Health care organizations are collecting a lot of data. This ocean of information, when providers apply analytics, can offer invaluable insights to improve chronic disease management, track at-risk patient populations and improve operational efficiency. When genetic data and gene sequencing are added to the mix, providers have powerful tools to spot health risks within patient populations. These tactics are the Holy Grail for improving health care.

There are still barriers to overcome — notably concerns about patient privacy or ownership of data. And it might be a while before we realize truly sophisticated, data-driven and interoperable population health management nationwide. But forward-looking organizations are plunging into the big data realm with the goal of improving overall patient outcomes.

Moving the Needle on Population Health Management

According to the Centers for Disease Control and Prevention, chronic diseases such as heart disease, stroke, cancer, diabetes and arthritis are among the most common, costly and preventable of all health problems. Yet they continue to cause seven out of 10 deaths among Americans each year, despite the health care system's spending more per patient than most developed countries.

This model is unsustainable, and the results attained thus far are unacceptable. But there is a silver lining: Providers recognize that harnessing data is key to managing the health of patients, especially those with chronic disease. Below are three main considerations for improving population health management:

Interoperability and openness. Creating secure and free-flowing information between providers and other stakeholders to ensure that consistent, high-quality care is available to patients regardless of their location.

Actionable data. Generating accurate data and valuable insights, with discrete elements that providers can mine easily for trends, will be critical in helping clinicians make better informed decisions at the point of care.

Patient engagement. Encouraging patients to take a proactive role in not only exchanging information about their health, but also using that information to engage actively with their care team and adopt healthier habits through wellness programs.

A Case Study

Signs of early progress in population health management are visible in places such as the University of Pittsburgh Medical Center (UPMC), one of the largest health systems in the United States. UPMC includes 22 hospitals with 4,000 beds, more than 400 outpatient locations and nearly 3,500 employed physicians. UPMC is aggregating and integrating patient data from 48 major clinical systems.

The health system also manages its own health plan, which serves more than 2 million patients through a network of 125 hospitals and more than 11,500 physicians. This not only puts UPMC in a unique position to address concerns about patient outcomes through newer care models, but also to evaluate financial and operational efficiency from the payer perspective.

For UPMC, connected health care makes it easier to find relevant data in a system that handles a large amount of information from both payers and providers. The system's 39,000 users and 5 million patient records generate a massive amount of data totaling 5.4 petabytes, and this is expected to double every 18 months. For some perspective, one gigabyte is worth about seven minutes of high-definition video, while one petabyte is about 13.3 years' worth.

It's no meager task, and UPMC's interoperability strategy is ambitious. But extracting meaning from data and defining patient populations is critical to identifying gaps in care. UPMC's strategy is to “filter the noise” and help clinicians to easily access key nuggets of harmonized data with population health management tools.

More than seven years ago, UPMC decided to pursue a strategy to ensure that systems were speaking to each other. The first step for UPMC was to ensure interoperability that enables the aggregation of relevant data from different sources within the UPMC health system. Then, the data were organized into manageable buckets that are clinically meaningful. For example, if there's a population of diabetics or patients with congestive heart failure, UPMC used the harmonized data buckets to tag the patients and determine the ideal clinical and evidence-based guidelines as well as best practices to manage these populations more comprehensively. For the patients, this has helped with pre-emptive diagnosis and earlier intervention. From the payer perspective, these data have helped to identify gaps in cost and research outcomes, as well as physician performance, which help to reduce operational waste.

While UPMC has made strides toward managing population health, there are more opportunities throughout the continuum of care to build an even richer data set: genomics.

The Hope and Promise of Genomics

Genetic and genomic information is becoming more accessible and valuable to clinicians and even patients. The cost of gene sequencing is getting lower — websites now offer genetic analysis that, for a nominal fee, gives consumers a full genetic profile outlining the diseases they are predisposed to and even what medications and treatments would work for them. This development is part of a growing trend in the consumerization of genomic data.

But for the provider, capturing and analyzing all genetic data — some of which may not be included in the electronic health record — continues to be a challenge. Some perspective: A single human body contains more than 140 million petabytes of information. Imagine gathering those data for an entire population.

At UPMC, genomic research started at the Children's Hospital of Pittsburgh with newborn screening, as part of one of the largest public health genetic programs in the world. UPMC is using data from newborns to facilitate early detection, diagnosis and intervention for a multitude of genetic, endocrine and metabolic disorders. Gene sequencing also has been critical for cancer patients at UPMC, helping to diagnose breast cancer pre-emptively and determine a more feasible course of treatment.

But there is a darker side to genetic data. What if gene sequencing finds a propensity toward chronic disease or identifies the patient as a carrier of mutant genes associated with a higher likelihood for certain types of cancer? Will disclosing the information make the patient a high risk for insurers? Would having this knowledge impact patient wellness and lifestyle management? There are also privacy and security concerns about the management of this information. Do patients own their data or does the provider? Would patients have the choice to opt out of providing their genetic information? There are also challenges in presenting genetic data to the right clinicians in the manner that makes these data useful.

The full repercussions of sharing this information are unknown but, gradually, the industry will create the right regulatory controls and define roles and responsibilities for following best practices.

Toward a Healthier Future

What's more exciting for clinicians is the possibility of using genetic data to improve clinical decision-making at the point of care. For example, knowing about a patient's biomarkers and other genetic information, a physician can determine which medication or treatment would be more effective for the patient before prescribing it. The promise of interoperability lies in connecting the dots between a multitude of data silos to pull together the complete patient story.

While the industry may not have smoothed all the wrinkles in the data conversation, the technological advancement in population health management opens up the doors for intelligent insights and conversations among patients, providers and even payers. There are plenty of opportunities to use critical data to manage health in a better way, focusing not only on disease, but also on overall wellness.

Rasu Shrestha, M.D., is the vice president of medical information technology at UPMC, headquartered in Pittsburgh. Martha Thorne is the general manager of performance and care logistics at Allscripts, headquartered in Chicago.