Healthcare leaders are committed to population health management, yet are cautiously moving forward to implement risk-based financial models.
This article first appeared in the October 2016 issue of HealthLeaders magazine.
Population health management continues to be a work in progress for the healthcare industry, with the majority of providers involved in investigating risk-based financial models, exploring strategic initiatives, and investing in patient engagement and IT infrastructure. In fact, a significant number of healthcare organizations appear to be doing their homework in this area: A combined 76% of respondents in our survey say that they are either fully committed and underway or have an experimental or pilot program underway for managing the overall health of a defined population. Another 11% of respondents are planning to pursue population health but have not begun, and 9% are examining how or whether to pursue this type of health management. These responses are indications that population health is top of mind for healthcare leaders.
However, while many providers are engaged in evaluating the feasibility of population health programs and some have initiated programs, few have managed to reap meaningful financial rewards because most are still in the early stages of implementation. One indicator of this modest financial progress is net patient revenue from risk-based population health management activities that have exposure to profit and loss: Nearly half of respondents (48%) say they have risk-based revenue of less than 10% (some have no risk-based revenue at all).
The adoption of risk-based financial models is an essential aspect of population health, and it is not to be taken lightly. Healthcare providers that take on exposure to financial risk face real consequences if they misjudge their capabilities or lack insight into the quality of their risk pool. And it is for these reasons and more that many providers are taking a cautious, steady-as-she-goes approach.
Managing health of a defined population
As stated previously, approximately three-quarters (76%) of respondents say that they are either fully committed and underway or have an experimental or pilot program underway for managing the overall health of a defined population, up seven percentage points over last year's survey. The results are higher this year for fully committed and underway (47% versus 41%) and basically flat for experimental or pilot programs (29% versus 28%). Only 2% say they do not plan to pursue this, an indication that population health has crossed a tipping point.
Perhaps because of the complexity of the undertaking and the scale of the investment, there is a correlation between organizational size and respondents who indicate they are fully committed and underway with population health management. For example, a greater share of health systems (56%) than hospitals (43%) and physician organizations (38%) say they are fully committed and underway, and based on net patient revenue, a greater share of large organizations (59%) than medium (46%) and small organizations (42%) is fully committed. This correlation also holds true for number of beds, number of sites, and number of physicians.
Bertram Scott is senior vice president of population health at Novant Health, a nonprofit integrated healthcare network with 2,585 licensed beds, 14 medical centers, and more than 1,380 physicians in 530 locations as well as numerous outpatient services, headquartered in Winston-Salem, North Carolina, and the lead advisor for this Intelligence Report. He agrees that the industry has likely passed a tipping point for population health.
"I think there is an increasing realization by physicians and hospital systems that population health is going to be inevitable, and they want to try to get ahead of the curve and get some learning in," says Scott. "I believe that this is also driving the need and desire for them to begin thinking about risk contracts. You don't want to have the switch turned on two years from now and you've had no experience in that space, right?
"I would also say the impetus for that tipping point is the coming together of several things at one time. For example, the most recent MACRA changes—the Medicare changes to physician reimbursement—are going to be a huge catalyst for change."
Revenue and compensation
Net patient revenue from risk-based financial models is one of the indicators that separates basic forms of population health management from the more robust versions. As such, it serves as a barometer for the level of progress providers are making in embracing risk in their organizations.
Interestingly, while survey results indicate that participation in population health programs has increased since last year's survey, the percent of net patient revenue attributed to risk-based population health management activities has remained fairly flat. Twenty-two percent of respondents say that 25% or more of their organization's net patient revenue is attributed to risk-based population health management activities, down from 27% in last year's survey.
Further, nearly half (48%) say this activity is less than 10% of net patient revenue, which includes 11% that report no at-risk revenue of this type. This response is up six percentage points from last year's survey (42%). The results indicate that respondents are taking a cautious approach when it comes to risk-based population health activities.
Another barometer of population health adoption is the extent to which employed physician compensation is at risk for quality-based outcomes. Responses indicate that only 16% of respondents say that 100% of their employed physicians have at least some portion of their compensation at risk for quality-based outcomes; more than one-third (38%) have between 50% and 100% of employed physicians with at least some portion of their compensation at risk. On the other hand, 31% of respondents report having no compensation at risk for quality-based outcomes.
