The unifying principle for care delivery in every situation should ideally be centered on complete clinical information, coupled with evidence-based medicine, and filtered through clinical decision support tools.
Ira M. Klein, MD, MBA
Winston Churchill once said, “Americans can be counted on to do the right thing, once they’ve exhausted all other alternatives.” In our healthcare delivery system, we’ve embraced this principle through many false starts and unfulfilled promises in the fields of clinical informatics and integration. A few shining stars were mentioned in the recent presidential debates, but not every system can be the Mayo Clinic or Cleveland Clinic. Healthcare delivery is like politics: it’s all local. The unifying principle for care delivery in every situation should ideally be centered on complete clinical information, coupled with evidence-based medicine, and filtered through clinical decision support tools for best healthcare outcomes tailored to individual patients. That is where we want to be overall as a healthcare delivery system. In clinical oncology, that era of informatics may be coming very shortly.
Source: Reprinted with permission from National Cancer Institute. Cancer costs projected to reach at least $158 billion in 2020: new NIH study projects survivorship and costs of cancer care based on changes in the US population and cancer trends. cancer.gov. Accessed October 25, 2012.
The background on why oncology practice would act as a leading edge for medical informatics includes three interdependent factors. These are (in no particular rank order): new and expensive drug therapies, Centers for Medicare & Medicaid Services (CMS) responses to cost rise trend pressures in oncology, and advances in health information technology (HIT) coupled with federal legislation (Affordable Care Act). This confluence of clinical, cost, HIT, and legislative factors is immensely powerful, squeezing the structure of oncology care delivery into something new. The resulting yield is a huge potential expansion of efficiency and quality through the application of informatics.From 2000 until 2011, 40 new cancer drugs were approved by the US Food and Drug Administration (FDA) in the United States. As documented by a recent New York Times editorial,1 the typical new cancer drug coming on the market a decade ago cost about $4500 per month (in 2012 dollars); since 2010 the median price has been around $10,000. The majority of new drugs in the pipelines of every major pharmaceutical company in the world are targeted at either cancer or a rheumatologic/immunologic condition. The costs of new drugs approved by the FDA continue to rise, with an annual increase of 20%. No sign of abatement is seen, given the swollen phase II and phase III pipelines and associated companion diagnostic biomarker tests for many of them (Figure 1).2The CMS response to the rising cost of cancer care was to completely change drug reimbursement to oncologists in 2004. As of 2005, Medicare recipients constituted 70% of Part B drug reimbursement by most oncology practice volume,3 with an incidence of cancer in the 6 to 8 per 1000 range of Medicare population, versus a commercial population incidence of 1 to 1.5 per 1000. The shift in drug payment pricing over to an Average Sales Price (ASP) + 6% methodology has ratcheted down the profitability portion of drug “buy and bill” hugely, with a net effect being an average effective payment cut on drugs of 46.9% as of 2010.4Community oncology practices, which were the predominant site of service early in the previous decade, responded in two ways. One, they sought to increase the volume of patient flow through their offices. Two, those practices that could not wring out more efficiencies or carefully manage the millions of dollars of oncology drugs in their infusion services inventory elected to merge with other practices or sell their practices to local hospital systems.
This was happening at the same time that hospitals were increasingly applying for and obtaining “340b” certification. The government certification program was designed to allow for more access to oncology care in underserved populations by requiring drug manufacturers to sell to hospitals at significant double-digit discounts. Hospitals could then reprice these drugs in their newly purchased oncology practices at the richer hospital outpatient site-of-service pricing and generate large windfalls of profit.As the site-of-service moved from the community to hospital outpatient—and drug prices soared— costs to private insurance payers and self-insured employers doubled or tripled.
During this same time frame, the amount of private payer precertification oversight increased to manage the costs of the pipeline of newer, very expensive, and highly specific drugs coming onto the market.
