We do not know the extent of the trouble about to rain down on healthcare information technology, but it is out there, and will start pelting providers within weeks, not months.
Laura Kreofsky, Principal Advisor at Impact Advisors
A year from now, things will look different.
A year from now, ICD-10 will be paying the bills.
A year from now, Meaningful Use Stage 2 will be standard operating procedure for the considerable number of eligible hospitals and professionals who attested to Stage 1 in 2011 and 2012.
But between now and then, there is oh so much work left to do.
For some, peril will strike early in the year, as they realize the electronic health record software they bet on back then will still not be certified for Stage 2. We will see EHRs ripped and replaced. We do not yet know the extent of the damage about to hit healthcare, but it is out there waiting, and the peril will be upon providers within weeks, not months.
For others, achieving Stage 2 will feel more like death by a thousand cuts. That's what I hear from Laura Kreofsky, a principal advisor at Impact Advisors, based in Chicago, although Kreofsky hangs her hat in Portland, OR.
For two years, Kreofsky was Meaningful Use program director at Sutter Health. Now, at Impact, she advises other providers "while catching up on two years of sleep," she jokes.
"Organizations that are more mature on their EHRs, or have been using health IT for longer, may actually be more challenged in this measure than people who just went up on an EHR," Kreofsky says.
She illustrates the point by explaining that seven or eight years ago, when those providers implemented their EHRs, some standards now baked into Meaningful Use Stage 2 did not exist. "LOINC was something you kind of talked about, but nobody was using very robustly, and then all of a sudden you're in a place where you have to implement it, and you've got 12 years of lab data that's not LOINC-coded and you've got all these downstream interfaces, and you need LOINC for your lab results, which is one component of that summary of care record, so it has this cascading effect," she says.
LOINC is but one example. There are others. "It's the little things that quickly become big things," Kreofsky says.
Portal Perils Some providers built what, at the time, was innovative portal technology. But in doing so, they deployed part of their EHR vendor's patient engagement portal, overlaid with data pulled from another source, overlaid with another user interface, and a different login procedure than the vendor's own, Kreofsky says.
"I'm not saying they're going to have to step back or rethink their plans, but they may have to," she says. "They may suddenly discover that they need to rework something that, while it's not broken, is not in compliance."
The ease with which some of these organizations attested for Stage 1 of Meaningful Use may have instilled a false sense of confidence, Kreofsky says.
I suggested, and she agreed, that the essence of good healthcare IT governance is being able to put out fires while at the same time performing the kind of rigorous risk analysis and systems analysis that keeps the bigger picture in mind.
Otherwise, you end up with something like healthcare.gov, a system that had all the necessary pieces in place, but didn't have the overall project management in place necessary to insure anything like success. "The challenge is that many of these analyses and potential reworkings are very time- and resource- intensive," Kreofksy says. "It's not like you just check a box."
Like Building a House Analytics will remain another hot healthcare IT topic in 2014, but the pitfalls are considerable there too.
"Most organizations are really struggling to move beyond just pure retrospective operational analytics," Kreofsky says. "They're trying to figure out where they can deal with their existing vendors and where to supplement with other tools or technologies. They're wrestling with data integrity, and quite importantly, the right skill set for this type of work. I have seen some data points, heard some pain points, talking with clients. It's a different type of mindset from I'll say building order sets to understanding utilization trends."
I mention to her the widespread disenchantment I've heard from providers with the analytics technology they have been acquiring.
"You have to ask the question, why are you so disenchanted?" Kreofsky says. "It's probably a third or a fourth the product's problem. It gets right back to data quality, data integration, staff capabilities, and governance.
Kreofsky characterizes healthcare IT as being in an unfinished state. "My analogy is often one of building a house," she says. "You have to buy the land. You have to invest in all of the architecture. And this is all like pre-implementation planning, and then you put in the foundation and the frame of your house, and that's your EHR, and that's a lot of money."
Although providers want to move in to the EHR and settle down, they need to keep spending on the other parts and pieces to have a finished house in order to really get the full value of that house, Kreofsky says.
