- Hospitals and health systems looking to reduce unnecessary spending by improving population health management have a difficult enough time getting any of their patients to adhere to chronic disease management programs, adopt healthier behaviors, and stay connected to the continuum of care. When those patients are also their employees, as is the case with self-insured healthcare systems, that daunting task has even more significant financial implications.
At Covenant Health, a not-for-profit Catholic health system with three hospitals in Maine and New Hampshire, becoming more effective as both an insurer and a service provider is a top priority. Despite being long-term EHR users with a certain level of data analytics savvy, the health system was struggling with outdated claims information that put truly meaningful population health management out of reach for their employees.
Retooling their infrastructure to embrace a more real-time approach to claims data analytics has significantly changed the way Covenant addresses preventative care, chronic disease management, and care coordination for their patients – and put them on track to save nearly $2 million by the end of 2015.
“Collectively, we have just over over 6,000 employees, and then, of course, we have their dependents. As a self-insured system, we're in the funny space of being the provider, the employer, and the insurer, which creates an interesting dynamic,” explained Richard Boehler, MD and CEO of St. Joseph Hospital, to HealthITAnalytics.com in an interview. St Joseph Hospital, with 208 beds, is Covenant’s Nashua, New Hampshire facility.
Due to the system’s multi-state structure, Covenant contracts with five different third-party administrators (TPAs) for their employee coverage, a set-up that can be challenging for those charged with identifying utilization patterns and ensuring that preventable spending is kept to a minimum.
“It’s not always easy to figure out what’s happening,” Boehler admits. “I recently had a meeting with our TPA in New Hampshire to look at what our experience was in 2014. So it's the middle of May in 2015, and under the traditional program, we're finding out what we should have been doing last January or February in terms of cost and utilization.”
“You can't be very effective at managing the care of employees if you don't understand your utilization patterns,” he added. “You can't have a good benefit design based on data that's a year and a half old. We have found that if we want to be effective at improving both wellness outcomes for our employees and reducing expense, that we need access to timely data that would give us the opportunity to influence change.”
Covenant decided to tackle some major plan redesign in an effort to drive waste out of their system, and started by extracting the data it needed to make better decisions regarding its employees’ care. “We recognized the need for a data warehouse that would be able to give us ready access to adjudicated claims for health insurance on the part of our employees,” Boehler said.
“And we needed a partner who could do this in a couple different states, dealing with a variety of different systems, so we chose MedeAnalytics. We have the clinical data, but we need to have the claims data on a timely basis so we could perform some risk stratification for our employees and help our care coordinators put in the effort to drive impactful change.”
That change is based on a two-pronged attack: using motivational wellness programs to promote lifelong behavioral changes and developing more immediate care coordination and patient engagement programs to ensure that patients with complex needs are getting the primary care necessary to prevent expensive and serious crisis events.
“We have a lot of traditional wellness programs in our organization,” Boehler said. “Those are great, but they represent more of a long-range approach. If I can engage somebody who has high blood pressure or who was just diagnosed with diabetes, we might prevent some problems five or ten years from now. That’s helpful, but you don't necessarily reap the benefits the minute that you start.”
“But when you do care coordination and population health management and more data-driven risk stratification, you are able to identify those individuals who are at risk of having a catastrophic problem,” he continued. “Or you're able to identify individuals who are utilizing emergency services frequently, or who aren't filling their prescriptions. They're not getting their blood tests for diabetes checked on a regular basis.”
“If you have access to data closer to real-time, you can intervene through care coordinators to be able to help them sooner, as opposed to down the road when you find out about it a year later from old insurance claims. So not only did we have to get all the data into a central repository, but we also had to find a way to get it back in a timely fashion to influence meaningful change.”
“It's meaningful to be able to identify which patients with diabetes have a hemoglobin A1C greater than eight, and understand what we can do to influence that,” he said. “It’s meaningful to identify the patients that have high cholesterol but aren't on a statin or some other approach to managing that condition. We can take actions to help those patients if we have the data.”
The ability to understand the needs and concerns of patients isn’t just a clinical issue, Boehler added. Because Covenant is an employer as well as a healthcare provider, maintaining the privacy of personal information is a double-barreled issue. Reassuring employees that their information is being used appropriately has been challenging.