As a result, oncology practices needed to hire more full-time employees to manage patient flow. Struggling community oncology practices saw office practice salary expenses grow just as profits sagged. Hospital-based systems, however, absorbed those costs easily using their 340b arbitrage profits.In the midst of these changes in oncology business economics, the federal government sought to promote technology solutions as an overall balm to healthcare cost rises. The 2010 Affordable Care Act (ACA) facilitated new delivery and payment systems, such as accountable care organizations (ACOs). ACOs can only function successfully in an environment of data sharing and interchange, with processing and distribution of data linked to outcomes and evidence-based medicine. Government funds were disseminated to create this environment. Since early 2012, $2 billion has gone to physicians and hospitals to increase the use of electronic health records (EHRs), and another $28 billion has been dedicated to informatics anticipated by 2015.5
However, an informal survey in 2010 of oncology practices showed that a significant proportion lacked complete oncology-specific office practice systems (non-clinical decision support—focused). Thus, a gap in technology and oncology-specific business solutions was being recognized at the same time as the need for sophisticated and personalized medicine for oncology care was becoming acute.
The shift to accountable care models and other integrated delivery systems that pay for value instead of volume will accentuate a lack of business viability to any provider entity not able to input, process, and output the full spectrum of member-specific clinical and demographic information. Many in the informatics industry recognize this point of inflection in oncology care and the opportunity to build administrative and clinical decision support tools that can support population health.
Early trials using electronic clinical decision support in oncology include the US Oncology study with Aetna in 2009 on non—small cell lung cancer patients. Using an evidence-based pathways software system, the outpatient costs of care in the community setting were reduced by 35%,6 with no evidence of change of outcomes (Figure 2).
Subsequent studies by P4 with CareFirst and other evaluations have validated the use of pathways as part of an overall and bigger goal of using informatics to deliver efficient, holistic, and evidence-based cancer care. Sometimes the results are not straightforward, as drug cost savings may evaporate when new, high-priced, and truly differentiating therapies become available as best care options.
Average 12-Month Cost
(n = 166)
(n = 34)
0.44 to 0.88
Overall Survival Probability
(n = 1409)
(n = 1095)
(n = 314)
Purpose: Evaluate the cost effectiveness of evidence-based treatment pathways for patients with NSCLC.
Conclusions: Results of this study suggest that treating patients according to evidence-based guidelines is a cost-effective strategy for delivering care to those with NSCLC.
NSCLC indicates non—small cell lung cancer.
Source: Reprinted with permission from Neubauer MA, Hoverman JR, Kolodziej M, et al. Cost effectiveness of evidence-based treatment guidelines for the treatment of non-small-cell lung cancer in the community setting. J Oncol Pract. 2010;6(1):12-18.
Some would argue that drugs should be completely voided from the efficiency portion of the care equation (drug costs zeroed out), per the episodes of care pilot. The argument is that avoidable hospital and emergency department visits will certainly diminish with improved coordination of care, and savings should start in this area. However, market benchmarks, patient case acuity mix, and practice starting points differ. Additionally, the cost and utilization parameters of measurement of oncology care quality and efficiency for any one practice, outside of a Kaiser and Mayo Clinic environment, resides among multiple private healthcare payers and CMS. Therein lies both the key problem and the seeds to a universal informatics solution.Vertically integrated, staff-model health maintenance organizations that own provider assets and self-contained integrated delivery systems have a powerful informatics advantage: a single data highway through which all patient-specific information can flow. What if all payers of healthcare services could use a payer-agnostic data highway to portage information to and from provider sites of care? This would allow an oncologist’s office to see a range of patients with all types of insurance coverage, unencumbered by a heterogeneous set of inputs. They could obtain pretreatment information populated on a patient-specific electronic medical record (EMR), with knowledge of lab values, prescription drug history, and radiology test results available. The ideal system would be able to plug into hospital EMR systems such as Centricity and Epic.
Oncologists would have a holistic view of complex patients who may require multiple hand-offs to different specialists. Different private payers could also craft their own pay-for-value contract terms with providers, without creating a need for multiple payer-specific work flows in the office. Furthermore, this health information highway, or exchange (HIE), could be the place where consult and referral applications reside, to be pulled off of “the cloud” at any time for use with colleagues. Payer-specific applications related to networks and services would also help providers help patients. Lastly, this HIE becomes the foundation for interactions with ACOs and other risk-bearing entities (Figure 3).
Communication and services are arrayed around the patient’s care; simplified and streamlined, connected to all involved in care, and ACO capable.
