I saw this article and it brought up some mixed emotions. Please share your thoughts on whether you agree or not.
Why Isn’t Innovation Helping Reduce Health Care Costs?
JUNE 4, 202010.1377/hblog20200602.168241
National health care expenditures (NHEs) in the United States continue to grow at rates outpacing the broader economy: Inflation- and population-adjusted NHEshave increased1.6 percent faster than the gross domestic product (GDP) between 1990 and 2018. US national health expenditure growth as a share of GDPfar outpacescomparable nations in the Organization for Economic Cooperation and Development (17.2 versus 8.9 percent).
Multiple recent analyses have proposed that growth in the prices and intensity of US health care services—rather than in utilization rates or demographic characteristics—is responsiblefor thedisproportionate increasesin NHEs relative to global counterparts. The consequences of ever-rising costs amidubiquitous underinsurancein the US include price-induceddeferral of careleading to excess morbidity relative to comparable nations.
These patterns exist despite arobust innovation ecosystemin US health care—implying that novel technologies, in isolation, are insufficient to bend the health care cost curve. Indeed, studieshave documentedthat novel technologies directly increase expenditure growth.
Why is our prolific innovation ecosystem not helping reduce costs? The core issue relates to its apparent failure to enhance net productivity—the relative output generated per unit resource required. In this post, we decompose the concept of innovation to highlight situations in which inventions may not increase net productivity. We begin by describing how this issue has taken on increased urgency amid resource constraints magnified by the COVID-19 pandemic. In turn, we describe incentives for the pervasiveness of productivity-diminishing innovations. Finally, we provide recommendations to promote opportunities for low-cost innovation.
Net Productivity During The COVID-19 Pandemic
The issue of productivity-enhancing innovation is timely, as health care systems have been overwhelmed by COVID-19. Hospitals in Italy, New York City, and elsewhere have lackedadequate capital resourcesto care for patients with the disease,sufficient liquidityto invest in sorely needed resources, andenough staffto perform all of the necessary tasks.
The critical constraint in these settings is not technology: In fact, the most advanced technology required to routinely treat COVID-19—the mechanical ventilator—was inventednearly 100 years ago in response to polio (the so-called iron lung). Rather, the bottleneck relates to the total financial and human resources required to use the technology—the denominator of net productivity. The clinical implementation of ventilators has been illustrative: Health care workers are still required to operate ventilators on a nearly one-to-one basis, just like in the mid-twentieth century.
High levels of resources required for implementation of health care technologies constrain the scalability of patient care—such as during respiratory disease outbreaks such as COVID-19. Thus, research to reduce health care costs is the same kind of research we urgently require to promote health care access for patients with COVID-19.
Types Of Innovation And Their Relationship To Expenditure Growth
The widespread use of novel medical technologies hasbeen highlightedas a central driver of NHE growth in the US. We believe that the continued expansion of health care costs is largely the result of innovation that tends to have low productivity (exhibit 1). We argue that these archetypes—novel widgets tacked on to existing workflows to reinforce traditional care models—are exactly the wrong properties to reduce NHEs at the systemic level.
Exhibit 1: Relative productivity of innovation subtypes
Source: Authors’ analysis.
Content Versus Process Innovation
Content (also called technical) innovationrefers tothe creation of new widgets, such as biochemical agents, diagnostic tools, or therapeutic interventions. Contemporary examples of content innovation includespecialty pharmaceuticals,molecular diagnostics, and advancedinterventions and imaging.
These may be contrasted with process innovations, which address the organized sequences of activities that implement content. Classically, these include clinical pathways and protocols. They can address the delivery of care for acute conditions, such ascentral lineinfections,sepsis, ornatural disasters. Alternatively, they can target chronic conditions through initiatives such asteam-based managementof hypertension andhospital-at-homemodels for geriatric care. Other processes includehiring staff,delegating labor, andsupply chain management.
Performance-Enhancing Versus Cost-Reducing Innovation
Performance-enhancing innovations frequently createincremental outcome gainsin diagnostic characteristics, such as sensitivity or specificity, or in therapeutic characteristics, such asbiomarkers for disease status. Their performance gainsoften lead tohigher prices compared to existing alternatives.
Performance-enhancing innovations can be compared to “non-inferior” innovationscapable ofachieving outcomes approximating those of existing alternatives, but at reduced cost. Industries outside of medicine, such as the computing industry, haverelied heavilyon the ability to reduce costs while retaining performance.
