Integrating specialty care can be key for high performance and financial success for ACOs, as the most complicated and costly patient care usually is managed by specialists. A large portion of the cost in our health care system is driven by specialists. Specialists are responsible for most medical and surgical procedures, involving both inpatient and outpatient care. Even for office visits, spending for specialists is higher than for primary care physicians. For example, while 45.7 percent of visits to office-based physicians in 2009 were to primary care physicians in general practice, family practice, internal medicine, or pediatrics, they accounted for only 30.4 percent of spending.

To understand the challenges and opportunities of integrating specialists, we spoke with eight health care leaders (1 CEO, 1 President, 3 Vice Presidents, 2 Medical Directors, and 1 Partner — 6 of these 8 respondents are medical doctors), during the summer and fall of 2015.  We targeted respondents from organizations across the country who are currently working in value-based arrangements in different geographic markets. They operate in various organizations including integrated delivery systems; health systems with Medicare ACOs, commercial ACOs, and organizations advising on ACO strategy.

We found that most of these health care leaders have developed strategies, which we discuss below, to help primary care physicians (PCPs) find and direct patients to specialists with practice patterns that are less resource intensive, although they’ve faced challenges with getting the data to do this. However, less work has been done to engage or provide incentives for specialists to actively change practice patterns to reduce costs and develop new care models for their patients.

While organizations are aware that they need to bring specialists into their care design, they are still at the beginning of this important effort. Thus far, most specialists in these arrangements have not had accountability for population health. Rather it’s been very limited to their specialty and not the overall care for the patients they are treating. As challenging as it may be, holding specialists accountable is essential for a next step.

Here is an overview of the insights we gleaned from our discussions with eight health care leaders, which we will discuss in more detail below. These insights are relevant to any organization operating or preparing to operate in a value-based care setting.

  1. Focus on both indirect and direct improvement levers.
  2. Engage primary care physicians in care process redesign.
  3. Start small and expand incrementally.
  4. Leverage data.
  5. Risk, payment, and incentives for specialists are the next frontier and likely a critical ingredient for full success.

Lesson 1: Focus On Both Indirect And Direct Improvement Levers

ACOs are changing primary care referrals and patient behavior first, focusing more on influencing rather than integrating specialty care at this early stage.

Influencing Primary Care Referrals And Care Management

Before trying to change incentives and practice patterns more fundamentally, most organizations we interviewed are focused on influencing referral patterns and care management around specialty care by encouraging primary care providers and patients to use specialists only when truly necessary and finding specialists who have less resource-intensive (i.e. keeping costs low) practice styles.

One of the ACOs had involved both primary and specialty care physicians at the start, but then quickly realized that it needed to focus on primary care to influence referral patterns. By compiling and sharing data with individual primary care providers or small provider groups, ACOs have been able to raise awareness of practice variation and generate discussion about leading practices and strategies to get there. Its next step was to separate primary care physicians from specialists and focus on building blocks – medical homes, clinical outreach, and population health.

Difficulties can arise in delineating responsibilities for specific actions in care processes. Many of these organizations wrestle with expectations of the specialist vs. the referring primary care physician, especially in complicated patient cases. They may also find that the power of primary care referrals in controlling spending may become more limited if the trend toward narrower provider networks continues.

Changing Patient Behavior

Some are putting energy into improving care management programs to guide patients in their use of follow-up specialty care services. This may be a promising strategy, particularly for ACOs participating in Medicare, where patients are free to self-refer to any physician.

Lesson 2: Engage Primary Care Physicians In Process Redesign

One ACO loops primary care physicians in with quarterly reports and holds periodic meetings to look for issues and gaps. Another respondent agreed with this as a useful strategy, and added that face-to-face interactions are best, including having discussions outside of larger meetings. Involving representatives from finance, IT, and analytics can also help the leadership team consider the complexities from all angles.

By including primary care physicians in process redesign, ACO medical leadership can achieve all three Triple Aim goals – not just reduced spending.

Lesson 3: Start Small And Expand Incrementally

The ACOs we interviewed are starting to address health care conditions where they can get traction for early success: areas where cost or quality is already of concern, care pathways are well defined, and accountability is relatively clear. Some organizations are prioritizing core service areas identified by Healthcare Effectiveness Data and Information Set (HEDIS) measures, working to ensure that patients across all providers receive the minimum of what is defined as high value/quality care.

Conditions involving high rates of hospitalization and readmission often top the list (e.g., cardiology, orthopedics) because of the potential savings. Some are also focusing on areas where employers are interested in savings.

The scale of early efforts to reduce spending and improve quality seems largely set by what is currently manageable given the organization’s size, geographic dispersion, and degree of interconnectedness through IT systems and care teams.

Starting small with a pilot approach has allowed some population health teams to become familiar with organizational nuances and work out unexpected kinks. One organization used a pilot focusing on the practice patterns around low-back pain, because of huge variation in cost and low patient satisfaction.  Once the physicians understood the “why” of the efforts, adoption of process and workflow improved. As physicians begin to see the benefits of their efforts, ACO leaders think they will build momentum, and they will be able to expand into other areas.

