As Health Care Changes, so must its technology
We are in the early phases of an unprecedented tidal wave of change in the healthcare delivery system. The prior method of delivering care, characterized by a vast cottage industry of small, independent practices functioning in a fee-for-service environment, is being replaced by coordinated delivery systems charged with managing populations and functioning in a performance-based environment.
The trends have been progressing over the past decade. According to the Center for Studying Health System Change, small clinical practices (solo and 2 physicians) have been steadily declining in their percentage of overall practice types, and are being replaced by mid-sized single-specialty practices.
Further, the percent of practices owned by physicians, as opposed to owned by hospitals or other healthcare institutions, has also been declining. This is especially true of medical specialties, less so among primary care physicians. The data from HSC, however, is a few years old (published in 2007). One needs to look at other data to gather a more recent picture of how physicians are deployed.
The Medical Group Management Association (MGMA) studies physician practice patterns, payment methods and compensation every year. Their studies are, therefore, of their member medical groups (not individual practices), and thus represent a certain bias. However, their study of the overall trends in ownership of physician practices (or, at least, physician group practices) has shown a dramatic shift to hospital ownership since 2005. This is best illustrated by a graph in a presentation by BDC Advisors in 2010, using MGMA data:
This trend is not surprising. Smaller, independent practices that may have flourished in a fee-for-service environment are hard-pressed to perform in the sophisticated kinds of ways expected of the emerging health care marketplace – clinical quality measures, outreach to patients who “fall through the cracks” for wellness and chronic-disease interventions, coordination with community, home-health and hospital services in a systematic way, are all very difficult to achieve in a small-practice environment – especially one with a weak technology infrastructure.
Paying attention to the “total cost of care” is something that only a coordinated approach between hospitals, primary care physicians, specialty physicians, health plans, and community resources can begin to achieve. The incentives for each of these elements in health care need to be aligned such that none of them see the others as “competition” for the same dollars.
Hospitals cannot see their business rely on “filling beds” as their source of revenue – the higher the census, the better they do – and therefore see owning physician practices as “volume-funnels” to fill the beds in their hospital systems. Physicians cannot see hospitals as “regional monopolies” forcing health plans to pay higher payments (thus driving up premiums), and competing for a bigger piece of the pie with physicians. Bundled payment approaches, deployed through Patient Centered Medical Homes (PCMHs) and/or Accountable Care Organizations (ACOs), may be the best way to date in aligning incentives such that when everyone does a good job of keeping a population healthy – using primary care and home-outreach services effectively – then the most-expensive way of serving a population (emergency rooms and hospitalizations) can be kept to its minimum.
What does this mean for health information technology?
Transformation of the healthcare delivery system will be long-term, difficult, and perhaps generational. And the kinds of tools needed to facilitate such change will need to evolve as well.
Traditionally, clinical documentation has been on paper. The earliest Electronic Medical Records (EMR) systems were mainly ways of replacing paper with an e-chart. They were primarily focused on documenting clinical encounters, and generating bills for those encounters in a fee-for-service environment. “Efficient coding” has been a key concept in these systems, making sure to capture every service done in order to get fairly paid.
But the new environment asks more from its technology tools. Of course, modern Electronic Health Records (EHR) need to cover the basics. They must be able to be efficient, intuitive, and work alongside the clinical workflows of everyone in the practice in order not to be obtrusive (and therefore resisted). They need to allow order entry – from sending electronic prescriptions directly to pharmacies, to ordering lab tests electronically, to ordering x-ray and imaging studies electronically as well. And they need to do these things in ways that allow for obtaining insurance (or medical group/PCMH/ACO) authorization when necessary. They need to capture outside data, like lab test results, imaging results, and correspondence from consultants, and integrate these into the EHR record.
However, that is not enough. The upcoming new organizational forms that physicians will find themselves working in also require other capabilities from health IT. These can be grouped into 3 different categories: (1) a collaboration platform, (2) clinical quality measures with linked clinical decision support and prompting at the point of care, and (3) robust data reporting.
A collaboration platform. As medium-sized medical groups and coordinated delivery systems evolve and emerge, the need for everyone to be able to communicate with each other becomes paramount. A seamless (yet secure) way to work as a team for everyone within the network needs to be part of the routine of care. The situation where everyone is on a different system, and each system can’t talk very well to the others, has been a vexing problem heretofore. Technology must be able to bridge that gap and create a collaboration platform that works with the business-to-business workflows found in health care.
CQM and CDS. Traditionally, clinical quality measures have been collected by others (not the clinicians themselves), and have been based on billing data. Modern EHRs must be able to collect clinical quality data at the level of the clinical practice, in order to answer questions like “what percentage of my diabetic patients are poorly controlled, as seen by glycohemoglobin blood tests of >9%?” and “how does my performance here compare to my colleagues?” Such measures are at the population level, and are important to a medical practice. At the individual level, clinical decision support can serve as CQM-driven prompts: “Mr. Garcia is here for seasonal allergies, but is a diabetic who has not had a diabetic eye exam in over a year – time to order one, while you have an opportunity.”
Of course, more sophisticated clinical decision support – such as Order Sets and “clinical pathway” protocols, found mostly in hospital settings – are also expected from modern health IT. But the starting point of CQM-based CDS is something health IT needs to offer to office-based physicians in these new settings.
Robust reporting. Clinical practices want to be able to look at their own data. This has been one of the traditional arguments for hosting EHRs and the EHR data locally within a hospital or medical group. However, web-based hosting (which offers better data security and safety, as local computers don’t have any residual PHI left on them when the web session is over) can also offer similar tools for looking at one’s own practice data.
Certain “canned” reports are, of course, expected – lists of patient who have a given diagnosis, or who are on a given medication, or who are candidates for a given wellness intervention (such as mammography, or influenza vaccination) and are due. However, larger groups will also be able to hire local IT consultants (or even employees) who will want to examine the practice’s data in new and different ways that were not part of the library of “canned” reports. Modern EHRs, whether they are locally housed or are located securely on the web, need to make available such access to the data in order that ad-hoc reports can be created at the practice level. When the data is on the web, a web-services API is the most likely way that such data can be made available.
Health care is evolving into new forms that are based more on the overall management of populations, and keeping them healthy, and away from the traditional transaction-oriented fee-for-service system that has dominated the past. As a consequence, physicians are gravitating towards larger groups (generally, single-specialty groups), and such groups are gravitating more toward hospital-owned settings.
The new coordinated delivery networks, such as PCMHs and ACOs, are starting to emerge. Under performance-based compensation and global, bundled payments, the relationship between hospitals, primary care physicians, specialists, and health plans is changing (albeit reluctantly, and slowly).
The kinds of technology needed by this change in health care is evolving as well. Beyond the legacy of simple chart note capture and billing, and beyond the e-prescribing, order entry, and outside-data capture of the EHRs we have today, the new frontier for health IT will be in anticipating – even leading – the changes that are coming in health care. The first things we need to build into our EHRs are (1) a collaboration platform that works; (2) a clinical decision support system present at the point of care, closely inter-related with clinical quality measures captured by the EHRs directly; and (3) a robust reporting capability which goes beyond “canned reports” and offers way of accessing one’s own EHR data so that ad-hoc reporting can be done as desired by the local practice.
This is the path forward for us all. With a clear vision, we can build it.