Editor’s note: In addition to Shubham Singhal (linked bio above) this post is authored by Rohit Kumar and Jeris Stueland. Rohit Kumar is a consultant in McKinsey’s Chicago office. Jeris Stueland, an expert in McKinsey’s Healthcare Systems and Services Practice, is also in the Chicago office.
In recent years, much of the payor industry has transitioned away from a medically underwritten model to a guaranteed-issue, community-rated, risk-adjusted model. As a result, managing the total cost of care has become the dominant imperative for achieving competitive advantage.
As payors have sought ways to develop effective managed-care approaches for this new environment, interest in vertical integration has increased considerably. In the four years between 2005 and 2008, payors announced just 12 vertical integration M&A deals. In the subsequent four years, the number jumped to 26, and recent targets have largely been physician groups and integrated care organizations. These deals have not just been attempts to create competitive advantage—they have also been defensive plays to counteract potential challenges from provider consolidation and the acquisition of physician practices by hospital systems interested in vertical integration of their own.
Our research suggests that the economics of vertical integration makes sense for payors in only a minority of markets. For example, when we identified the markets in which the acquisition of physician groups appears to create economic value for payors, the total population included only about 80 million Americans. Furthermore, the significant execution challenges involved in integrating payors and physician practices at scale suggests that the markets we identified likely represent the outer limit of the feasible set.
We describe below the economic drivers of net value creation (or destruction) through vertical integration, the market conditions that indicate a given area may be fertile ground for positive value creation, and the execution challenges that must be overcome to capture the value.
The Economics Of Vertical Integration
The economic drivers of vertical integration for a physician practice must be analyzed on a granular level for each line of business that might benefit significantly from such an acquisition—typically, Medicare Advantage or another risk-adjusted line. A separate granular analysis must also be conducted for each type of provider being targeted.
We have found that any acquisition of a physician practice results in three major types of incremental cost, each of which is necessary if the desired outcomes and potential cost savings of the integrated system are to be achieved:
- a decrease in physician productivity once they are employed, coupled with the cost of the performance incentives required to align their behavior;
- the additional staff support needed to enable better risk coding, case management, referral management, etc.;
- and the cost of capital for investments in technology, data, analytics, and physical infrastructure.
The associated costs would be quite different if the potential target were a hospital system or pharmacy chain, but the principle remains the same: Only once all these costs have been calculated can a payor determine whether the potential savings might outweigh them.
The positive value created for a payor from integration with a physician practice comes from better revenue and medical-cost management. For example, the drivers of value under Medicare Advantage could include the incremental revenue gained from better risk-coding accuracy and higher star quality ratings, plus the decrease in medical costs that would result from better population health management and the steerage of members to appropriate hospitals and procedures (Exhibit 1).
Market Conditions That Enable Value Creation
In our experience, the critical yardstick for whether vertical integration is likely to create value is the density of potential value in each local market—in other words, the sum of potential incremental revenue upside and avoidable medical costs per square mile. Value density depends on a handful of market characteristics, including risk-score density, the level of compressible costs, and whether the market structure is conducive to steerage. (Note: Other factors that influence value density include the availability of physician groups that can be acquired and the acquisition price (driven by competition) to acquire such practices; both of these factors can vary greatly across markets. For this post, we did not analyze those factors in each local market.)
Risk-score density is the prevalence of health conditions, especially chronic illnesses, in the market. Because the effective management of chronic health conditions can often prevent their progression to more complicated acute events that require more expensive treatment (usually hospitalization), it provides the greatest opportunity for value creation through reduction in medical costs. However, risk-score density is highly variable across the United States. (Note: In our research, we used the Dartmouth Atlas’s hospital service areas, or HSAs, as our geographical unit. The Dartmouth Atlas defines an HSA as “a collection of ZIP codes whose residents receive most of their hospitalizations from the hospitals in that area.”)
The second important characteristic is the proportion of costs in a market that are compressible. These are costs that can be mitigated through clinical excellence, which typically includes reduced clinical variability, shorter average length of stay, greater use of evidence-based treatment protocols, and the substitution (when appropriate) of generic for branded medications, among other things. The result is generally lower use of clinical resources and higher productivity. One indicator that provides a good sense of the extent of compressibility of costs across markets is the difference in hospitalization rates for ambulatory care-sensitive conditions. Higher rates often indicate greater compressibility. We normally recommend that a payor repeat this analysis using multiple other granular factors to ensure that it has assessed all the major elements of cost compressibility, but in our experience hospitalization for ambulatory care-sensitive conditions can suffice for a broad market scan.
Finally, payors must determine whether the market structure is conducive to steerage. The presence of excess capacity (as measured, for example, by hospital beds per thousand), adequate hospital system choice, and independent physician practices that make it possible to shift referral patterns to higher-quality, lower-cost hospitals are some of the factors in each market that suggest that managing costs through owned physician practices is likely to be feasible.
Only by analyzing these characteristics in all local markets across the United States is it possible to identify the markets that are the best candidates for vertical integration (Exhibit 2). Once we did this, we found that vertical integration was likely to create value in less than one-third of all markets (Exhibit 3). We believe that in the remaining markets, payors will therefore need to deploy additional approaches, such as virtual integration, pay-for-value programs, and improved network contracting methods. Although these approaches are potentially less effective in creating upside value, they are likely to be less costly as well and so more likely to create net positive value in the remaining markets.
Execution Challenges That Must Be Overcome
Three additional factors should be considered carefully in each market before a payor goes ahead with a vertical integration deal. Each of these factors can present challenges that must be overcome before the deal can succeed.
The first is how willing providers are to align with a payor. Surveys we conducted in late 2011 revealed that most physicians did not view working for an insurance company as an attractive option (Exhibit 4). However, physician attitudes toward employment—by payors or others—is likely to vary both over time (as the benefits and costs of independence evolve) and by region (based on whether other options, such as independent physician associations, are present or absent). In our survey, even the physicians who did not view employment favorably mostly thought it was likely that they would become employees within the next few years.
The second factor that must be considered is how probable it is that the payor can change practice patterns. In our survey, many physicians indicated that they were willing to alter their practices to reduce health care waste, but most of them thought they had only a low ability to influence the amount of waste (Exhibit 5).
Furthermore, it is not yet clear whether a physician’s stated willingness to change really translates to a willingness to follow a payor’s new protocols. Remember: employment does not ensure alignment. To maximize the likelihood that it can alter practice patterns, a payor should take into account all the factors that encourage physicians to change (Exhibit 6). For example, it should include respected physicians on its board and in its C-suite, and make certain that its medical director is both clinically credible and business savvy. It should also have a streamlined IT system and effective data analytics capabilities that make clinical and claims-data sharing and communication back to physicians easy. In addition, the payor should offer physicians incentives that are strong enough to motivate real change and provide them with training and support to help them improve care delivery.
Finally, the payor must appraise—realistically—whether it has the capabilities and capacity needed to ensure a successful deal execution and subsequent integration effort. (Note: More details about the capabilities needed for deal execution can be found in the July 2013 white paper by Celia Huber et al, “Riding the next wave of healthcare payor M&A.” For a copy of this paper, contact Karen_Harden@mckinsey.com.)
The integration effort should focus on six key dimensions: strong physician alignment that enables steerage to network hospitals and medical facilities, a full suite of clinical and medical management programs, tracking and management of star ratings, compliance/CMS reimbursement management, training and incentives for support staff, and back-end support systems (especially integrated IT, data and analytics).