"'The challenge for CIOs and for organizations now is to recognize that the big capital spending era is over, but to continue investing appropriately to get the full value of the technology."
Fatigue May Set In With $17 billion already handed out in incentive money for Meaningful Use, it's hard advice to hear. "Quite honestly, if we look ahead a year, we're going to be really tired as an industry," Kreofsky says.
"Everybody's going to have to push to get to Stage 2. We're going to be dealing with ICD-10, and there's going to be increasing pressure across the board for cost management, both capital and operating, so it then becomes incumbent for organizations to really look to find the value, the innovation, the return on investment, on things like analytics and patient engagement, population health."
ICD-10 will happen. It has to. Getting paid depends on it. But Kreofsky notes that if an eligible professional or eligible hospital fails Stage 2 MU in 2014, they would not attest and would not get their incentives for the 2014 reporting year, and they would be penalized—a 1% reduction in Medicare reimbursement—which would be incurred in 2016. That's the biggest looming threat.
It's no mere coincidence that Kreofsky used to train for triathlons with a CIO friend. Next year will feel like one too.
The electronic health record vendors in the first series of CommonWell Alliance pilots include Allscripts, Athenahealth, Cerner, CPSI, Greenway, McKesson, RelayHealth and Sunquest. Testing will begin January 1.
CommonWell Alliance announced Wednesday three regions in Illinois, North Carolina, and South Carolina where pilot testing of its cross-vendor EHR interoperability services will commence January 1.
Participating providers include Lake Shore Obstetrics & Gynecology in Chicago; Hugh Chatham Memorial Hospital in Elkin, N.C.; Maria Parham Medical Center in Henderson, N.C., and Midlands Orthopaedics and Palmetto Health, both of Columbia, S.C.
The electronic health record vendors in this first series of CommonWell pilots are provided by CommonWell founding companies Allscripts, Athenahealth, Cerner, CPSI, Greenway, McKesson and Sunquest. RelayHealth, a co-founder, is providing health information exchange services between participating EHRs.
According to CommonWell, RelayHealth and the participating provider sites will be validating a patient-centric identity and matching approach, as well as a consent-driven record sharing and retrieval process across care facilities.
Specifically, the initial participating providers will support the launch by enrolling patients into the service and managing patient consent protocol. They will identify whether other provider participants have data for a patient that is enrolled in the network when the patient is at their setting of care, and transmit data to another provider that has consent to view data on that enrolled patient.
Initial use cases will focus on ambulatory practices, although data exchange between a broader range of healthcare systems is expected over time.
CommonWell plans to announce preliminary results from the initial service launch at the Health Information and Management Systems Society (HIMSS) conference in February. The initiative was announced at the last HIMSS conference in early 2013, and numerous other vendors are in a queue for CommonWell.
"We do expect customers to put this into productive use, and we get all the associated benefits from testing this in a live environment, volume testing, and validation of all the technical components working as expected," says Walter Reid, vice president of product management at McKesson.
Those pilot sites using McKesson software are using a specific version and release level of the software which is CommonWell-ready, Reid says.
"It's a record locator service for patient identity management," Reid says. "EMR vendors have committed to build this into their workflow, so the member firms are also delivering to the customer product changes to show how we're incorporating the CommonWell services into the actual workflows. It's not just a technology component. There are software changes on the vendor side as well."
CommonWell's consent management technology allows patients to authorize data sharing across participating providers.
CommonWell's announcement was accompanied by several prepared statements by participating providers.
"We know firsthand how the lack of seamless data flow between clinicians can be a challenge in delivering excellent care. Palmetto Health has patients that interact with a number of different providers," stated Tripp Jennings, M.D., system vice president, medical informatics officer of Palmetto Health. "We are pleased to be part of this interoperability effort with CommonWell because breaking down health data silos and having patient health data available to providers, no matter the setting of care, will improve quality and the patient experience."
"Despite tremendous advances in electronic health records, communication among diverse systems is poor," stated AnnMargaret McCraw, CEO of Midlands Orthopaedics. "We've signed up for CommonWell because we believe that achieving true interoperability among EHRs is fundamental to improving healthcare."
