Tiered Networks: Strategies of Engagement for Patients and Physicians

February 5, 2010
Margaret E. O'Kane
Margaret E. O'Kane

Volume 16, Issue 2

Tiering of networks cannot be viewed in isolation, but rather as a piece of a larger value agenda.

Anna Sinaiko and Meredith Rosenthal1 produced a useful analysis of the way health plan members reacted to an effort by the Massachusetts Group Insurance Commission (GIC) to experiment with tiered provider networks, offering members lower copayments for selecting physicians who met certain quality and utilization criteria. The experience offers some encouragement for advocates of tiering networks; it also points to a number of issues that must be addressed to make tiering a stronger means of improving quality and affordability. Finally, tiering of networks cannot be viewed in isolation, but rather as a piece of a larger value agenda.

Tiered pharmacy benefits are one of the few successful costcontainment strategies in the past 2 decades. Evidence indicates success on 2 fronts: (1) a positive impact on pharmacy costs and (2) acceptance by members that their choices will affect their own out-of-pocket costs. Against the background of other health plan cost-containment strategies, some of which were subsequently thwarted by public policy and others of which suffered from poor execution and/or lack of provider and patient acceptance, drug tiering stands out as a singular success. It is both ethical and logical to structure choices so that costs are more transparent to consumers and to ask them to bear at least part of the differential cost for a comparable but more expensive product, as long as individuals are not forced to accept lower quality in the favored product. As Americans, we are used to the idea that if we choose a more expensive product, we pay more for it. Third-party payment interrupts the connection between purchasing choices and resultant costs to the individual, but the idea of reference pricing (the third-party payer pays for an equivalent choice at the lowest price, and if the individual chooses a more expensive version, he or she pays the difference) is familiar, and well accepted by customers, in scenarios like auto insurance.


The insurance marketplace has overwhelmingly accepted the concept of a tiered product—the preferred provider organization (PPO). However, the PPO always has been handicapped by an excessively simple design: the absence of quality criteria and the basis of cost criteria for network inclusion. Network entry is determined by a willingness to accept a discounted fee-for-service payment. Although network participants may look economical from this “piecemeal” perspective, they often practice in ways that make them anything but efficient from a total cost perspective. As a result, traditional PPO design has failed to demonstrate any meaningful impact on costs. Consumer-directed models of PPOs have succeeded in reducing costs to third-party payers in the short term, mainly through a shift of costs to consumers in the form of a high deductible.

The GIC project moves beyond the traditional PPO to a network design based on quality criteria as well as episode-based costs. By design, all GIC health plans were required to participate in the tiering initiative. This allowed for several advantages often absent from early tiering experiments: first, all the plans used the same quality and utilization metrics; second, because the collective membership of all plans was so large, economic signals were more apt to transmit beyond the plans to the delivery system; and third, because all options were tiered, plan selection and churning of enrollment were less likely.

Commentary on Findings

As the authors note, the difference in copayment between tier 1 and tier 2 physicians was modest across all plans. The gradual introduction of this new network design is a reasonable change management strategy, although it presumably limited the economic impact of the tiering (which this article does not address). Only about half of the enrollees (49.5%) were aware of their health plan’s tiered network, in spite of vigorous efforts on the part of the GIC and its associated plans to educate them beforehand and to inform them of the new network design at the point of introduction. The lack of awareness among enrollees is likely linked to the small gradient between the preferred tier and tier 2. There also were very few complaints about tiering from members.

However, a less positive set of findings involved trust: 35.5% of enrollees did not trust the tiers, and 22.5% did not know if they trusted the tiers to tell them which doctors were better than others. It is possible that some of the lack of trust relates to the novelty of the plan design. If this is the case, trust could be enhanced through experience and by continuing to inform patients of how tier 1 doctors are selected—based on cost and quality, already understood by 59.3% of respondents. Encouragingly, half of respondents found the information about their physician’s tier very or moderately important to their decision to see that physician. If they received the information before their first visit, they tended to rate the information as even more important. The authors rightly comment that the experience with tiering in the GIC project is limited because of the short duration of the implementation.