Lesson 4: Leverage Data

The organizations we spoke with are looking first at total rates and levels of service use for a small set of conditions, with a focus on identifying outliers, inconsistencies, and gaps. Sharing that “high level” data with physicians has prompted the physicians to want to dive deeper into the sources of differences.

One ACO did not have enough information to know where to start, but once they began comparing orthopedic surgical rates among provider practices in their system, they discovered a surprising level of variation. Further investigation suggested that differences in referral rates were the main source of the variation, so they were able to narrow the focus of their improvement efforts. They believe that they are developing better decision-making supports for physicians and for patients. This shows having sufficient detail on inpatient and outpatient data is needed to be able to identify the underlying source of variation.

Data Limitations and Challenges

Data are essential for guiding process redesign and behavior change, but several challenges are slowing progress:

  • Identifying and gaining consistency across specialty metrics: Organizations face the challenge of trying to meet a myriad of measurement requirements expected by payers. Accommodating competing demands and negotiating which metrics to use can be a real challenge. Organizations wishing to engage specialists should consider identifying the core set of measures most affected by those specialists and use them to identify the specialists to include in the ACO’s network. Selecting these measures should follow a systematic approach considering the organization’s strengths, desired results, and how to achieve population health aims most efficiently. For example, after culling through all metrics to measure outcomes for heart failure /stroke, an organization may determine the following to be most salient: cholesterol screening, hypertension control, and tobacco assessment.
  • Data integration: Organizations face difficulties pulling together data from disparate IT sources. Integrating claims and electronic health record data is a challenge. One organization mentioned that they have more than 10 different platforms, so they find it difficult to extract data. One of most challenging aspects of interoperability is integrating ambulatory data with inpatient data.
  • Data quality: No data are perfect, but improving the data organizations do have is a priority. Missing or erroneous data can erode physicians’ trust that measures of their performance are accurate and complete.
  • Benchmarking data: Organizations are making some headway in developing internal benchmarks for physicians within their system, but are facing challenges getting external data on practice patterns for specialists in their area or their state. As more data are collected across multiple providers over time, benchmarking data may become more readily available. Fair comparisons of performance require taking patient differences into account, so more may need to be done to ensure case mix adjustment approaches are adequate.

Lesson 5: The Next Frontier

ACOs are cautiously moving towards linking specialty care performance to risk, payment, and incentives.

Engaging specialists in the process of improving care delivery and population health should take place before payment can be tied to outcomes. That said, once payment is tied to outcomes, specialists will likely come to the table quickly to engage. Two key parts of the issue have already been discussed –greater agreement about standardized approaches to care and reporting structures that are trusted by both sides.

There can be other hurdles to engaging specialists in improving outcomes:

  • No skin in the game yet: At this point, many physicians are still operating within a fee-for-service model. While they may understand that there will be a “grand accounting” at some point down the line, they don’t currently have a tangible sense of how they will participate in the model.
  • Too much uncertainty for ACOs and physicians: Physicians have difficulty estimating the financial effect of participation, including whether and how the ACO will share gains and losses fairly. Negotiations are likely between the ACO and physicians, and among physicians. ACOs themselves may be unsure how to share risk with specialists: “Most [Track 1] ACOs do not believe that they have enough certainty about how to succeed in a shared-risk arrangement to progress to shared-risk payments now.”
  • Agreement on contract details: As noted earlier, it can be difficult for ACOs to reach agreement with physician groups on which metrics are to be used to gauge performance and what benchmarks must be achieved to trigger financial rewards such as shared savings or bonuses. While many organizations have extensive contracting experience, negotiating terms of this kind may be new.
  • Don’t have sufficient economies of scale to try new payment models (yet): As one CEO said, “We’re trying to reinvent care on a fee-for-service chassis. We would be happy to pay on something other than fee-for-service, but we don’t have the economies to do so.”

It can be hard to get started on efforts to integrate specialty care, because neither the “top” nor the “bottom” is in a strong position to (or motivated to) press for change. ACOs face a fundamental challenge at both the system level and the individual level. At the system level, leaders of the largest unit in the ACO – typically, a hospital – may seem best positioned to champion a value play because of their centrality and size in the system.

But, the problem is that part of the effort to increase value will likely involve reducing hospital spending, which will reduce revenues. With stability at risk, it’s not really a surprise that the biggest players are moving cautiously on integrating specialty care. The core challenge at the individual level is that many specialists’ livelihoods depend on the spending ACOs are trying to reduce.

Moving Forward – What Are The Next Steps?

Exhibit 1: Steps For Integration Efforts


Movement toward greater integration of specialty care into ACO models is likely to continue incrementally as the market shifts away from fee-for-service care and data, management, and leadership improvement. Momentum could pick up if more ACOs demonstrate financial success.

Authors’ Note:

We would like to thank Deloitte leadership for sharing their market experiences with us. Special thanks to Mitch Morris, MD; Bob Williams, MD; Es Nash, MD; Brian Flanigan, Bill Copeland, and Dorrie Guest. We would also like to thank Laura Eselius for her help on this research.