November 2016

HIT's Public Health Report Card
By Selena Chavis
For The Record
Vol. 28 No. 11 P. 18

An analysis of an EHR implementation reveals there's much work to be done in this sector.

Public health departments across the nation are well aware of the role that effective and timely response play when a crisis presents. When stakeholders within and across communities come together, share critical information, and collaborate, the fallout of emergency situations is often minimized and major crises are averted.

It's understandable then why the industry has placed great hope in HIT and health information exchange becoming catalysts for improving public-private health partnering and speeding crisis response. Yet, some research and analysis of recent public health events suggests that the industry still has a long way to go before it fully realizes the benefits of health information exchange within the public health sector.

Specifically, disease outbreaks such as the Zika virus and Ebola are delivering key insights into the gaps and barriers that exist throughout the HIT spectrum. University of Maryland researchers recently completed a two-year "intensive analysis" of an EHR rollout in the public health sector, uncovering a number of challenges and barriers to effective use. The technology was implemented within the Montgomery County Department of Health and Human Services (DHHS) and a public–private network of safety net clinics supported by the Primary Care Coalition of Montgomery County (PCC).

Following the effort, Ritu Agarwal, a senior associate dean, professor, and codirector for the Center for Health Information and Decision Systems in the Robert H. Smith School of Business at the University of Maryland, was quoted as follows:

"Health departments have lacked guidance to effectively strategize about appropriate IT investments, and incidents like the current Zika crisis bring the issue to the forefront. From intensive analysis of the rollout of an electronic health records system [in a nearby county], we uncovered a host of barriers and obstacles to effective use of information, including the complexity and usability of the software, the inability of the software to support certain unique public health reporting needs, the learning curve for public health workers, and the lack of standards for effective data exchange. All of this does not bode well, either for crisis response or for proactive crisis anticipation."

In tandem with these insights, a recent essay published in PLOS Medicine titled "Make Data Sharing Routine to Prepare for Public Health Emergencies" points to the importance of medical and public health research in accelerating outbreak control and underscores the importance of open science—a movement focused on making scientific research and data accessible to all levels of society. Lead author Jean-Paul Chretien, from the Integrated Biosurveillance section of the Armed Forces' health surveillance branch, and colleagues also point to data sharing as a key impediment to realizing the advances of HIT during a public health crisis. Fueled by such challenges as long-standing academic norms and human and technical resource limitations, the analysis uncovered three key issues: a lack of standards to credit data providers, hesitation by some scientists to share data, and inadequate technology, standards, and human capacity.

Marc Probst, vice president and CIO at Intermountain Healthcare, agrees with the assertion that there is greater need for industry standards. In fact, as an original member of the Office of the National Coordinator for Health Information Technology's HIT Policy Committee, he has been carrying the standards flag for some time now. He points out that efforts to develop common data sets for exchanging information have not gone deep enough. "For us to have the kind of reaction time we want in public health or the ability to gather the kind of broad set of data to really do public health well, standards are necessary," Probst says. "They are not looking at a single encounter or a single surgery or a single procedure—they are looking at a population and what's happening in that population. To do that without data that are standardized … it's difficult."

A Deeper Look
Agarwal notes that the goal of analyzing the EHR rollout in Montgomery County was to understand the process and challenges associated with implementation. The county has engaged in ongoing efforts to improve public health services by taking advantage of new HIT systems, an effort designed to support coordination across social, somatic, dental, and behavioral health services.

"The EHR was intended to provide greater visibility of patient information across service areas and more efficient communication and management of patient health and administrative information both internally and externally," Agarwal explains. "We used a variety of different data collection methods to understand the issues and opportunities that the EHR system implementation created."

Data were collected from 12 different facilities across DHHS and PCC. Methods included interviews (61 in total), observations (16), patient focus groups (three), surveys (55.5% overall response rate; 602 surveys completed) of EHR users before and after the EHR implementation, and client chart reviews (67); Agarwal indicates that this provided a rich qualitative record. Participating personnel included DHHS and PCC clinical providers, administrative and client services staff, and managers at multiple levels across worksites.

Agarwal says that the key finding from the analysis is that EHR system implementation causes significant organizational disruption and upheaval that must be carefully managed. "Users face significant roadblocks in learning how to incorporate the new systems into their clinical and administrative workflows, especially when some features may be cumbersome or nonintuitive," she explains, adding that the learning curve can be significant and may lead to user resistance if training and other support are inadequate. "Training should involve use of the system for the actual processes and tasks users will be conducting."

The analysis also revealed that the systems themselves are not designed optimally, especially for public health settings. "Public health departments have unique information management needs that commercial off-the-shelf EHR systems designed for general clinical practices do not always support effectively," Agarwal says. "In a public health department with diverse users and activities, it can be difficult to find the balance between uniform conformance and flexibility of system use."

Probst agrees, pointing out that EHRs were not built to support public health. "They were built to manage a hospital. In fact, their roots were built in billing if you really go back to when these systems were originally designed," he says. "Over the years, they have built in electronic medical records, but their roots were not in public health. In the day-to-day, that's not where their money is coming from. If you are an EMR vendor, you are not getting a lot from public health; you are getting a lot of money from organizations like Intermountain Healthcare."