Still unknown are details of the fees expected to accompany CommonWell services once commercially deployed.
"Part of what we want to do is learn from this experience, [and] figure out the value proposition firsthand by working with customers," Reid says. "There's a backlog of requests for us to work through additional vendors that want to collaborate and participate with the alliance, so that's a next step."
A medical group's call for allowing licensed physicians, without vendor interference, to designate any recipients or senders of messages using the Direct protocol puts a spotlight on nagging EHR interoperability issues.
If 2013 has taught us anything, it's that eventually, every area of healthcare IT becomes enmeshed in politics sooner or later.
This week, speculation is bubbling that the low-cost interoperability promised by the Direct protocol is the latest vehicle for the continuation of expensive, business-as-usual interfaces between different healthcare IT vendors' systems.
Evidence of this rising concern surfaced on December 7, when the Massachusetts Medical Society House of Delegates' Committee on Information Technology passed a resolution calling for "a more open, affordable process to meet technology mandates imposed by regulations and mandates."
The resolution goes on to urge that "that all Direct secure e-mail systems, mandated by Meaningful Use Stage 2, including health information exchanges and electronic health record systems, allow a licensed physician to designate any specified Direct recipient or sender without interference from any institution, electronic health record vendor, or intermediary transport agent."
To try to find out what about Direct is rubbing physicians the wrong way in the Bay state, I spoke with Eugenia Marcus, MD, a pediatrician in private practice in Wellesley, MA and chair of the committee that adopted the resolution.
Although the Massachusetts state HIE has promised licensed doctors Direct access to the HIE for $5 per doctor per month, Marcus and her colleagues are troubled by chatter that vendors will continue to be able to charge for interfaces to immunization, disease and allergy registries on top of Direct. If true, that would thwart some early promises by Direct's advocates that the secure messaging protocol would put an end to usurious vendor interoperability fees.
So far, Marcus and her colleagues in private practice have not been issued Direct secure e-mail addresses by the state HIE, so they don't really know what to expect yet. They also worry that the ground rules for granting access to those addresses by those who assist physicians, such as nurse practitioners and other designated staff, could make the doctors possessing the Direct addresses a potential bottleneck of information flowing to and from the HIE. Those rules have yet to be communicated.
While large institutions in Massachusetts and elsewhere have already ironed out the logistics of transferring large amounts of information via Direct, a lot of the rules governing how things are going to make it out to small practices haven't been determined yet, and rumors are swirling, Marcus says.
"One of the things I have found out [is] that even though the federal government has provided funding to write the interfaces, most vendors have not taken advantage of that," Marcus says. "They're writing interfaces, but in their own shop[s], without taking advantage of the funding that is available to help support that."
Small practices—those defined by CMS as containing from one to nine physicians—are particularly challenged to implement a technology such as Direct because they lack the stable of information technology resources available at larger practices and hospitals. Since Direct is one of the requirements of the not-to-be-delayed-after-all Meaningful Use Stage 2, physicians such as Marcus who lack sufficient Medicaid patients, and thus cannot qualify for Meaningful Use payments, view Direct as yet another unfunded mandate from the federal government.
The basic aims of Direct are simple enough: to replace the fax machine with a HIPAA-compliant secure messaging protocol that expedites transfer of all or part of a patient's medical record with other providers. The challenge is to determine who holds the keys to encrypting and decrypting those secure messages.
By some measures, Direct is already a success. But to use the protocol, ONC has defined a chain of trust centered around designating health information service providers (HISPs) as holders of the right to issue Direct e-mail addresses and hold those encryption and decryption keys. It's this necessary chain of trust that has some critics concerned that Direct is a new way for large organizations and vendors to exert control over individual physicians and patients.
That's the concern of Adrian Gropper, MD, a Massachusetts physician (and chief technology officer of the nonprofit organization Patient Privacy Rights) who argues that Direct is merely "paving the cow path of our current system" as he put it in a comment to a story I wrote in April about ONC's grant of $280,000 to DirectTrust.org, and other $200,000 to the New York eHealth Collaborative, to act as HIE Governance Entities to support their HIE efforts and promulgate use of Direct nationwide.