In health care though, this pattern of innovation is rare. Since passage of the 2010“Biosimilars” Actaimed at stimulating non-inferior innovation and competition in therapeutics markets, only 17 agentshave been approved, and only seven have made it to market.More than three-quartersof all drugs receiving new patents between 2005 and 2015 were “reissues,” meaning they had already been approved, and the new patent reflected changes to the previously approved formula. Meanwhile, the costs of approved drugshave increasedover time, at rates between 4 percent and 7 percent annually.
Moreover, the preponderance of performance-enhancing diagnostic and therapeutic innovations tend to addressnarrowpatient cohorts (such asrare diseasesorcancer subtypes), withlimited clear clinical utilityin broader populations. For example, the recently approved eculizimab is a monoclonal antibody approved for paroxysmal nocturnal hemoglobinuria—which effects 1 in 10 million individuals. At the time of its launch, eculizimabwas pricedat more than $400,000 per year, making it the most expensive drug in modern history. For clinical populations with no available alternatives, drugs such as eculizimabmay becost-effective, pending society’s willingness to pay, and morally desirable, given a society’s values. But such drugs are certainly not cost-reducing.
Additive Versus Substitutive Innovation
Additive innovations are those thatappend topreexisting workflows, while substitutive innovations reconfigure preexisting workflows. In this way, additive innovationsincrease the useof precedent services, whereas substitutive innovations decrease precedent service use.
For example, previous analyses have found that novel imaging modalities are additive innovations, as theytend not todiminish use of preexisting modalities. Similarly, novel procedures tend toincompletely replacetraditional procedures. In the case of therapeutics and devices, off-label uses in disease groups outside of the approved indication(s) can prompt innovation that is additive. This isespecially true, giventhat off-label prescriptions classically occur after approved methods are exhausted.
Eculizimab once again provides an illustrative example. As of February 2019, the drughad been usedfor 39 indications (it had been approved for three of those, by that time), 69 percent of which lacked any form of evidence of real-world effectiveness. Meanwhile, the drug generated nearly $4 billion in sales in 2019. Again, these expenditures may be something for which society chooses to pay—but they are nonetheless additive, rather than substitutive.
Sustaining Versus Disruptive Innovation
Competitive market theory suggests that incumbents and disruptorsinnovate differently. Incumbents seek sustaining innovations capable of perpetuating their dominance, whereas disruptors pursue innovationscapable of redefiningtraditional business models.
In health care, while disruptive innovations hold the potential to reduce overall health expenditures, often they run counter to the capabilities of market incumbents. For example, telemedicinecan delivercare asynchronously, remotely, and virtually, but large-scale brick-and-mortar medical facilitiesinvest enormous capitalin the delivery of synchronous, in-house, in-person care (incentivized by facility fees).
The connection between incumbent business models and the innovation pipeline is particularly relevant given that 58 percent of total funding for biomedical research in the US is now derivedfrom private entities, compared with 46 percent a decade prior. It follows that the growing influence of eminent private organizationsmay favor innovationssupporting their market dominance—rather than innovations that are societally optimal.
Incentives And Repercussions Of High-Cost Innovation
Taken together, these observations suggest that innovation in health care is preferentially designed for revenue expansion rather than for cost reduction. While offering incremental improvements in patient outcomes, therefore creating theoretical value for society, these innovations rarely deliver incremental reductions in short- or long-term costs at thehealth systemlevel.
For example, content-based, performance-enhancing, additive, sustaining innovationstend to addlayers of complexity to the health care system—which in turn requireadditional administrationto manage. The net result is employment growthin excess ofoutcome improvement, leading to productivity losses. This gap leads to continuously increasing overall expenditures in turnpassed along topayers and consumers.
Nonetheless, high-cost innovationsare incentivizedacross health care stakeholders (exhibit 2). From the supply side of innovation, for academic researchers, “breakthrough” and “groundbreaking” innovationsconstitute the basisfor career advancement via funding and tenure. This is despite stakeholders’ frequentinability to generalizeearly successes to become cost-effective in the clinical setting. As previously discussed, the increasing influence of private entities in setting the medical research agenda is also likely to stimulate innovation benefitting single stakeholders rather than the system.
Exhibit 2: Incentives promoting low-value innovation
Source: Authors’ analysis adapted from Hofmann BM.Too much technology. BMJ. 2015 Feb 16.
From the demand side of innovation (providers and health systems), a combined allure (to provide“cutting-edge”patient care), imperative (to leave“no stone unturned”in patient care), and profit-motive (toamplifyfee-for-service reimbursements) spur participation in a“technological arms-race.”The status quo thus remains as Clay Christensenhas written: “Our major health care institutions…together overshoot the level of care actually needed or used by the vast majority of patients.”
Reference
Retrieved fromhttps://www.healthaffairs.org/do/10.1377/hblog20200602.168241/full/