Tuesday, December 16, 2014

Healthcare Management Forum, Winter 2014: Guest Commentary by Colin Goodfellow

(Cet article ne est pas disponible en anglais.)

The dramatic dehospitalization of health services is a prerequisite for a sustainable and effective health system

Using the general precepts of integration, Lean thinking, and patient centredness, this article highlights the potential for dramatic dehospitalization of health services as a prerequisite for a sustainable and effective health system.

Watch as Colin Goodfellow challenges the traditional notions on the utility of hospitals and more

All those in favour of integrated services, Lean design, and patient-centred health, say aye. These three watchwords are themes for much of the action around health system change and health improvement in Canada. Listening to change leaders from Britain, Australia, and the United States over the past few years, it is clear that Canada is not alone in the pursuit of these directions. This commentary proposes that, for public or government-dependent health systems, results will be disappointing, unless we move away from our cultural fixation on hospitals and move towards digitally enabled distributed specialty service delivery channels. In short, our institutional pendulum needs to swing in a dramatically new direction, taking out large chunks of hospitals as it goes.

Using the general precepts of integration, Lean thinking, and patient centeredness, this article highlights the potential for dramatic dehospitalization of health services as a prerequisite for a sustainable and effective health system.

But first, two words about integration: system or clinical? These two types of integration (excluding the usual notions of vertical and horizontal, which are flavour choices in the system integration discussions), currently popular in the health services field, are confused, considered as interdependent, or most commonly system integration is presented as a precondition for clinical integration. System integration 1 occurs when most health services are under the operational direction of a single agency—theoretically enabling silo busting service optimization; for example, Kaiser Permanente or the Department of Veterans Affairs in the United States. The second form, clinical integration, exists when all the appropriate diagnostic and treatment modalities and health-promoting interventions seamlessly line up and are readily accessible to the patient for optimal health outcome; for example, the “right care, right time, right place, right outcome” view of the world.

These two forms of integration are obviously different, but a significant point that seems to be lost is that they have no necessary relationship to each other. Consider most provincial health systems in Canada: it is obvious that one can have system integration without clinical integration. Clinical integration can occur within both integrated and non-integrated health systems, as there are great examples of integrated clinical management in both integrated and non-integrated delivery systems. In my neck of the woods, I point to the Ottawa Heart Institute’s ST-elevation myocardial infarction protocol as good clinical integration within a non-integrated health delivery system. Therefore, system integration does not lead to clinical integration, and clinical integration does not require system integration. These two forms of integration have no necessary relationship to each other. According to Stephen Shortell of the HAAS School of Business at Berkeley, Integrated Delivery System performance is less than what could be achieved if there were a greater focus on clinical integration.

A Canadian study on health system integration 2 points to 10 principles required for health system integration, two of which relate to clinical integration. The first one is standardization of care and the second is the use of integrated information systems. If one were to achieve both of these, frankly, delivery system integration would be immaterial. In a single payer system with standardized clinical protocols and integrated information systems, value optimization (patient experience and clinical outcome) becomes transparent, becomes evidence-based, and, in turn, achieves many of the end results envisaged by policy-makers and funders.

That first principle, standardization of care, is a staggeringly tough objective and an obvious precondition to clinical integration. The barriers to clinical standardization go beyond the health delivery system and provider organizations; they are endemic to health and medical care professions and practitioners. Healthcare is the only industry in the developed world that remains organized and funded through “guilds.” Although practitioner reimbursement, barriers to entry and highly specialized work would benefit from Lean methods, the nub of the issue is variations in standard care enabled by individualized compensation. If you would like to fact check this reality, ask local radiologists what they think of someone else conducting readings outside their hospital/clinic, outside their province, and outside Canada. It may be acceptable to have medical transcription taking place in Mumbai but suggest off-shoring guild work and you will understand that health providers in Canada are not embracing our digital age.

In Ontario, the current response to the challenge of standardizing care is increasing regard for evidence of clinical efficacy when determining funding and to deploy different models of primary care through roster-based rather than transaction-based funding. Both of these ideas, which (1) stop paying for activities not supported by evidence and (2) encourage wellness over treatment in funding models, are attractive approaches to standardization. The fact that these two initiatives are in their early stages, remain physician guild dependent, and have little relationship to each other or major delivery components underscores how far healthcare is from the standardization of care required to support clinical integration.

The second requisite for effective clinical integration is the use of integrated information systems. According to Esther Suter, many of the perceived benefits of an integrated delivery system “are only possible with an information system that tracks utilization and enhances communication flow across an integrated pathway, links consumers, payers and providers and can be accessed anywhere.” The adoption and use of electronic medical records and standardized pathways is slow for many reasons.3 However, it is almost by definition that clinical integration is achieved with the advent of accepted clinical standardization and ubiquitous digital access to these accepted pathways/protocols and individual digital patient records. Information becomes the backbone of the system, rather than buildings or governance structures. Our health reform conversation is trapped in a “bricks-not-clicks” perspective. This conversational dead end may just be a generational hang up or a Canadian cultural fetish with the big blue H, regularly being whipped into frenzy by vested interests. The need to concentrate intellectual capital by way of care standards rather than financial capital by way of expensive buildings and complex management systems will define healthcare in the digital era. It is the path to sustainability.