Gropper says Massachusetts already is the most consolidated healthcare state in the nation, with 80 percent of care aggregated into three hospital systems and three insurers, when they should be giving more power to independent physicians to refer around high-cost providers.
Both Marcus and Gropper are concerned that the HIEs being deployed by states and by vendors won't give physicians discretion to send messages under the authority granted to them by their medical licenses, but instead rely solely upon the aegis of the HIE or vendor itself.
And yet, the way Direct address are being allocated is very similar to the way doctors acquire the means to send e-prescriptions, counters Wes Rishel, an analyst at Gartner and longtime participant on ONC's HIT Standards Committee.
Gropper is also convinced that doctors should be able to self-certify their own security credentials in a way that deployment of Direct currently does not allow. "Direct is based on e-mail," Gropper tells me. "You don't need HISPs at all. You can run Direct using Mozilla Thunderbird on a $35 Raspberry Pi to do the encryption and decryption and white lists. The mail servers can be blind intermediaries with no filtering or encryption function if you want."
Yet, Rishel and I share concerns that such a scheme opens a massive hole for fraud. "If Sam's Endoscopy Club wants to self-certify, would you as the physician be happy to send your patient's data there?" Rishel asks. Conversely, he asks, "Would you as a physician only want to communicate with those physicians you know personally?"
Rishel and others say HISPs are necessary in order to scale secure messaging to the dimensions necessary to enable widespread trust even among providers who have never met and otherwise know nothing about each other.
Under Gropper's model, "The HISP would be instructed to keep track for each physician of who else the physician trusted," Rishel says. "For Massachusetts alone that would require keeping track of 319,600,000,000 trust relationships. While the database of [that many rows of information] for Massachusetts is feasible (although not cheap) with today's process, the administrative task of setting it up is imponderable."
Gropper's response: "Even if you do want to centralize things at the server, 10,000 physicians on a server would have 10,000 rows with only a few entries in each row. Even more, any reasonable institution would typically allow any physician to whitelist every physician in the other institution. This is how faxes and postal mail work."
Both Rishel and ONC Chief Scientist Doug Fridsma contend that Direct was never intended to treat individual physicians, or patients for that matter, as equals in the chain of trust, and always meant to rely upon HISPs as the keepers of whitelists and blacklists and all the other things that service providers do to create the chain of trust.
"Direct is an important way of exchanging information securely, but we should never let our technology and other things like that get in the way of patients having access to information that is rightfully theirs," Fridsma says. At the same time, ONC judged that providing individual certificates for providers and patients was "too onerous" and "very challenging to scale," he adds.
Fridsma and ONC seem open to tweaking health information exchanges to deal with the concerns of Gropper and others. Meanwhile, however, vendors have baked their own solutions into Meaningful Use Stage 2-compliant software, and states continue to build out their HIEs.
I have a feeling that very soon, we will see just how warranted the concerns of the Massachusetts Medical Society turn out to be.
An announcement from CMS on Meaningful Use implementation deadline changes obscures the fact that the agency fully expects all Stage 2 start dates to remain unchanged.
For a few hours last Friday, healthcare providers thought they were going to get some relief from next year's Meaningful Use Stage 2 implementation deadlines. But what observers are calling a poorly worded announcement from CMS hid the fact that the agency fully expects all Stage 2 start dates to remain unchanged.
Instead, the CMS announcement proposes to push back start dates for Meaningful Use Stage 3, by adding a third year of Stage 2 for those providers who attested for Stage 1 in 2011 or 2012.
Organizations which attested to Meaningful Use Stage 1 during the first year, 2011, and the second year, 2012, are still expected to attest to Meaningful Use Stage 2 during 2014. Those attesting to Stage 1 this year are still expected to attest to Stage 2 in 2015.
What CMS' proposal would mean is that start dates for Meaningful Use Stage 3 attestation would commence not in 2016, but in 2017 at the earliest.
