Historically, public health has been viewed as something of a poor stepchild when it comes to the billions of dollars in investments that have been aimed at transforming clinical care. Public health funding continues to be an issue, and the capabilities of local public health departments vary widely.
In fact, that is a problem even in the largest urban local health departments, according to a paper in the January issue of the Journal of Public Health Practice. According to a survey of 45 public health leaders in 16 large urban public health departments, timely information and data on chronic disease that are available in smaller geographic units are still difficult to obtain. Technology such as electronic health records (EHRs), claims data and hospital discharge data, was cited as useful data sources, but fewer than half of local health departments polled used systems that collected or disseminated data from EHRs or health information exchange.
The question is: Has the time arrived for public health to benefit from the technological transformations that are benefitting clinical acute care, and what are the barriers that stand in the way of data sharing that will enable a better partnership between public health and clinical care? In interviews, several public health experts give a qualified “yes.”
BRIDGING THE INFRASTRUCTURE GAP
Brian Lee is chief public health informatics officer, Office of Public Health Scientific Services, of the Centers for Disease Control and Prevention (CDC) in Atlanta. “New techniques and new technologies will allow us to amplify the work of a single smart epidemiologist or practitioner by allowing him to use tools that didn’t even exist 10 years ago,” he says. Although there will more data, and more timely data, available, the challenge for public health is going to be filtering out the relevant data, and especially making sure that the people who are interested at the local level have the ability to see that it’s relevant.
Public health departments will increasingly get data from EHRs, and it will be important that various systems that collect data for particular conditions work together, Lee says. He notes that investments are needed in EHRs to ensure that data systems are interoperable to public health, and that actionable public health decisions can be made from that data. He adds that standards are needed that allow for exchange of the right data and the right meaning of that data.
Lee adds that the CDC is developing standardized tools and procedures for rapidly collecting data, using it and tying it back so it is useable at the local, state and federal levels. One example of that effort is the CDC’s BioSense program, part of the national Syndromic Surveillance Program, which is aimed at developing additional analytics capabilities and more capacity to send and receive data as the level of syndromic data increases.
Jeff Engel, M.D., executive director of the Council of State and Territorial Epidemiologists in Atlanta, says the technology infrastructure requires setting up a platform that works within government IT. Security issues, firewall issues, and working with other departments such as public safety face every jurisdiction, he says. He also sees an emerging requirement for what he terms public health informaticians—highly skilled epidemiologists, who can write code and translate the English language into machine language, particularly for reportable conditions.
Brian Castrucci is program director of the de Beaumont Foundation, a Bethesda, Md.-based not-for-profit that “works to transform the practice of public health through strategic and engaged grantmaking,” and has a background as an epidemiologist with local health departments. He says the U.S. has a health system that is purely set up to manage the consequences of disease. Clinical practices have tremendous depth, but that is only half of the equation, he says. In his view, public health workers have the skills to analyze data around the upstream causes of diseases, but there has been a disconnect in that clinicians do not contextualize their information at the community level.
“If we can bring this together, use the multi-billion dollar investment in health IT to bridge the gap between those looking for the consequences of a disease and the causes, we could leverage better partnerships between clinical medicine and public health,” he says. He adds: “We have this Tower of Babel throughout health; we have a lot of information, but it’s in different formats, different EMRs, and in different organizations. We have to harmonize the information for the public health good it can do,” he says.
Castrucci calls attention to a public health concept called foundational capabilities, which is aimed at developing a national standard for what local health departments need to deliver their services, and is aimed at identifying the right data points for local public health. “EHRs get a lot of attention but they are insufficient. We don’t have the data in the right places to make the right choices for our citizenry,” he says. “We must figure out a way to get sub-county data in a real-time way, so our public health leaders can make the same informed decisions that our clinicians make when treating individual patients.”
BIG DATA: RELEVANCE ON A LOCAL SCALE
Getting the right data—both clinical and non-clinical—that is relevant to public health is also the central challenge of using Big Data, according to the CDC’s Lee. “The challenge is in working with healthcare, and also the broader health system, to identify what is helpful to public health,” he says, adding that a lot of what takes place at the state and local levels is useful to public health.
What is new in public health is an onslaught of novel data sets that require novel techniques, he says. “When you look at public health practices, and look at surveillance, we have incoming data streams, we have ongoing surveillance activities and we need to analyze that in real time or near real time, which requires Big Data approaches,” he says.
Oscar Alleyne, director of epidemiology and public health planning at the Rockland County Department of Health, Pomona, N.Y., who also chairs the Biosurveillance Workgroup of the National Association of County and City Health Officials, agrees. He says that epidemiologists “love data,” but local health departments require useful data. “We may have terabytes of data streaming in, but if only a gigabyte of that terabyte is useful data, it is of little use,” he says. He adds that that there has been progress from the perspective of what data points are relevant to public health.
“Local health departments can and do identify key indicators that are of value to the health of the community. When Big Data says ‘We have these elements,’ unless they can articulate how the data collected allows us to have actionable items, that’s the progress to which I am speaking,” Alleyne says.
Incorporating Big Data, improving bio-surveillance and incorporating electronic health record information in a meaningful way, will improve public health, according to those interviewed. Nonetheless, they acknowledge that significant hurdles remain. Engel, for one, is optimistic that new technologies will enhance public health, but he urges patience: “It’s not going to happen overnight in government because of the lack in resources and workforce. It’s going to take time.`