ACO indicates accountable care organization.
As pay for value supplants traditional fee for service as the predominant method of financial interaction in the public and private marketplace, having the capabilities to build and deploy HIT in an oncology office and connect to an HIE becomes critical to maintaining a growing business. Creation of a payer-agnostic HIE, coupled with ownership of additional technology and actuarial analytics, will furnish both the platform for doing business and the outcomes-based reimbursement requirements that can be payer-specific. The traditional functions of disease management and care management will migrate to these information-rich platforms, with more informed and incisive engagement opportunities.
Most importantly, by providing this technology at low or no cost in a revenue-neutral or revenue-positive arrangement for oncology providers, the private market is removing a major barrier for adoption of healthcare informatics at the provider level. The key oncology points are the need for a multi-payer solution, housed on a single platform, with a solution that keeps pace with the science, and is holistic in supporting a team approach, biomarker technology, patient education, and care management and bioethics capabilities.
Addressing oncology-specific needs in a streamlined approach to evidence-based cancer care is the goal of any informatics solution specific to oncology practice. The practicing oncologist cares about delivering the highest quality of care to his or her patients, doing it in a practice environment that has simplified work flows, generating enough revenue to stay in business or grow, and maintaining viability for the foreseeable future.
As explained earlier, the old “buy and bill” environment is going away. The revenues generated in the past allowed practices to create widespread access to community-based office infusion services (economical and patient-centric). However, knowledge-intensive services (payment for cognitive services) were de-emphasized as a revenue driver. A system that captures all costs, granular patient-specific information, and all related care activities can measure and reward cognitive services on a comparative basis. This could change contracting terms with hospital-based systems, and help community oncologists maintain autonomy through pay-for-value incentives. Now, an HIE-empowered practice, with applications that have oncology clinical decision support capabilities, can have revenue opportunities and work flow simplification.
At the office practice level, the most flexible software tools are those that are Web-based, requiring only Internet access to use. Understanding the recent history of oncology practice and the present business environment, Aetna is piloting a “next generation” oncology software application by eviti, a clinical decision support vendor; the program offers the most complete and unbiased digital library of more than 1200 evidence-based treatment regimens for more than 120 cancer types. The regimens are compiled from government and industry data and take into account costs, effectiveness, side effects, and outcomes. In addition, the library includes more than 9000 federally registered clinical trials. The eviti software is used as a Web-based application or as a part of EHR software. Aetna is supplying the health information highway in the form of the payer-agnostic iNexx platform from Medicity. Both services are provided free of charge, with only a commitment-for- use agreement necessary by the practices entering the pilot.A significantly attractive feature of the eviti tool is the cross-walking of Aetna’s clinical policy bulletins into the software that allows for realtime, automated preauthorization approval when evidence-based protocols that are patient specific (including genomic markers and performance status) are chosen. This reduction in administrative work will allow an oncology practice to dedicate less staff to preauthorization tasks, and more to patient education and monitoring.The flexibility of pulling “apps” that provide sorted data in useful and actionable information from an HIE portal, such as iNexx, and using them to improve connectivity among healthcare providers, payers, and patients is a transformative event. Oncology practice is one of the most clinically intensive practice activities in medicine, made more complex and expensive with huge growth in biologic drugs. Oncology is ideal to lead the way in this technology construct because clinical benefits will be yielded in the short term. The economics of oncology practice also pressure a change from pay-for-volume to pay-forvalue. And now, a platform exists that can take on the needed monitoring and measuring. Lastly, the entire architecture of healthcare delivery is changing, and creating the voluntary HIE pipes in oncology will lead to more rapid adoption elsewhere.
Many vendors focused on clinical decision support and consumer engagement are eager to jump on the health information highway. We are eager to support them and the innovation that comes from competition. All agents in the system—payer, provider, and patient—win when better, cheaper, effective, and easier-to-use tools are brought to market. We support these efforts, and look to work collaboratively with others who think likewise. Innovation is best done with an understanding of the history and workings of the marketplace. Aetna is a change agent and committed to building a better healthcare delivery system for everyone.
Author Affiliation: From Aetna, Hartford, CT.
Funding Source: None.