Confusion sown by the Friday announcement even had HIMSS erroneously congratulating CMS for a Stage 2 delay that in fact was not in the cards. On Saturday morning, HIMSS put out a corrective statement expressing the same concern that CHIME had expressed the night before.
While CMS did not issue any clarification, officials from ONC as recently as last month stated there would be no changes in Stage 2 compliance deadlines, although those officials did say the decision was ultimately CMS's to make.
But last Friday's announcement threw the industry into confusion that lingered through the weekend.
"They could have easily said, this has no effect on the start date and implementation requirements for Stage 2 in 2014, which is what our concern is, but they didn't say that," says Russ Branzell, CEO of the College of Healthcare Information Management Executives (CHIME).
Instead, Branzell says, CMS is proposing the same sort of flexibility for Stage 3 that it provided for Stage 1, while denying any further flexibility on a Stage 2 start date.
"We still believe there is some opportunity for flexibility with them for the beginning of Stage 2," Branzell says. "It's just going to require some hard work."
Toward that end, Branzell will be in Washington D.C. Monday and Tuesday meeting with CMS, legislative, and administration officials to once again press providers' plea for relief from being required to comply with Meaningful Use Stage 2 in 2014.
With ICD-10-compliant technology and processes required to be paid for care reimbursement after October 1, 2014, providers may simply be unable to comply with Stage 2 requirements in 2014, "which is even a greater concern on our part," Branzell says.
"If we mess up Meaningful Use as an organization, just pick any organization out there, that hurts," Branzell says. "You [don't] get your reimbursement from the government. You mess up ICD-10, you don't get paid. If you don't get paid for a few months, with today's cash flows in a lot of these smaller organizations, [your] viability's at risk."
In the wording of its Friday announcement, CMS may have been trying to make it appear that it was trying to satisfy the needs of the provider community without really doing so, says Laura Kreofsky, principal advisor at Impact Advisors, a healthcare IT consulting firm.
Kreofsky referred back to HHS Secretary Kathleen Sebelius' December 2011 announcement delaying the start of Stage 2. "It was, in my opinion, a more clearly worded press release," she says.
"Whether it was intentional or not, the initial ripple was very easy to have misconstrued," Kreofsky says. "The message has got to be in the provider community: Nothing's changed around Stage 2, and that everybody needs to stay the course under their Stage 2 or Stage 1 work.
"CMS purported to do this to basically make sure we're all ready for Stage 3. It was more defensive on the part of CMS than responsive to the needs of the provider organizations and vendor community."
Friday's announcement also introduced another potential complication which could make the job of healthcare system CIOs even more complicated, Kreofsky said, referring to CMS's call for frequent updating of criteria under its HIT Certification Program.
According to CMS' announcement, "This approach is designed to provide public input on policy proposals, enable our certification processes to more quickly adapt to include newer industry standards that can lead to greater interoperability, and add more predictability for EHR technology developers. We also anticipate that this new approach would spread out over a longer time period the certification requirements to which EHR technology developers have previously had to react."
The first step under this new approach would be to publish a proposed rule for a 2015 Edition of certification criteria.
"We intend for the 2015 Edition certification criteria to improve on the 2014 Edition certification criteria in several ways," the CMS announcement states. "We expect the 2015 Edition would be responsive to stakeholder feedback; would address issues found in the 2014 Edition; and would reference updated standards and implementation guides that we expect would continue momentum toward greater interoperability.
CMS also stated that it expects to propose that the 2015 Edition would be voluntary in the sense that providers participating in the EHR Incentive Programs would not have to upgrade to 2015 Edition EHR technology, and no EHR technology developer who has certified its EHR technology to the 2014 Edition would need to recertify its products. CMS' intention would be for the 2014 Edition to remain the baseline certification criteria edition for meeting the Certified EHR Technology definition.
Kreofsky questions the need for the proposed voluntary 2015 Edition. "Why wouldn't you push people to use the 2015 Edition?" she asks. These criteria won't come soon enough to be of any use to software vendors struggling to meet the 2014 criteria, and will add another layer of complexity to existing healthcare IT challenges, she says.