Physician Strategy

The GIC project focused heavily on specialists, with most plans declining to tier primary care physicians. This seems sensible, because specialists and the costs they generate constitute a large proportion of spending for health services. Moreover, primary care physicians tend to be in short supply, so tiering might effectively force patients into using tier 2 physicians because of lack of access to those in tier 1. The strategy seems to focus on minimizing disruption while focusing on the highest opportunities to drive quality and reduce costs through network selection.

The authors speculate that the intrinsic motivation of providers may be a source of change in the future. However, I would go beyond this and suggest that a strategy enlisting the cooperation of willing physicians might generate a less controversial path to cost savings. As a first step, plans could incentivize and reward primary care physicians for referrals to tier 1 specialists and provide feedback to specialists who find themselves in tier 2 on steps they could take to qualify for tier 1 status. Admittedly, it would be folly to build a strategy on the assumption that all physicians would be willing to cooperate to achieve high-quality, affordable healthcare, but perhaps it is worth the effort to enlist those physicians who are willing to be part of the solution.

An important limitation of the focus on individual physicians is the fact that the cumulative behavior of the constellation of providers who treat a patient can give rise to considerable inefficiency. Although each team member may be well within group norms for utilization and quality, many opportunities for true efficiency will present themselves in handoffs, transitions, reorganization of the care model, and other improvements that exist only at the collective level.

Patient Engagement

A more vigorous strategy of patient engagement should be helpful in taking programs like this one to their true potential. This area of research is not well understood, and efforts to engage patients often have focused more on patients’ own behavior than on their care-seeking patterns. An ambitious program of patient education coupled with incentives should lead to more progress toward value.

Looking to the Future

In July 2009, the Massachusetts Special Commission on the Health Care Payment System recommended that the state move to a system of global payments for state providers. If achieved, this model holds great promise for effecting a major change in provider behavior. Unfettered by the visit- and procedure-based reimbursement system, providers will be free to reorganize care to take advantage of technologies like e-mail, telemedicine, and patient self-management to achieve efficiencies that are not possible today. Yet we cannot expect the “discipline of the market” to bring about these transformations without a way of recruiting enrollees to be part of the needed changes. Efforts at tiering networks today can create the necessary consumer awareness and understanding only if consumer choices and incentives can be structured to drive value in years to come.

Overall, though, national healthcare reform proposals hold little apparent potential to drive a sustained and serious value agenda. Assuming that this situation is unlikely to change, enrollees will bear an increasing share of coverage costs—and engaging them in the drive to achieve quality, affordable healthcare for all Americans is urgent. Consumer cost-sharing can be used either as a blunt instrument or as a means of strategic buying. An ethical approach to the cold reality of this cost-sharing must ensure that consumers can purchase quality, affordable healthcare within the top-performing provider tier. This approach will both provide patients with affordable choices and send a signal to the delivery system that quality and value are the new path to success.

Author Affiliation: From the National Committee for Quality Assurance, Washington, DC.

Funding Source: None reported.

Author Disclosure: Ms O’Kane is the President of the National Committee for Quality Assurance (NCQA). She reports that a member of NCQA’s Board of Directors, Dolores L. Mitchell, also serves as the Executive Director of the Commonwealth of Massachusetts Group Insurance Commission, the employer group that was studied in the paper to which this editorial refers.

Authorship Information: Drafting of the manuscript; critical revision of the manuscript for important intellectual content; and supervision.

Address correspondence to: Margaret E. O’Kane, President, National Committee for Quality Assurance, 1100 13th St, NW, Washington, DC 20005. E-mail: okane@ncqa.org.

1. Sinaiko AD, Rosenthal MB. Consumer experience with a tiered physician network: early evidence. Am J Manag Care. 2010;16(2):123-130.