“One of the biggest issues has been making sure that the employees understand that I'm not sitting here with unrestricted access to their personal health information,” he stated. “That's the last thing that I want. As an employer, that's not something that I want any access to, lest somebody think we might be using information about their health to influence their employment status or anything like that.”
“So we created a firewall that allows me to look at global information about utilization to help shape plan design and answer questions about things like increasing the number of services we do in-network to reduce costs,” he explained. “I use it at that broad level, as opposed to drilling down to an individual patient. I don't have that access. I don't want it. The only people who need that kind of detail are the care coordinators and primary care physicians who are caring for that individual.”
Care coordinators are integral to the health system’s population health management efforts, and help to forge stronger relationships between providers and patients who would benefit from sustained contact with their care teams.
“We have found that having care coordinators working on behalf of payers on the health plan level can play a certain role,” Boehler said, “but it's not nearly as effective as having somebody within your primary care practice who's got your clinical information and knows you personally to reach out and say, ‘For some reason, you haven't gotten your blood tested recently,’ or, ‘You haven't filled this prescription. Is there a problem? You were in the emergency department over the weekend. What's going on? Can we get you in to see your PCP soon?’ That level of involvement with the patient is really important.”
Providing care coordinators with close to real-time access to admission, discharge, and transfer (ADT) data has also helped patients to receive appropriate services after a hospitalization or a visit to the emergency department. “My care coordinators now get notified when somebody gets admitted, and they're able to get those patients into the primary care practice at the time of discharge for follow-up,” Boehler says. “That wasn't a structured process before. We're making sure that patients visit their PCP within seven days of discharge to reduce the probability of readmission.”
Those steps are part of the reason why Covenant Health is on track to save millions during their first year of plan redesign. Ensuring that patients seek the majority of their services from in-network providers is another way that the health system is cutting costs.
“When we looked at our utilization data, we found that a significant number of our employees were going out of network for services at a much greater cost to us. So we set up incentives in terms of more domestic utilization, said Boehler. “We created a stronger relationship with a tier-one provider to make sure that our employees would preferentially choose a high-quality, cost-effective organization for their care.”
“That’s been a big part of our savings,” he said, and is an important stepping stone for the healthcare industry as a whole to embrace value-based reimbursement and a higher level of financial responsibility for their actions.
“Our region, like many around the country, is rapidly evolving to a much greater level of risk-based provision of care. As a self-insured healthcare system, we’re taking on full risk,” he added. “But because we have a blend of different types of relationships with insurers and different incentives, we can learn how to accept financial risk for every patient. If we can do it well with our employees, we certainly have the capability to influence a commercial population, too.”
“Our care coordinators will tell you that it’s pretty impractical to do these things just for one subset of our patient population,” noted Boehler. “We don’t just say, ‘Well, this patient is an employee, therefore, we're going to provide this set of services. But this other person who's just commercially insured or has Medicare, we're going to do everything differently.’”
“I think better population health management programs improve care for all, because you really can't close your eyes to all the other patients who fall outside of some specific criteria. That’s not how you do medicine.”
As real-time, data-driven population health management becomes a core competency for healthcare organizations looking to capitalize on the rapid shift towards value-based reimbursement, self-insured healthcare organizations can no longer afford to make clinical or financial decisions based on year-old information. Understanding the technology options available to bolster real-time data analytics capabilities and seeking a qualified partner to develop more robust infrastructure will help healthcare organizations get ahead of the risk-based curve while providing optimal population health management services to employees and commercially insured patients alike.
“If you’re a self-insured healthcare system, you’re going to want to get an appraisal of your data capabilities,” Boehler suggests. “What kinds of information are you getting, and how timely are your third-party administrators? And you getting actionable information in a format that will allow you to intervene appropriately with patients?”
“That’s where the big difference is for us. Now I can respond more quickly to our needs, whereas before I was waiting a year and a half to get results. And I think that waiting a year and a half to get results is more characteristic of what organizations have been used to. That needs to change if we’re going to provide better care to our employees and our patients.”