Medical informatics is playing a significant role in a unique, newly-launched partnership between Durham-based Duke University and the Durham, NC, community. The goal of the partnership, known as Durham Health Innovations (DHI), is to improve the health of everyone living in Durham County by using medical informatics to identify interventions for community members whose needs aren't being met successfully by conventional methods.
"The informatics side of this project is deeply embedded," says Lloyd Michener, MD, chair of the department of community and family medicine at Duke. "The entire project requires a very robust backbone and system. This is an example of what you can do with really good information systems and analytic tools. You're basically doing real-time epidemiologic studies."
The data the partnership uses comes from Duke University's electronic medical records (EMR) system, which is a variation of the system developed by San Francisco-based McKesson. Duke has implemented the system in all of their office practices.
The university collects the data from the EMR and runs customized software on it that assigns geographic locations to the data, also known as geocoding. This process, which is HIPAA-compliant, allows them to look at areas of disease clustering in order to treat specific diseases or disorders prevalent in areas within the community.
"What we're doing that's unique is that, rather than looking at one disease or one subpopulation, we're doing this as a system experiment of how we can look at the health needs of our entire quarter-million people in Durham County and redesign how the system works to improve health," says Michener.
The plan includes 10 individual projects aimed at reducing death or disability from specific diseases or disorders prevalent in the community. The focus of the teams on particular health problems was determined by the community, not by Duke University.
The projects include: adolescent health, asthma, cancer, heart disease, diabetes, HIV and sexually transmitted diseases, maternal health, pain management, substance abuse, and seniors' health.
Each project team receives $100,000, which is distributed among team members to help them accomplish the work. Funding for the projects comes from grants from the National Institutes of Health (NIH) and Duke Medicine. One surprising aspect of the project is that they are not asking for new money to implement the projects. According to Michener, all the teams must reallocate existing money to achieve better outcomes.
A model project
Communities currently track their health status through periodic health department or Centers for Disease Control and Prevention (CDC) surveys. These surveys typically rely on data that comes from individual practices, hospitals, and emergency rooms, which aren't aggregated to look at disease patterns or clusters.
By the time the survey results are released, they may no longer contain timely and relevant information. By comparison, Michener says that Duke has primary data that they can aggregate to look at disease frequency pattern, and severity, so that they can intervene immediately.
According to DHI, the health of Durham county residents is comparable to that of most other residents in other parts of the state. However, North Carolina still ranks in the bottom 20% of all states in terms of life expectancy and functional status.
Michener believes that the project could become a model for the entire country. While health systems such as Kaiser and Intermountain have conducted similar projects, Michener says that they only benefit those seen by that particular health system.
"We're trying to do it for the whole county, including folks who aren't primarily our patients," he says. "We don't know of anyone else who's tried to do a scale experiment for a quarter of a million people. You have to be a big enough healthcare provider that you can affect the whole community."
Data helps project members impact ER rates
The data has already been used to implement successful changes in the community. Recently, the data revealed that there were children with asthma who were visiting the emergency room once or twice every couple of years. However, when they viewed the information in aggregate form, the numbers were much larger in Durham than in other communities.
When researchers ventured out into the community to find out why, they discovered that families were being given different asthma control by different healthcare providers, causing confusion. As a result, they were visiting the emergency room more often.
"We came to an agreement across all the practices, both Duke and non-Duke, on common management plans for asthma that we all use now," says Michener. "And ER rates dropped."
In another example, Michener says researchers looked at low-income senior citizens in Durham County and discovered that they, too, were visiting the emergency room far too frequently. He says the elderly residents were putting things off and weren't getting their medications refilled; therefore, they were getting sick and visiting the ER more often.