Data sharing and interoperability remain a challenge for the public health sector, according to Chretien's essay. While scientists did share genetic sequences during the Ebola and Zika outbreak by using familiar databases such as GenBank, similar centralized databases that are widely accepted for other types of research data do not exist. The essay points out that "clinical trial data, for example, mostly reside in independent databases and are collected under various standards."

In addition to the industry standards challenge, health care and government organizations have struggled to make smart, informed determinations about IT investments for more than four decades, according to Agarwal. From a general perspective, she suggests that insufficient time is spent developing the business case for identified technology. "Quite simply, what problem or opportunity are we trying to solve or address, and if we do it right, what do we hope to achieve?" she offers as appropriate questions that should be asked prior to the decision to invest in HIT. "Sometimes executives and managers have misconceptions about the capabilities of the technology and also about what users want."

Agarwal says organizational and behavioral change present another barrier. Asserting that this area is as important as the technology rollout itself, she emphasizes that lackluster change management strategies can derail implementations.

Finally, there are usability and technical issues that must be addressed through early due diligence. For example, are the data needed for the system available? If not, what is the plan for capturing them? How interoperable is the new technology with other existing or planned systems? Is there a good testing strategy in place to ensure that the system does not produce bad data and that its business logic and rules are accurate? Is the technology easy to learn, intuitive to use, and a fit for the desired purposes?

Improving the Outlook
Industry professionals agree that while there are no simple fixes, the problems are not insurmountable. Probst believes that industry standards play a large role in advancing effective information sharing and response to public health crises across communities, regions, and the nation. "I feel we are making very superficial headway around interoperability," he says, pointing out that the technological format of patient information in one city or region is unlikely to match that in another. "We don't even have a common patient identifier. Public health systems needing to take in lab data and see what's happening with a population will find that it's nearly impossible to aggregate needed information."

Probst notes that there are success stories in the industry that can be applied to public health crises such as Zika or Ebola. For instance, industry leaders collaborated to develop data standards around childhood leukemia, and the results have been impactful. "Essentially, dozens of the leading pediatric hospitals decided to gather and share standardized data about pediatric oncology," he says. "Because of this, they were able to see treatments and the outcomes from specific protocols. This gave them the data to really sharpen the protocols for specific kids based on age, sex, type of leukemia, etc."

Until the larger challenges of interoperability are addressed on the industry level, Probst believes that focusing on small data sets for individual diseases—such as Zika or Ebola—may be an interim solution. "Someone needs to make it a priority, so [organizations like] Intermountain Healthcare, which provides 60% of health care in Utah, cooperates with all the other organizations," he says. "If we all did that, if someone took the leadership to do that around Zika, we could be in a position where those data are ubiquitous across the country in a matter of months."

However, Agarwal says limited resources and expertise present roadblocks to this kind of development. "It's important to recognize that becoming a tech-savvy public health system is a complex endeavor," she points out, adding that public health departments lack effective guidance regarding how to make informed decisions about IT investments that will deliver the desired return. "Public health departments are frequently underresourced, and they deliver a wide array of critical services. Fulfilling the public service mission while simultaneously dealing with the implementation of a complex system requires not only additional resources but also careful planning and execution."

In light of this reality, Agarwal and her team recommend an ongoing process of assessment-feedback-improvement. Her team has also developed a tool called the "Public Health IT Maturity Index" (PHIT Maturity Index) designed to fill this gap. A public health department can deploy the solution to measure how well it's using HIT to support its mission. Essentially, the PHIT Maturity Index defines the department's journey toward maturity.

Achieving the public health mission, Agarwal notes, typically involves a health department working to serve the community across 10 essential services, each of which has unique information and system requirements. To put perspective on these complexities, the PHIT Maturity Index helps health departments conceptualize these services in conjunction with the aspects of technology needed to accomplish them. For example, the tool provides guidance into the functionality needed to diagnose and investigate a community's health problems and hazards, including measurements of information timeliness, quality, and use.

The PHIT Maturity Index also can be leveraged to identify gaps in infrastructure investments from community partners or state systems. From a policymaking perspective, this information can help inform initiatives and programs to develop community PHIT capacity, Agarwal says.

Additionally, the PHIT Maturity Index measures human capital needs, helps gauge information quality dimensions, and assists in the identification of issues related to information accuracy, usefulness, and usability. For instance, while a system may be easy to learn, it may not be useful for the required tasks, suggesting changes to the PHIT feature set are necessary.

Looking Forward
Agarwal acknowledges that limited published research exists regarding public health EHR rollouts and that much more research is needed to document the technology's potential benefits and limitations. "There are no shining examples of public health EHR rollouts we have seen—not to say they don't exist, just that there has not been published findings about it," she says, pointing out that EHR adoption in public health departments has been quite low compared with health care systems. "For more than a decade, researchers have promoted how EHRs can improve syndromic surveillance efforts by increasing the breadth, detail, timeliness, and completeness of public health surveillance, and, anecdotally, it is working in some communities, but sadly, there are not many well-known rigorously studied examples."

Probst emphasizes the need for leadership, suggesting that public and private health entities should not rely on IT to lead the charge. "It needs to come from the clinical side or the government side," he says, pointing out that people don't necessarily follow IT but rather look to IT to implement a predefined solution. "We've seen it with EHR implementations today. The successful ones are led by clinicians because clinicians follow clinicians."

— Selena Chavis is a Florida-based freelance journalist whose writing appears regularly in various trade and consumer publications, covering everything from corporate and managerial topics to health care and travel.