Over the last decade, pharmaceutical companies have been aggregating years of research and development data into medical databases, while in the USA at least, there have been many government dollars put into standardizing electronic medical records. The US government and other stakeholders have also been looking to utilize the significant volumes of health-related information, including data gathered from clinical trials and patients covered in public insurance programs.
At the same time, recent technical advances have made it easier to collect and analyze information from several sources which is a big deal in healthcare, since data for a single patient may come from all sorts of sources like; Insurance companies, hospitals, laboratories, physician offices, etc.
All of this information and how the sheer volume of it can be used is the purview of Big Data applications.
Currently, big data applications are not widely used in Life Sciences, though one of the areas where the techniques have made some inroads is Biometrics and in particular clinical data research.
In Clinical Trial Patient Recruitment one concern is potential patient recruitment overlaps, i.e. Two trials for the same indication but slightly different subpopulations.
Big Data techniques can provide an overview of the scale of the overlap, and allow insights into whether the studies will be in direct competition with one another. If so, this can change the site and/or sample for both studies.
Risk plays a major part in all safety analyses, and the more information collected on risk, the safer the study should be. Big Data currently helps by
- Allowing the Creation of Benefit/Risk profiles which feed into Risk Management Plans
- Helping to monitor risks associated with in populations taking certain compounds or with certain diseases in order to gauge potential impact.
- Assist in ethical decision making – given what is known about the molecule under test, deciding if it is ethical to treat patients with a certain medical history.
New Uses Just Starting
In the field of Health Economics new techniques can be used to prepare analyses (mainly using the insurance claims data sources) on charges and expenditures relating to certain treatments.
Funding bodies, including government agencies in many countries need to make decisions about whether to pay for a drug. Increasingly, the information is gathered from many sources, cost/benefit ratios can be analyzed and comparative studies can be performed against competitor compounds.
This is where the use of multiple data sources becomes very useful. Clinical trial data combined with that from any other source relating to the patient’s particular situation may indicate a particular treatment regime, and assist in identifying the strategy with the highest chance of success.
As more and more sources of information become available, the techniques of sorting out what is useful and what is not become increasingly sophisticated. It is likely that there will be an increased use of:
- Volumes of genomic data and opportunities to link this data to patients in the real world, e.g. to include Ancestry, Health condition risks, Predictive profiling around ageing and diseases.
- Text analytics to mine for information from online publications, specific web site (PubMed, Clintrials.gov, etc.), and disease profiling, i.e. Information on Co-medications, Co-morbidities, and Expected and actual standards of care.
- Different Risk Mitigation strategies, e.g. potential risks to patients to be avoided.
Added environmental factors (proven or otherwise) such as Diet, Local High voltage lines, Traffic, Pollution, Climate change, Pollen counts, Temperature data, and Pesticide use – etc.Adding the potential of a feedback mechanism for patients, the use of data from Social media (a more reliable measure of patient behavior?), and the potential for input from the “internet of things”, wearable technology, and “traditional” mobile apps there is huge opportunity for Big data in the Healthcare sector
Big Data is already making something of a breakthrough in the clinical trials area, helping to bring down development costs, increasing the chances of new discoveries success, increasing patient safety and reducing the time to market, but this is just the beginning.
As more and more “things” become connected to the internet, more and more information is becoming accessible with ever increasing complexity and links between caches of data.
In conclusion, Companies need to think now about their strategies in these areas. There is a lot going on, and with so many possibilities the key is going to be targeting the right data for answering the right questions.