"For an organization that upgrades their EHR in the middle of a reporting period, it's incredibly confusing just to manage and keep track of which edition of the criteria you're at, and to have them changing constantly isn't going to help providers and organizations," Kreofsky says.
She also notes that the period for public input into many of these proposed rules will hit next fall, at just the time when the pressure is on to comply with ICD-10 requirements and, for those who attested for Meaningful Use Stage 1 in 2011 or 2012, to migrate to Stage 2.
"It takes a considerable amount of time and energy to review and respond to proposed rules," Kreofsky says.
Patient survival of ICUs is 26% more likely when backed by eICU services, and patients are discharged faster from the ICU and the hospital, a new study finds.
New research supports telehealth within intensive care units as improving patient survival rates and speeding discharge.
"A Multi-center Study of ICU Telemedicine Reengineering of Adult Critical Care" examined the impact of remote intensive care units (eICUs) on nearly 120,000 critical care patients who were treated at 56 intensive care units, 32 hospitals, and 19 health systems over a five-year period. The research demonstrated reductions in both mortality and length of stay. The results were statistically significant on both an unadjusted and severity-adjusted basis.
The study, which was published Thursday in CHEST Journal Online First, was authored by Craig M. Lilly, MD, professor of medicine, anesthesiology, and surgery at the University of Massachusetts Medical School and Director of the eICU Program at UMass Memorial Medical Center in Worcester, MA.
"The number one thing was that either your care plan was created by an intensive care specialist [working remotely], or within one hour of you arriving at the ICU, an intensive care specialist went over the plan and made sure it was right, almost like a second opinion if you will," Lilly told HealthLeaders Media. "These [eICU] systems generate an awful lot of information about performance, and folks that use that information in nearly real time to improve their performance, to understand what they were doing, and how to do it better, had a lot greater improvement than people that reviewed it quarterly or annually or [only] when they had a problem."
Participants were drawn from 19 healthcare systems participating in Philips Healthcare's eICU program. The eICU program is a technology and clinical reengineering program that enables physicians and other healthcare workers at a centralized eICU center to provide around-the-clock care for critically ill patients. The eICU program utilizes bidirectional audio/video technology, population management tools, clinical decision support, real-time and retrospective reporting tools, and targeted process redesign.
The key findings of the study were that, compared to patients receiving standard ICU care, patients who received their ICU care from a hospital that utilized the eICU program were:
26% more likely to survive the ICU
Discharged from the ICU 20% faster
16% more likely to survive hospitalization and be discharged
Discharged from the hospital 15% faster
Also important was whether clinicians responded to alerts and alarms within three minutes, Lilly says. "The bedside people do it just fine most of the time," Lilly says. "But occasionally they're busy and diverted and they're not paying attention as they need to. The offsite team can help to make that happen, and that decision support piece was important."
Another key element: using telemedicine to enable multidisciplinary rounds to help busy doctors and nurses talk to each other at the right times, Lilly says. "That ability to communicate with one another and do decision-making in a multidisciplinary way is also a form of decision support," he adds.
When asked if telemedicine alone would have the same benefits as adding decision support, Lilly says "the more of those things you did, the better you got."
While a paper and pencil is still better than nothing in clinical care, "having an offsite team that double-checks to make sure that the order's right, that the nurse is doing it right, and that the patient actually got what they needed to get, makes a difference," Lilly says.
The American College of Chest Physicians, which publishes CHEST Journal Online First, consists predominantly of pulmonologists and pulmonary care doctors, thoracic surgeons, anesthesiologists, and cardiologists. Philips officials say their eICU product currently supports approximately 12% to 13% of all adult ICU beds in the United States.
Seven healthcare IT issues that divide IT leaders, users, and patients, including Healthcare.gov, Epic Systems, ICD-10 and SNOWMED CT, and the BYOD trend.
It's time for those stories that look back at 2013. What better place to start but with those healthcare IT issues that divide us and create lots of topics for journalists like me to pick apart and try to put back together.
