Advent of Data sciences
Pharma companies’ tryst with handling electronic data traces back to 1970s since the time Pharma companies moved on to EDC and set up IBRD(Institute for Biological Research and Development) , Cambridge Structural Database (CSD) and the Protein Data Bank (PDB)to circumvent data clean-up problems arising due to manual clinical data forms. This was only a starting point for development of tools that could analyze data and effectively using statistical models for Toxicity prediction, 3D Molecular modeling and Clinical Data analysis.
Age of Big data and Predictive Analytics
With the data deluge that follows the end-to-end drug discovery and development cycle, the resultant new age mantra is “Big data”. Pharma companies are looking to explore this data that can provide answers to questions they may not have even thought to ask. And these answers come from the most untraditional and unstructured channels of data. The key to gaining a multidimensional insight is to couple Big data with the traditional types of information that is generally collected and analyzed since time immemorial. It is a no-brainer that no organization would want to “hold” so much of data with themselves due to the sheer volume that needs to be stored and this has led to the emergence of aggregators that offer platforms and services for acquiring, organizing and analyzing big data, with enterprise-class performance, availability, supportability, and security features. Technologies such as Hadoop(open-source platform) aid in consolidating, combining, and transforming large data volumes while programming frameworks such as MapReduce support processing large data sets on distributed nodes to generate aggregated results.
Role of Data scientists
Data scientists who analyze and make meaningful inferences from big data in the Clinical side have aided Pharma companies in Patient Selection based on targeted genetic information, Real-Time Monitoring of clinical trials for Safety and efficacy and for prediction of adverse event even before commencement of a trial by tapping into historic side-effect data. Data Science has aided in targeted Sales and Marketing through Population Health Management programs focusing on target therapeutic areas and geographies, these have been facilitated through real-time analytics enabled mobile platforms. From a post-marketing perspective patient follow ups have been enabled through analyzing data from Social Media and from smart phones and wired health apps for understanding the drivers for non-compliance. Discussion forums (Patient and Physician) present the opinions about drugs in the markets. The key point here is that there is data lying everywhere in all possible forms waiting to be explored by the data scientists.
The flip side to the big data exploitation is the rising concerns on patient data privacy. As stated by McKinsey, “Pharmaceutical enterprises must understand and mitigate the legal, regulatory, and intellectual-property risks associated with a more collaborative approach.” There are consortiums such as International Pharmaceutical Privacy Consortium (IPPC) to take up issues pertaining to breach of data privacy. If the issues of data privacy are completely addressed and FDA issues a clean chit to all processes of collecting and processing Big data and the use that the results are put to, Data Science is definitely here to stay!