Innovations in science and medicine have delivered to us a long list of promising miraculous treatment pathways. We are on the brink of being able to leverage these innovations for more personalized medicine. For example, genetic variations inform the choice of the correct drug dosage required to control epileptic seizures. Research has identified the gene variant that predicts type 2 diabetes and the gene that elevates the risk of Alzheimer’s. Routine newborn screenings that track the thousands of gene variants are used throughout the individual’s life to manage their healthcare. Blood and other samples collected during medical visits are routinely sent for comprehensive molecular screening for a large number of cardiovascular, neurological, and other diseases that may develop in the future.

Medical information on a smart card will soon contain our unique individual molecular profile. Doctors will consult the profile before treatments or drugs are prescribed, thus avoiding costly mistakes that jeopardize patient safety. The rigid dataset silos of yesteryear will give away to a new era of intense information collaboration that will accelerate the drug development cycle in progress worldwide and accelerate health improvements. This means that the healthcare ecosystem will have to work together to integrate all data in a consistent and reliable manner with robust linkages between pharmacy prescription fill, doctor visit, claims, and laboratory data. Personalized medicine will be enabled by foundational data management infrastructure that will finally be able to use individual genetic differences and other patient population reported data to inform treatment plans.

Where traditionally doctors used past family history, socioeconomic circumstances, and environmental factors they will now be assisted in personalized medicine by genetic testing, proteomic profiling, metabolomics analysis, and a host of other scientific techniques that will become ubiquitous in the immediate future. The old paradigm of reactive medical care where a diagnosis is made, a drug is selected, then the drug is switched multiple times until effective treatment is found is totally outdated. This practice must be replaced by efficient medical care where the genetic predisposition on clinical screening of the patient informs the doctor’s diagnosis and prognosis. The right drug is selected based on comparative effectiveness research and every patient is monitored on the journey to full health.

For a single individual, hundreds of gigabytes of information could be gathered from tissue and cells, at multiple time points, and under varying environmental conditions. The collected data sets provide answers relating to the causes of disease and the best treatments for disease both at the individual level and at the diseased subpopulation level.

This clinical and wearable/sensor generated data deluge will need to be managed with high-performance data lakes and sophisticated predictive analytics. Integrating such data and constructing predictive models requires advanced approaches now employed mostly by physicists and climatologists. Managed data lakes and self-service data sets will be a fundamental component of clinical data management and health informatics, bringing clinicians and caregivers closer to their patient popular on data.

Healthcare data ecosystem

The promise of personalized medicine can only be realized by marrying life sciences and biotech, and understanding the potential of the vast amounts of biological data that clinicians, the patients themselves and their wearable medical devices can gather and share. Patients working with their physicians and care team can proactively learn about their own medical conditions, various treatment options, and be guided by advanced analytics in managing their health effectively.

Excerpt from white paper, Managed Data Lakes: The Foundational Fabric of the New Health Ecosystem, by John Poonnen and Kelly Schupp.

Download the full ebook here.