Hospitals’ Twenty First Century Time Warp

There has been a lot of controversy in health policy circles recently about hospital market consolidation and its effect on costs.  However, less noticed than the quickened pace of industry consolidation is a more puzzling and largely unremarked-upon development:  hospitals seem to have hit the wall in technological innovation.   One can wonder if the two phenomena are related somehow.

During the last three decades of the twentieth century, health policymakers warned constantly that medical technology was driving up costs inexorably, and that unless we could somehow harness technological change, we’d be forced to ration care.  The most prominent statement of this thesis was Henry Aaron and William Schwartz’s Painful Prescription (1984).  Advocates of technological change argued that higher prices for care were justified by substantial qualitative improvements in hospitals’ output.

Perhaps policymakers should be careful what they wish for.  The care provided in the American hospital of 2013 seems eerily similar to that of the hospital of the year 2000, albeit far more expensive.    This is despite some powerful incentives for manufacturers and inventors to innovate (like an aging boomer generation, advances in materials, and a revolution in genetics), and the widespread persistence of  fee for service insurance payment that rewards hospitals for offering a more complex product.

Technology junkies should feel free to quarrel with these observations.  But the last major new imaging platform in the health system was PET , which was introduced into hospital use in the early 1990’s.  Though fusion technologies like PET/CT and PET/MR were introduced later, the last “got to have it” major imaging product was the 64 slice CT Scanner, which was introduced in 1998.  Both PET and CT angiography were subjects of fierce controversy over CMS decisions to pay for the services.

Arguably the last major technological advance in surgical technology was the daVincisurgical robot, introduced in 2000.  And the last new surgical product line was bariatric surgery for weight control, which hospitals began offering around 2000.  The last major advance in interventional cardiology/radiology was stenting and coiling, again introduced in the late 1990s.  (Drug-eluting stents, a technological refinement, but not a game changer, were introduced in 2002).  The last major advance in critical care: the eICU introduced by VISICU (now a part of Phillips Healthcare) in 1998. The last major innovation in hospital logistics was probably Pyxis (now a part of CareFusion), the automated pharmacy system, which was introduced in the mid-1990’s.

Though there’s been a lot of policy attention to electronic health records, the two dominant integrated hospital enterprise IT platforms are also nearing or past their fifteenth birthdays.  First-to-market Cerner began development of its Millenium suite in 1996, and began installing it in 1998.   Market leader Epic introduced EpicCare in 2000.  These suites, as well as those of competitors like MediTech, GE, Seimens and Allscripts, are subject to updates and refinements, as well as a steady stream of new applications.  But the architectures themselves, as well as their Windows 95 style user interfaces, have evolved little since they were introduced. Hospitals’ enterprise IT incumbents have become so wealthy and powerful that their incentive to innovate has diminished, as  has the power to disrupt or displace them.

It’s difficult to point to a single cause of this technological menopause, which is largely beyond hospitals’ control.    Device regulation was “modernized”  in the Food and Drug Administration Modernization Act  of 1997 (FDAMA). The law systematized device approvals and levied user fees to expedite the review process.    This law also made it far easier and cheaper for companies to modify existing innovations than to birth completely new ones.  Even so, the expedited 501k  (substantial equivalence) path to market has become increasingly difficult to follow. The percentage of 510k requests granted without requests for additional information fell from over 50% in 2000 to less than 20% in 2010, and approval times have increased significantly.

The glacial progress toward such potential game changing technologies as molecular imaging or mHealth can in part be traced to a much more demanding regulatory regime, as well as a more difficult pathway to payment once approval is granted.  For example, we will shortly have molecular markers for tumor aggressiveness or propensity to metastasize.  These markers will help patients avoid the anxiety of an inconclusive cancer diagnosis and subsequent diagnostic cascade.

Yet these biological markers and many other useful molecular imaging tools are being regulated by FDA essentially as if they were drugs, even though they may yield a tenth to a hundredth of the revenues per marker.   One wonders what path truly disruptive technological innovations such as endoscopy or CT scanning would have taken under this regulatory regime.

Colleagues in med tech venture investing believe that “the fix is in” on payment approvals for new medical technologies, not only for Medicare, but for the vitally important private insurance sector.  FDA cleared products face daunting hurdles to obtaining billing codes, which can take over two years.  Even if that happens, payment levels may not return the invested R+_D capital. Venture investment in medical device space has steadily declined in the past fifteen years, before falling off the table after the 2008 financial crisis.  US firms are moving their medical device and product R+D activities and workforces overseas.

The decline in hospital inpatient utilization and the marked flattening of outpatient activity in the past five years can in part be attributed to the lack of “got to have it” new clinical services.  While there is a tremendous amount of process innovation aimed at making hospital services safer and more reliable, the core product seems frozen in a 21st Century time warp.

Whatever the cause, it is inconvenient for hospitals to defend significantly higher societal costs for a care product that hasn’t changed in technological terms in almost fifteen years.   Hospitals’ adjusted census (a crude composite which adds in outpatient activity) was about 14% higher in 2011 than in 2000, according to the recently published 2013 Hospital Statistics by the American Hospital Association, while hospital revenues/costs to society have roughly doubled.

I’m not certain what the path to accelerating innovation in the hospital world is. But it’s going to be a difficult period for hospitals politically in the next two to three years, as millions of new people make their way to health coverage under the Affordable Care Act.
Hospitals are going to be pressed to defend their rising costs, and that will be more difficult to do if their product is not changing meaningfully.

Originally published on The Health Care Blog.