"In my experience, most physicians believe in the quality element of this and they want to participate," says Scott. "If they are improving outcomes for patients and for large swaths of populations, they would like those dollars to increase for them versus the payers."
Strategic initiatives
The responses for strategic initiatives that healthcare organizations are engaged in or exploring to improve the health of a defined population are nearly identical to last year's survey, an indication that the scale of population health activities is likely incompatible with sudden changes in direction. For example, respondents indicate that the top three strategic initiatives are clinically integrated networks (63%, down one percentage point), patient-centered medical home–related activities (57%, identical to last year), and alliance of providers (45%, down four percentage points).
One exception is the response for merger with or acquisition of providers (31%), which is 11 percentage points lower than last year's result. This suggests that, while the industry is currently experiencing heavy merger and acquisition activity, population health is not necessarily the biggest driver.
Financial risk structures
Survey results for financial risk structures that organizations currently use in caring for an identified population continue to show a reluctance to assuming downside risk on the part of respondents. Bundled payments and shared savings programs with payers are tied at 43% as the top financial risk structures mentioned by respondents by a large margin; they are followed by capitation (23%) and direct contracting with employers (22%). Note that results are relatively comparable to last year's survey, except that the response for bundled payments increased nine percentage points, which moved it into a tie for the top position.
The response for shared profit and loss arrangements with payers (18%) shows little change from last year's survey (17%), an indication that respondents are in no hurry to adopt risk-sharing financial models with downside risk.
Another risk structure that receives an equal level of response is owning insurance companies (18%), which is down four percentage points over last year, where it was in the top three for financial risk structures. According to Scott, many providers think about having their own insurance company as a way of participating more actively in risk-sharing, but they may be overlooking some of the difficulties.
"It is not for the faint of heart. It is a difficult business. I think sometimes providers believe that it looks pretty easy from the outside. But it is a difficult thing to do—to govern and police yourself when you're that close to the action. Because if you think about it, you're both the manufacturer as well as the distributor of the product."
Looking ahead three years, respondents are fairly bullish about their tolerance for risk. They expect that bundled payments (61%) and shared savings programs with payers (61%) will remain the top financial risk structures their organizations use in caring for an identified population, and they expect direct contracting with employers (41%) to also play an important part in their financial risk structure.
Significantly, capitation (32%) is no longer in the top three responses. Shared profit and loss arrangements with payers (36%) increases to the fourth position on the list of responses, an indication that respondents expect to gradually take on more financial risk in the next three years. Another indicator of respondent interest in risk are the results for none (6%), down 12 percentage points from what financial risk structures organizations use in caring for an identified population now.
Biggest barriers to population health
Respondents indicate that the top three barriers to successfully deploying population health programs are up-front funding for care management, IT, and infrastructure (42%), engaging patients in their own care (39%), and aligning independent physicians/providers (38%). Responses for financial risk assessment capabilities (36%) and getting meaningful data into providers' hands (34%) follow closely behind.
Making up-front investments in infrastructure while facing uncertainty about risk-based reimbursement models requires providers to make a certain leap of faith that the numbers will work out in the long term. This explains why so many are investing in analytics, with the hope that analytics will generate meaningful data to guide strategy.
According to respondents, the top IT infrastructure capability investments that are directed toward population health management are analytics using payer claims data (62%), analytics using population data (58%), patient registries (56%), and data warehouses (55%). Respondent interest in analytics is increasing; in last year's survey, analytics areas held the third, fourth, and fifth positions for responses. This year, they occupy the first, second, and fifth positions.
Within three years, respondents indicate that the top five investment areas fall in a narrow range between 63% and 76%. The top IT infrastructure capability investments directed toward population health management are expected to be analytics using population data (76%), analytics using payer claims data (71%), patient registries (67%), analytics to identify gaps in care (64%), and data warehouses (63%).
Respondent expectations for future IT infrastructure capability investments show the greatest increase compared with the results for now for analytics using population data (up 18 points), analytics to identify gaps in care (up 15 points), and risk stratification (up 15 points).
As mentioned above, patient engagement received the second-highest response for barriers to population health. Note that the top three patient engagement areas in which respondents say their organizations are investing in population health are patient portals (83%), wellness- or condition-related outreach programs (64%), and patient access to medical records (64%). Responses for all engagement areas show a moderate to high level of investment interest and support, with no area falling below 31%.
