No one expects a scalpel to perform surgery by itself. Similarly, no one should be surprised by the conclusion of a widely cited article in the January 2013 edition of Health Affairs (“What It Will Take To Achieve The As-Yet-Unfulfilled Promises Of Health Information Technology”) about the unrealized promise of health IT to lower costs and improve care.

Electronic health records (EHRs) are a tool, not a solution. But our experience at the New York City Department of Health and Mental Hygiene with the Primary Care Information Project (PCIP) demonstrates that with the proper support, EHRs can be a powerful tool for improving clinical care and managing population health.  Now we are about to turn another corner on our journey, to see if we can use the data we collect from New York City EHRs to paint an accurate picture of the population at large.

Since 2005, PCIP has helped more than 3,200 providers implement EHRs, and we currently provide assistance to more than 7,700 providers in New York City. Our staff members help with everything from choosing an EHR to billing, coding, and documenting workflows, to assisting providers with navigating the federal EHR incentive program, to reviewing their clinical quality measures and helping them with numerous quality improvement (QI) activities.  We are the home of New York City’s Regional Extension Center (REC), one of 62 such centers nationwide that are funded to support primary care providers through the transition to an EHR and help them achieve the federal Meaningful Use standards.

Using EHRs To Drive Quality Improvements And Increase Understanding At Multiple Levels

Individual practices. Unlike most RECs around the country, we have been doing this work for many years. In that time, we’ve developed data-based, quality-focused interventions that make a big difference to providers. We use aggregate (not patient-identifiable) monthly data from 3,000 providers in 550 practices with more than 3.3 million patient records and send them dashboard reports on the quality of care they are delivering. We can also look at data about a practice’s entire patient panel and determine if they need specific practice improvement coaching to, say, improve the care they provide their patients with diabetes.

By using these QI interventions, we are able to conduct in-depth interactions with providers, and assist practices in making improvements that facilitate improved quality of care. For instance, we help develop new clinical workflows that don’t just “computerize” existing paper workflows but take advantage of the unique functions and features provided by the EHR to improve care. A recent Health Affairs article points out that such technical assistance is essential to helping small practices use their EHR to improve quality. The Health Affairs article asserts that “sluggish adoption of health IT systems” is in part due to “the choice of systems that are neither interoperable nor easy to use; and the failure of health care providers and institutions to reengineer care processes to reap the full benefits of health IT.”

All patients in participating practices.  But what we’ve developed is a flexible system for querying practices and returning aggregated counts on vital public health questions, known as the Hub Population Health system, or the Hub. Patient-level data is not accessible through the Hub but its capabilities allow us to ask questions of our Hub patient population like, “How many patients in each borough have high blood pressure?” or “What percent of women in the Bronx over the age of 65 with diabetes have controlled blood sugars?” We do this by querying clinical information input into the EHR, such as diagnoses, blood pressure readings, or prescribed medications, and can specify the population of interest by borough, zip code, gender, or other characteristics.

We can also communicate with our practices via the EHR and tailor our messaging based on patient population to offer decision support. For example, if there was an outbreak in a specific area of New York City, we could target providers in those areas and send them messages within 24 hours instructing them on proper protocol.

The New York City population. We are working now to see if we can use this data to understand the population health of New York City as a whole, not just patients in participating PCIP practices.  To that end, we are working in partnership with the City University of New York School of Public Health with funding from the Robert Wood Johnson Foundation, the de Beaumont Foundation, the Robin Hood Foundation, and the New York State Health Foundation to benchmark our data against other gold standard data sources, such as the upcoming 2013 New York City Health and Nutritional Examination Survey.  We are calling this new EHR-based population health surveillance system “The NYC Macroscope.”

By comparing our EHR data with gold standard data, we can identify and measure some of the potential biases inherent in tracking population health using EHR data.

Like all surveillance systems, there are limits, and this project seeks to explore the potential of this data source and to quantify what is possible.  Once we benchmark and study the results, our goal is to develop strategies to generalize our findings to a larger segment of the NYC population — for example, to all New Yorkers with a regular source of medical care (approximately 82 percent of city residents) — much as opinion pollsters reweight their survey samples to reflect the nation. We also hope to help other regions and organizations learn from our work using EHRs for population health surveillance.

The Hub provides a rich data set for learning things that would otherwise take years to understand, and it enables us to explore potential connections, complications, and influences of factors that impact health, especially at the population level. Most importantly, we turn those data into actionable information that we deliver to our practices to help them improve the care they deliver.

PCIP offers demonstrable evidence of the promising potential for health IT. The sooner everyone understands that implementing EHRs in physicians’ offices is the starting line, not the finish line, the sooner EHRs can realize their potential as a tool to lower cost and improve care.