaced the traditional SAS only programming approach, which was time-consuming.
This ensured the customer met tight timelines for anonymised datasets submission as per their data disclosure agreement.
The automation resulted in 60% savings on effort, significant cost reduction, improved turnaround time, and high-quality deliverables.