While traditional approaches receive the highest responses, large organizations in particular are looking to technology to enhance their patient engagement efforts. For example, based on net patient revenue, a greater share of large organizations than medium and small is investing in technology-supported techniques such as telemedicine for clinician-patient consults (80%, 53%, and 44%, respectively), social media (59%, 43%, and 41%, respectively), remote monitoring (57%, 28%, and 22%, respectively), and telehealth to track patient health status (54%, 32%, and 21%, respectively).
"Patient engagement is hugely critical because patients' out-of-pocket expense over these last few years has increased," says Scott. "So they want to be much more active consumers, but we have to find ways to get patients engaged in decision-making—decision-making about their own health, and how they want to gather, collect, and discuss their own information.
"It's improving dramatically, but providers are still more inclined to tell them how to receive stuff versus listening to how they want to engage with us and then reacting."
Delivery of care redesign
Care redesign is an important element of population health management, and the majority of respondents are engaged in some form of this activity. Care management with risk-based patient panels (55%) receives the highest response, followed by systems to identify gaps in care (50%) and team-based care in patient-centered medical homes (50%) in a tie, clinical programs organized by disease state (48%), and care goals, incentives aligned across the continuum (44%).
Provider financial resources appears to play a role in care redesign. For example, based on net patient revenue, a greater share of large organizations than medium and small has redesigned the delivery of care for all areas mentioned in the survey. The results are: care management with risk-based patient panels (71%, 62%, and 48%), systems to identify gaps in care (61%, 52%, and 45%), team-based care in patient-centered medical homes (69%, 48%, and 43%), clinical programs organized by disease state status (63%, 52%, and 40%), care goals, incentives aligned across the continuum (54%, 40%, and 45%), and care registries organized by disease state (44%, 40%, and 26%), respectively.
As with the results for care redesign now, the responses for the top five delivery-of-care areas intended to support population health management in three years are clustered in a narrow range. Care management with risk-based patient panels (67%) receives the highest response, followed by systems to identify gaps in care (65%), care goals, incentives aligned across the continuum (63%) and clinical programs organized by disease state (63%) in a tie, and team-based care in patient-centered medical homes (61%).
While the results are clustered in a tight group, certain areas had large increases in response rate compared with the results for what areas organizations have redesigned delivery of care with the intent of supporting population health now. For example, the top three increases are found for care goals, incentives aligned across the continuum (up 19 percentage points), with systems to identify gaps in care (up 15 percentage points) and clinical programs organized by disease state (up 15 percentage points) in a tie.
The strong response rates and relative parity among care redesign areas indicates that respondents are considering all options when it comes to population health. Because there is uncertainty as to the best tactics, there is an opportunity for analytics to help focus activities going forward.
Triple aim goals
Another method for evaluating the current state of population health competency is to look at respondent strength on its triple aim goals. Respondents indicate that their level of strength is stronger for improving patient experience of care (85% very strong or somewhat strong, and only 15% very weak or somewhat weak) than improving the health of populations (66% very strong or somewhat strong, and 35% very weak or somewhat weak), and reducing the per capita cost of care (54% very strong or somewhat strong, and 46% very weak or somewhat weak).
The results for improving the patient experience of care are not surprising given the attention that providers and the Centers for Medicare & Medicaid Services have given patient experience in recent years, and the results for improving the health of populations also indicate that positive progress is being made. This is particularly true given that the population health trend is still in its early stages. However, results for reducing the per capita cost of care indicate that lowering costs remains a challenge for many providers, and that there is still work to be done.
Scott has some general guidelines for providers to consider as they take on population health. "One, get your staffing right. Go identify and bring in the resources you need to be successful at this. Don't presume that people in your organization have the right skill sets to do this work. You wouldn't go out and begin to do surgery and not bring in a surgeon, right?
"Number two, be really clear about what you're trying to achieve. When you say 'population health,' what exactly do you mean? Because there's a lot of things that could go under that umbrella. The third thing is take on risk at the same rate that you build the infrastructure to manage risk. This requires behavioral change by many and particularly the physicians. Make sure that when you take on risk, that infrastructure and physician change is analyzed, because it's big.
"And go talk to people who have done this and have done it successfully. Don't exclude people because they are either a competitor or they're a payer and you've had a bad relationship. This is the time when we should be trying to learn as much as we can, and everybody's a resource for that."
Jonathan Bees is a research analyst for HealthLeaders.