The reason to underscore that clinical integration is independent of system integration or institutional integration is that it enables us to think freely about how to best offer healthcare through a Lean and patient perspective. These two declared system goals (Lean/efficient, and patient-centred) would be best served by smaller hospitals and fewer of them. By hospital, I mean a facility with an Emergency Department (ED), inpatient medical and surgical beds, and the associated diagnostics.

Although Lean precepts have been around for approximately 100 years in manufacturing, modern Lean did not really gain favour in North American manufacturing until the rust belt meltdown of the 1980s. And the practice has only been kicked about in the Canadian health system for approximately 10 years,4 with Saskatchewan leading its adoption. At their core, Lean approaches examine the process, remove everything not essential to creating value, standardize the process, and repeat. Canadian health leaders are currently rushing to adapt and adopt Lean lessons learned from American health systems. This approach overlooks a vital advantage that public or government-dependent health systems have—the ability to apply Lean methodology across the entire delivery system, not just within its component functions and processes. Provinces and health authorities should seize the opportunity to apply the principle of Lean to the design of delivery system rather than just within portions of the system. The redundancy of large hospital infrastructure would become obvious as would the more cost-effective opportunity to have patient-friendly sized physical infrastructure broadly distributed and woven into communities. (A model primary care, urgent care centre with first-tier diagnostics and robust ambulatory clinics is approximately 20,000 ft2. A day surgery centre can accommodate four operating rooms in 13,000 ft2 as there are many of these in operation in the United States.)

For the most part, our health system still functions as it was conceived in the 1960s. There have been some events around rescaling or right sizing, extending or contracting what services are covered (long-term care vs pharmacare), and, of course, structural integration–horizontal, vertical, or as in the case of Alberta absolute–which means centralized command and control with few observable links to health outcomes. With Lean tools and approaches and the digital age fully launched, we have the potential to see past our industrial era view of healthcare. Replacing worn-out hospitals with bigger hospitals, adding beds and deploying more state-of-the-art technology is a misallocation of infrastructure dollars. It also impairs the adoption of Lean methods and models for a different healthcare system. We are rebuilding yesterday׳s system when tomorrow is already here.

If we embrace Lean thinking for system design, an early design question must be—do we need hospitals at all? Although the answer is a qualified yes, we need fewer of them, they should be much smaller and they should be much more focused. The activity removed from hospitals would go to different places: (1) dedicated ambulatory care and community surgery centres, (2) primary providers, (3) the patient via e-based self-management, (4) reactivation or soft rehabilitation, and (5) some things would simply stop being done. Think of eliminating the non–health-producing people-management activity required in our large complex organizations. In any Lean analysis, these structures would be seen as inessential, non–value producing for the patient.

Consider the following two examples to appreciate the magnitude of the potential of a Lean-conceived, digitally enabled health system. As you read through this, “patient-centred” is the patient as a whole person, with a life outside the service delivery system; our “job-one” is to help that person access and enjoy as much of that life as they can.

At least 80%-85% of individuals presenting in an ED do not need the services of an ED, and 50% of those presenting would be better seen a day or two later by a general practitioner.5 Imagine the following scenarios: if EDs were scaled only for emergencies; if group family practices had responsibility for the Canadian Triage and Acuity Scale level four and five clients; and if smaller, cheaper, faster ambulatory care and urgent care centres were located conveniently in neighborhoods or smaller towns.

In Ontario, this would mean that 4.3 million people annually could be diverted to less expensive and more appropriate care. Waiting room times of an hour in urgent care centres or primary care offices would replace the ED system׳s current performance goal of 6-hour waits for non–emergency ED care. Stated another way, we have the ability to remove 22 million hours of waiting room time—that is 33 people׳s lifetimes—each year. Although the negative health consequence of a 6-hour ED wait surrounded by other sick people may be hard to quantify, removing 33 lifetimes of waiting is a tangible result for a patient-centred system. Applying Lean techniques to embed fast-track clinics within overbuilt hospital EDs misses a real opportunity to bring care to a more appropriate urgent care setting closer to patients׳ homes. Admittedly the Canada Health Act is a barrier to doing the right thing for patients and efficiency with regard to EDs, as patients cannot be turned away.

To be clear, most patients seen unnecessarily in EDs are not the patients that emergency doctors are concerned for when they talk about wait times. Their concern is rightly for those patients who need access to the hospital services (12%) yet who are blocked (>0.3%) by other people inappropriately in beds. This unhelpfully leaves the emergency patient stuck in the ED or corridor waiting.