My top seven topics:
Healthcare.gov.Its problems were predictable, but is it salvageable? Lipstick on a pig? Now that the calendar has turned to December, mass numbers of signups must materialize for the site to be called a success—especially now that insurers have cancelled millions of old healthcare plans. The new site may do a good job (finally) of showing consumers how much each healthcare plan will cost, but as critics have pointed out, it still fails to make it simple to see exactly what you are buying for that money. How about a section showing the minimum coverage of each health insurance policy? Oh, and while we're at it, let's put all of this in plain English.
23andMe. The FDA's crackdown last week on this human genome test kit that arrives at your house in an Amazon-style box has split my friends seemingly down the middle, irrespective of their level of technical knowledge. Libertarians in general argue that the FDA should get its cotton-pickin' hands out of 23andMe's business; consumer privacy advocates rail against seen and unseen third parties getting their cotton-pickin' hands on our most personal of data; and healthcare cost-control advocates wring their cotton-pickin' hands over the medically unadvised surgeries and tests prompted by social media–driven scientific theories that don't always square with science.
Epic. Those who adopt this market-dominating EHR are legion, and some of them are willing to enumerate in excruciating detail the benefits of it, as will become clear when you read my cover story in the January-February issue of HealthLeaders magazine. Of course, those providers not in the Epic camp worry about being frozen out of the Epic ecosystem. Meanwhile, more and more doctors in small practices are waking up to find that every major hospital in town is running Epic, forcing them to make a tough decision—stay free of Epic—or an expensive one—join the crowd.
ICD-10 vs. SNOMED CT. Yeah, I know these technologies are supposed to solve different problems, complement each other, work together like peanut butter and chocolate. But in researching SNOMED CT, I am continually struck by how often ICD-10 dominates the conversation, and how many doctors are somehow expected to learn more than humanly possible about ICD-10 codes—all because ICD-10, not SNOMED CT, will pay the bills come October 1, 2014. Meanwhile, SNOMED CT addresses not just disease classification but also a range of clinical diagnoses and procedures. Clever software is out there that lets docs comfortably select medical concepts and terms that then generate SNOMED CT and ICD-10 codes under the covers. It seems too many CIOs are busy simply trying to turn the ICD-10 crank, and won't have enough SNOMED CT codes to show meaningful use Stage 2 auditors round about the same time. No wonder CHIME is screaming for relief from MU Stage 2.
Nurse practitioners and health coaches versus doctors. This one depends on whether you're still billing fee-for-service or if you've moved to risk-based payment. More and more EHR technology enables NPs and health coaches to flag and field patient care and follow-ups, medication management, or various and sundry check-ins, without bothering the doctors. But some state regulations still require doctors to sign off on virtually everything, creating a workflow traffic jam. Still other doctors are control freaks and, rules or no rules, aren't going to feel good about not micromanaging these workflows. Just because technology enables distributed care and decision-making doesn't mean healthcare is ready for it.
Patient identity. This one won't go away. We're all another year older and no closer, really, to a solution that protects privacy, assures authentication, enforces authorization, and doesn't require an IT forklift upgrade by the healthcare industry. The federal ONC is making some new baby steps in this area, but advocacy groups such as Patient Privacy Rights are watching the smallest of these gestures nervously, fearing that the very sort of biometric authentication that in theory could make healthcare ID fraud a thing of the past could also be used to harvest patients' PHI, subvert the intent of HIPAA, and exert control over the autonomy of independent physicians.
Bring-your-own-tech. Now that Amazon is delivering Microsoft desktops via a Chrome browser, the genie is well and truly out of the personal cloud bottle. Any healthcare worker who has a browser potentially is able to store PHI in the cloud. The opening BYOD debates were just skirmishes. The big battle, with Amazon and Dropbox in one corner and data loss prevention technology in the other, is bound to make its way to the square ring in 2014.
What polarizing topics are on your list? Meaningful use Stage 3? Clinical quality measures? Problem lists? Google Glass–distracted doctors? Zombie fax machines? Health information exchange business models? Please chime in below. One thing's for sure: there's no shortage of polarizing IT topics to kick around at this year's Christmas parties.
