Implementation best practices: Getting pop health off the ground
Increasingly, healthcare provider organizations across the country are launching population health management programs to improve health, enhance care delivery, trim costs and contend with the growing demands of value-based care.
But healthcare leaders cannot begin to get into population health without their CIOs and other health IT professionals effectively deploying information technology to support these programs. There’s a lot of EHR data that’s needed to successfully pull off population health, and analytics technology is key to making sense of it all.
Four experts in population health IT spoke with Healthcare IT News, offering comprehensive advice and best practices for launching systems that will support successful population health programs.
The full spectrum of health data
The success of a population health technology implementation rests on the underlying data foundation – if a healthcare organization has insufficient or low-quality data, the pop health program will fail, said Michael Gleeson, chief innovation and strategy officer at Arcadia.io, a population health and value-based care IT vendor.
“For most organizations, the full spectrum of data that will be required for you to be successful in value-based care is not easily aggregated nor is it all within your organization,” he explained. “The data will be in many different EHRs or databases within your own organization – and with other providers, health plans and health information exchanges.”
"For most organizations, the full spectrum of data that will be required for you to be successful in value-based care is not easily aggregated nor is it all within your organization."
Michael Gleeson, Arcadia.io
To be successful, a healthcare organization needs to map out exactly where all of the data is and then immediately start the process of acquiring access to those sources, he advised. As part of this, organizations need to get ahead of legal, policy and political challenges to make sure data can be aggregated.
“For example, you will need to work with your legal and security experts to understand how data sharing will comply with HIPAA requirements,” Gleeson stated. “In addition, you will need to work with other providers so they will permit you to see the data – this may require significant educational efforts to ensure they understand why you are requesting this data.”
Finally, a healthcare organization will need to determine what technical skills its team will need to access, extract and aggregate the data, he added. Without doing this, one will be paying for the technology and won’t have the data one needs to feed it, he said.
Bringing order to data chaos
Today, health systems are overwhelmed by data. There are data marts and data warehouses and data lakes and more. And there is still more data out there waiting to be aggregated.
“How to reconcile all the moving data flows?” asked Michael Barbouche, founder and CEO of Forward Health Group, a population health technology vendor. “By focusing on denominators, population health can begin to bring initial order to this chaos.”
Data from health plans – claims data – offers compelling insights on the patient care journey, the performance of the clinical network, and the behavior of the clinicians within the “four walls” of the health system, he said. Data from affiliated clinics – for example, an EHR, but not the health system’s EHR – can close the loop on required Centers for Medicare & Medicaid Services reporting measures or reveal a concerning decline in patient health status, he said.
“Marrying these external data sources with the health system’s EHR data provides a comprehensive view of all patients,” Barbouche explained. “But the marrying effort takes years of effort and enormous IT staff resources and several expensive staff augmentation consultants. That pathway is no longer valid for stressed CIOs.”
"Software vendors will often try a ‘dump and run’ software implementation style once a sale is completed. This suboptimal approach to implementation often fails."
Dr. Joseph Siemienczuk, Enli Health Intelligence
What to measure? The incentives that were just updated for the 14th time in the past 90 days still only impact 73 patients, he said.
“Focus on those 73, not on the universe of potential patients that one day might swap from this health plan to that government program,” he stated. “The denominator-centric approach delivers to CIOs a scalable, adaptable, flexible measurement solution. Others can work on the effort to boil oceans while the hard work of driving important changes in workflow is undertaken to deliver better outcomes for the 73 patients.”
Strategic planning is key
A critical consideration in the successful implementation and utilization of population health management IT involves strategic planning, said Dr. Joseph Siemienczuk, chief medical officer at Enli Health Intelligence, a population health IT vendor.
“Software vendors will often try a ‘dump and run’ software implementation style once a sale is completed,” he warned. “This suboptimal approach to implementation often fails. A better practice is to engage with the vendor in a formal and structured strategic planning exercise. This process should include organizational leadership at the highest level and operational leadership from all affected sectors.”
The healthcare organization’s goals, resources and desired outcomes should be discussed and documented, the doctor advised. This includes explicit implementation timelines, steps and success measures, he added.
“In addition, the software implementation must directly support identified goals while making optimal use of available and planned resources, with minimal impact to existing workflows,” he said. “Explicit success metrics should be defined prior to implementation, along with targets, benchmarks and clear accountability. Attendees should include the C-suite, with appropriate participation by accountable leaders in operations, care delivery and finance.”
Agreement on setting goals
Healthcare provider organizations implementing population health IT should first sit down and set clear, specific and measurable goals for what they want to accomplish with their investment in population health technology, said Gleeson of Arcadia.io, who agrees with Siemienczuk of Enli Health Intelligence on strategic planning.
“This should include ensuring it aligns with all value-based care contracts against which your practice is measured,” Gleeson said. “Having these goals will help you figure out the correct sequence in which to build and expand your data foundation, analytics capabilities and clinical workflow initiatives. As you assess your progress against your goals, you will be able to identify issues where you do not have enough high-quality data to fuel your initiative.”
"The demands on clinicians and care teams to fill out one more form, complete an additional assessment, or add on an extra measurement effort are taking an enormous toll on the healthcare workforce."
Michael Barbouche, Forward Health Group
Oftentimes, Arcadia.io sees healthcare organizations believe that HL7 or an HIE alone will solve their data exchange challenges. That is not the case, Gleeson insisted.
“Instead, there are varying levels of data quality, formats, volume limits and technical issues that must be worked through to establish the necessary data foundation,” he advised.
Everyone cannot use what they want
Another best practice for launching pop health IT: Use population health management technology to unify workflows, don’t just bolt it on, said population health IT vendor GSI Health Founder and CEO LeRoy Jones.