Emergency physicians׳ observations are quite telling: roughly 16% of Ontario׳s hospital beds are occupied by people who require an alternate level of care. A patient-centred, Lean reimagined system accommodates those patients who still need some level of care in smaller reactivation/rehabilitation facilities much removed from hospitals. These are cheaper to build, cheaper to run, and closer to home, with the potential for deeply skilled specialized providers to apply continuous innovation at multiple sites. The reimagined system allows distributed continuous improvement in specialized centres.

Ongoing deep continuous improvement can never happen throughout every service line in a multiservice facility like our legacy hospitals. Yet, quality patient care requires Lean approaches and their relentless focus on customer service and value. Although department store after department store goes bankrupt, specialty stores thrive. Our patients, you and I, live in the age of Amazon yet our health institutional planning perpetuates, replicates, and celebrates an era when people shopped at Eaton׳s.

As a second example, here is a naive order of magnitude on the 16% inappropriate bed utilization issue.6 If Ontario hospitals did not do what they should not be doing (ie, provide alternate level of care), some 4,750 beds could be removed from the system. That is a sufficient number to warrant the complete closure of a few hospitals and the conversion of many smaller ones to exclusively ambulatory care and reactivation centres. Even if 50% of the operating costs of this inappropriate utilization were redirected to other service options, an allocation that is very high by the current standards,7 the annual saving would approach $500 million.

An interesting observation is that since 1960 and the advent of Medicare, Canada׳s population has doubled and the absolute number of hospital beds has decreased by 8%. Although a comparison of the gross square footage of hospitals in Canada for the same period is unavailable, I suspect it has nearly doubled, rather than reduced. What this means is that the population’s need for the foundational purpose of hospitals–EDs and inpatient beds–has dramatically reduced, yet our conception of hospitals as the appropriate core expression of the healthcare promise has not moved. This reality is a barrier to achieving a reimagined sustainable health system.

These two order of magnitude examples of what is possible in the digital health age are a proxy for the potential of the application of Lean design at the system level. It is possible to go to the United States and visit systems that are much further along the clinical integration path than in Canada; as well, there are outstanding examples of smaller distributed specialized urgent care and surgery care centres. In Canada, an interesting glimpse of tomorrow׳s “non-hospital like” hospital is emerging in small rural communities. Initially by default but increasingly by design, they are becoming primary care continuum integrators with urgent care and reactivation capacity. None yet have chosen to go “bed free”–that would offend our Canadian healthcare sensibilities and, more practically, end their funding.

It is clear that health services system integration in the absence of clinical integration is of negative or low value. According to the Ontario Hospital Association, Integrated systems in Canada on average cost 10% more than non-integrated systems in Canada. And, that clinical integration requires two things: (1) clinical protocol standardization and (2) ubiquitous digital access to the protocols and the patient record. If we make these two conditions non-negotiable for the emerging system, we are both obliged and emboldened to imagine the best physical arrangement of this new standardized knowledge healthcare system enabled by a digital backbone. From the standpoint of efficacy and patient experience, the best arrangement is demonstrably not the legacy hospitals; it is most probably through distributed networks of specialty centres with very light footprints and short life cycles. We are in the early morning of the digital age in health services, yet we are culturally affixed to a delivery system that is overbuilt and underimagined for the age. We need to change the system and we need to change it now.

Member’s and subscribers log into Healthcare Management Forum to read the following responses:  
  • Plaidoyer pour une amélioration du système de santé québécois
    Lucie Tremblay et Alexis Rheault  
  • Clinical integration: A burning platform
    Katherine Chubbs 
  • Finding economies of scale and coordination of care along the continuum to achieve true system integration
    Maura Davies  
  • Provincial and regional accountability for performance is critical in non-integrated health systems
    Michael Sherar
  • Hospitals in the digital age
    Ron Sapsford

  1. Shortell, S.M. and McCurdy, R.K. Integrated health systems. Stud Health Technol Inform. 2010; 153: 369–382
  2. Suter, E., Oelke, N.D., Adair, C.E., and Armitage, G.D. Ten key principles for successful health systems integration. Health Q. 2009; 13: 16–23
  3. Closson, T.R. The capital health region׳s early experiences: moving towards integrated healthcare. Healthc Pap. 2000; 1: 88–95 
  4. McCarthy, M. Can car manufacturing techniques reform health care?. Lancet. 2006; 367: 290–291
  5. Schull, M.J., Kiss, A., and Szalai, J.P. The effect of low-complexity patients on emergency department waiting times. Ann Emerg Med. 2007; 49: 257–264 (264.e1 [Epub Aug 22, 2006])
  6. Health Quality Ontario in partnership with ICES. 2012 report on Ontario׳s health system. 〈http://www.hqontario.ca/portals/0/Documents/pr/qmonitor-full-report-2012-en.pdf〉.
  7. Boyle T, Welsh M. Begging for care—seniors find little care in provincial aging strategy. Toronto Star. 〈https://www.osla.on.ca/uploads/pdf/Begging%20for%20Care.pdf〉; 2013. 

No comments:

Post a Comment