The Future of Hospitals: Not BUH-Bye, but B2B

In 1950, business life looked good for the US Railroad system. With almost 100 years of dominance of the American transportation system, the US railroad system had 7 times the intercity passenger-mile traffic of the then-30-year-oldcommercial airline industry. 

However, by 1970 the tables had turned, with airline service carrying passengers then 9 miles for every mile traveled by rail services, which ended up bankrupt and nationalized. (Interestingly, the proportion of passenger traffic maintained by inner city buses during this time changed very little).

While many forces drove this change – the Interstate Highway System, the familiarity and comfort of returning WWII veterans with airplane travel, and shifts in mail sorting technology were all factors -- one of the biggest failures of the railroad industry itself was the way they defined themselves.

By not envisioning themselves as being in the transportation business, railroad executives created an artificial isolating barrier where airplanes were outside their operational frame of reference, and, in fact, ignored as a potential threat to their survival. If the railroads had decided to categorize themselves as being in the ‘transportation industry’, they could have invested in airlines and modified their rail service strategies based on market forces, allowing a more productive evolution, but hubris and legacy blinded them to the reality of social dynamics.

This phenomenon was largely repeated when the Internet emerged as a competitor to the print publishing industry and looms prominently in the future of the inpatient hospital industry.

As one historical reference on rail transformation notes: “An entire generation of rail managers had been trained to operate under (a restrictive) regulatory regime…Labor and their work rules were likewise a formidable barrier to change.”

Which could be rewritten thus: “An entire generation of healthcare administrators had been trained to operate under an increasingly restrictive fee-for-service model. Physicians and the existing operating model for care delivery were likewise formidable barriers to change” 

Reducing preventable primary and secondary hospitalization rates in a defined population is a well-proven competency for health plans (and their “Plan Sponsors”, usually self-insured employers); their financial incentives are aligned to do so (they get to keep any unspent money), and by and large, their success is driven by

  • Improvements in the quality of care delivered to covered patient-employees (e.g., closing evidence-based gaps in care and focusing on clinical risk-reduction strategies) and
  • Targeted shifts in patterns of resource utilization and patient engagement (introducing Care Management for example).

While the absolute levels of engagement and quality improvement by Health Plans is quite low (only closing 26% of identified gaps in care in one study; only reaching 13% of patients eligible for Care Management in another), the results still have been quite dramatic; a 9% reduction in hospitalization rates can save (tens of) millions of dollars; shifting emergency department utilization via telephonic and urgent care is both financially and operationally valuable under risk- or bundled payments models.

“Accountable care” shifts in the locus of control (and costs) of these care delivery strategies from the employer to the provider. However, as the health plans know very well, when the patients physician recommends care management, engagement and compliance levels rise, raising the potential for successful prevention of the ‘most impactable hospitalizations’ under the population health model. Combine this with “bundled payments” or "capitation+ bonus”, or full risk, there is an incentive for more intensive resource (cost) management by providers by sourcing and using lower cost/high efficacy tools and interventions.

When in the business life of a value-based health system does it stop making sense to own an inpatient facility? What happens when the costs of inpatient services are ‘out of its own pocket’ and do not have any direct associated revenue? How do value-based health systems make the determination as to which resources to own and which to contract for on an “as-needed” basis?

Another problem with the railroad industry (and the print publishers as well) is that even when there was a recognition that things were changing, there was a concerted effort to shape the change to retain the status quo. I once heard a publisher whose magazine was getting thinner and thinner every month, say “We’ll give away web banner ads in order to sell more print pages”. On the surface, 15 full-page ads at $20,000 does not have the same value as 3,000,000 ad impressions at $25 per 1000. The real problem was that his P&L had to pay for the print production infrastructure. The print margins were 15%, leaving $45,000 in profit on $300,000 in revenue; the web margins were 45%, leaving $33,750 profit on $75,000 in revenue. Divesting the print production infrastructure and leveraging the subscriber list to build web traffic would have resulted in lower revenue, but much higher profit venture -- as many newspapers and magazines eventually figured out often too late to maintain their market share.

With the gradual decline in healthcare fee-for-service revenue will come an associated sea change in the concept of “return on investment” (ROI). First, the administrative infrastructure for coding and billing will no longer be necessary, although some of the coding functions can be repurposed for clinical and operational analytics. No longer will the value of an infrastructure or resource-related expense be assessed on its dollar-for-dollar return. In value-based care, ROI calculations will have to be determined under an operational efficiency model. Performance consistency, flexible operations, clinical supply chain management and customer value will become new standards of health system operations.

The questions a CFO will have to ask are:

  1. What is the cost/efficiency value of a resource to my ability to meet my contractual goals for quality, efficiency and patient experience management?
  2. What is the portfolio of resources I need to deploy in order to meet those goals in the most efficient (highest effect/lowest cost) way possible?
  3. What are the criteria I need to escalate and de-escalate resource allocation with any individual patient (taking into account individual patient attributes, preferences, and goals)?
  4. When I must deploy significant or complex portfolios of resources against a patient (e.g., emergencies, exacerbations of existing conditions), do I need to own them, or can I use a ‘just-in-time’ inventory construct and contract for them “on-demand”?
  5. When I do contract for on-demand care delivery services, how can I assure those vendors (which may include inpatient services, imaging and medical and surgical specialists) will be accountable for their contribution to my contractual goals for quality, efficiency and patient experience management?

Today's hybrid fee-for-service + gainshare or 'pay for performance' models are offering a transition window of economic comfort to health systems. But as cost-shifting gets more traction, eventually there will be a tipping point where, even with a successful effort at “steerage” – assuring patients stay within the ‘network of accountability’, it will take great economies of scale – tens or hundred of thousands of patients – to generate enough inpatient volume to justify a health system retaining its hospital. The likelihood of any one ACO succeeding with the cost of an inpatient infrastructure on the books is very low. Why keep the building and staff on the books when you can divest it and competitively negotiate terms and rates?

The same is true, by the way, for specialty services; the only ones an ACO might want to invest in will be driven by the population analytics. One ACO I work with has no need for a bariatric obesity program because the vast majority of their population is underweight; while analytics may justify primary care and certain specialties 'on staff' there will be an entirely new forms of private practice for specialties where services are provided to system-attributed populations, not geographic communities.

The economics of ‘accountable care’ will naturally lead to a divesture (or abandonment) of the infrastructure and overhead costs associated with running inpatient services, some high cost imaging services whose success was based on fee-for-service ROI models, and some specialty services with limited 'need' in an attributed population. Since the population health model is also based on keeping patients in the ‘network’, hospitals (and some specialists) will not compete for share of mind of patients, but share of market from ACO’s.

The future of the inpatient hospital will be as a freestanding ‘on-demand’ service competing for business-to-business contract with value-based system of care under a population health management model (interestingly this is similar to what health plans do today). Even when an inpatient hospital and an ACO are under part of the same corporate parent, it will increasingly be a zero-sum game to have the profitable ACO offset the losses or required investment of the inpatient service; the hospital will have to become ‘market competitive’, wooing and negotiating with a all the local “ACO’s” to provide value-driven services at competitive levels of quality and cost

© Steven Merahn, MD 2015