“When you’re building an interdisciplinary network and each clinical, behavioral health and community-based organization is bringing their own technology to the table, it may seem to be the path of least resistance to let each organization use their own tools,” Jones said. “But simply bolting together information from different tools, contexts, sets of assumptions and data models requires significant work to interface across different technologies because the semantic meaning of data doesn’t translate neatly across platforms.”
More important, an “everyone use what they want” approach doesn’t provide a full and accurate picture of shared patients or the population as a whole and fosters inconsistencies in applying a coordinated care model, he added.
“When everyone follows an independent model with different workflows and tool capabilities, there is no shared accountability for following the planned population health management approach, making it harder for you to achieve your outcome and financial goals,” he explained.
Instead, look for a technology that provides a common platform for collaborative workflows across diverse teams, with configurable, program-specific actions to drive care management consistently across organizations, he advised.
"We’ve seen entire programs stall when it becomes time for implementation because somebody doesn’t want a neighboring organization they’re collaborating with to have full and open access to patient lists and records, and the tool they’d chosen wasn’t able to adequately reflect each organization’s privacy needs."
LeRoy Jones, GSI Health
“This approach will provide insights about how issues addressed by other organizations impact those you collectively serve, enabling the unification of workflows and a more efficient orchestration of care across your community,” Jones said. “All stakeholders can work from a single source of truth about a population, bringing a unity of purpose around how they jointly manage each patient and the population as a whole so each provider can deliver more informed care.”
Plus, the healthcare organization will have a much higher level of confidence in how it manages outcomes, makes course corrections and prescribes new courses of action, he added.
Getting physician engagement
An often overlooked best practice when implementing population health IT is physician engagement, said Siemienczuk of Enli Health Intelligence. Few, if any, technologies can truly succeed in a healthcare organization without the buy-in of the providers, he added.
“Begin by ensuring physicians agree that a solution is needed, and that the selected option is a good one,” he advised. “Critically, early physician involvement in the software selection process leads to stronger alignment and better utilization down the road. The most efficient and effective path is to engage physician stakeholders at the leadership level.”
The physician leadership approach varies by type of organization and includes political – for example, hospital medical staff – and employment organizational structures, he said.
“Optimal outcomes are enhanced by physician leaders who have excellent communication skills and who solicit input from all affected providers,” he explained. “Each physician must feel they at least had the opportunity to share their perspectives. Communication options include regular physician meetings, town halls and surveys.”
Durable physician buy-in is only assured if the pop health IT is consistent with prevailing and efficient care team workflows, Siemienczuk added. If a technology will be used in the care delivery process, especially by physicians, it should be integrated with the EHR platform to minimize disruption of workflows, he said.
Delivering value from the EHR
Once caregivers get into the population health management program, the one resource that is most under stress remains time, said Barbouche of Forward Health Group.
“The demands on clinicians and care teams to fill out one more form, complete an additional assessment, or add on an extra measurement effort are taking an enormous toll on the healthcare workforce,” he said. “The goal of any population health effort must be to deliver value from the existing electronic health record.”
Put more directly – population health technology is about delivering the ROI of the EHR environment, not creating an additional platform that competes for more of that scarce time, he said. Effective population health can be defined as the dovetailing of incumbent health IT platforms with rich insights about patient cohorts, gaps in care and opportunities to drive financial improvement, he said.
“The opportunity of population health begins and ends with helping care teams build, improve, enhance workflows in the EHR platform,” Barbouche advised. “No matter the patient acuity, no matter the disease type, population health is about delivering the right action step to the right actor at the right time in the EHR workflow.”
For an overscheduled physician, that means a comprehensive summary of her entire patient panel organized around the clinical, financial and operational outcomes that she and her team are trying to improve, he said.
“For her team, population health is about delivering efficient pieces of intelligence, whether in bulk – for example, for all our CHF patients, every patient that has not had an ejection fraction – or at the individual patient level – for example, has the plurality of new medications in the last three months had an impact on the patient’s blood sugar, blood pressure and creatinine,” he stated.
When deployed, population health IT can serve as a barometer for the pressure that a delivery organization is facing in its market, he added. When successful, population health technology can relieve the pressure and help deliver improved outcomes, he said.
Robust security/privacy controls
On another front, a pop health IT launch best practice is to seek highly flexible and granular security controls, said GSI Health’s Jones.
“If your population health management solution is going to successfully convene diverse and quasi-competitive stakeholders across the community to jointly manage patients, it is critical to have robust security/privacy controls in place that allow for very specific role-, organization- and consent-based access to information,” Jones advised.
Security concerns with interdisciplinary teams extend beyond HIPAA into the dynamics of competitive organizations working together to manage populations, he added.
“The flexibility to construct intricate data protection schemes is essential to building a level of trust across your network so everybody is comfortable with how the information flow is used and authorized,” said Jones. “We’ve seen entire programs stall when it becomes time for implementation because somebody doesn’t want a neighboring organization they’re collaborating with to have full and open access to patient lists and records, and the tool they’d chosen wasn’t able to adequately reflect each organization’s privacy needs.”
A technology that provides a high degree of granularity and flexibility in how it protects information enables a healthcare organization to put its data security concerns aside and focus on the partners and workflows that will most benefit its patients, he added.
“If organizations are confident they can rein things in as necessary through technology-enabled controls, they are more willing to put aside their concerns about patient churn and leaky networks, and share more data when designing workflows,” Jones concluded. “The result is a more robust and collaborative population health management program that ultimately improves how care is delivered to your patients.”
Health IT implementation best practices
This 20-feature series examines in-depth what it takes to deploy today's most necessary technology and tools.