Computerized physician order entry is a key requirement of meaningful use stage 1. Now that CPOE has reached a tipping point, healthcare IT leaders are at the beginning of a journey that takes the accumulated wisdom of a healthcare provider and codes it into decision-support systems.
This article appears in the November 2013 issue of HealthLeaders magazine.
A key requirement in stage 1 of meaningful use was that providers employ computerized physician order entry for at least 30% of medication orders entered by any licensed healthcare professional who can enter orders into the medical record per state, local, and professional guidelines.
Now that more than 80% of eligible hospitals have received reimbursement for adopting electronic medical record technology, CPOE has gone mainstream. It is also in some ways just the beginning of a journey that takes the accumulated wisdom of a healthcare provider and codes it into decision-support systems that, over time, make CPOE smarter.
"We've got much higher CPOE adoption than we had three or four years ago," says Bill Spooner, senior vice president and chief information officer of Sharp HealthCare, a not-for-profit regional healthcare system based in San Diego with 1,735 licensed beds at four acute care hospitals.
Providers such as Spooner observe that some of the productivity-enhancing technology that fuels CPOE is the same technology that gets criticized as potentially illegal shortcuts for providers.
In September 2012, U.S. Attorney General Eric Holder Jr. and Health and Human Services Secretary Kathleen Sebelius sent a letter to the American Hospital Association and other industry groups, stating that "false documentation of care is not just bad patient care, it's illegal … a patient's care information must be verified individually to ensure accuracy; it cannot be cut and pasted from a different record of the patient."
Spooner acknowledges cut-and-paste probably does get used in some situations. "If that's all that's happening, that's totally wrong; but if in fact the patient's condition hasn't changed in 24 hours, then do you want to write all this stuff, or do you want to use cut-and-paste?" Spooner says.
Templates are smart productivity-enhancing tools, Spooner says. And while he says the Holder-Sebelius letter "was kind of overreaching," he adds, "At the same time, there's no excuse for somebody using it to cheat the documentation in any way."
Meanwhile, rural hospitals such as 100-staffed-bed St. Claire Regional Medical Center in eastern Kentucky still struggle just to get CPOE and electronic medical record systems adopted by doctors.
Six months after implementing CPOE, "we have some physicians … who just can't get past that point of using a computer to take care of a patient," said Randy McCleese, CIO of St. Claire Regional Medical Center, speaking at a July 2013 hearing of HHS' HIT Standards Committee. "We have had huge issues with some of those physicians being able to use a computer to take care of that data rather than writing it down."
And yet, if EMRs have achieved anything so far beyond producing incentive checks, it is probably the banishment of illegible physician notes, which have been the scourge of healthcare since its inception.
"I think the biggest benefit of CPOE as it stands is that the orders are legible and clear, and it's menu-driven," says William Park, MD, senior general surgeon and former chief medical officer at North Hawaii Community Hospital, a 33-bed rural hospital in Waimea.
"Truly the biggest challenge with CPOE is it's not terribly intelligent," Park says. "It's not a time-saver, nor necessarily in my opinion a quality improver. It's become more of a clerical issue in legibility and efficiency of the systemswide approach to clinical orders, but I don't truly think that it's delivered the benefit to the extent that everybody thought it would. There's a lot of room for improvement."
Park says the way forward is for providers to campaign for better-designed, more intuitive, straightforward, simple CPOE.
For example, when a physician is preparing an order for IV fluids, "If you're writing something that's inappropriate for the patient's admission chemistry, the system ought to tell you immediately: 'Do you really want to do that?' " Park says. "If a patient's potassium is 3.7, do you really want to give this patient no potassium supplements in their intravenous fluids? It's not smart enough to do that, and it should be."
On the cut-and-paste issue, Park says the challenges of avoiding inappropriate shortcuts while retaining potential productivity enhancements affect his physicians as well.
"EHRs can be as dumb as the template and force you to check boxes, which makes it a chore," Park says. At the same time, if there is no box to check, physicians must still dictate or type clinical notes into the system. "It's a mish-mash that doesn't work well," Park says. At some point, Park believes that EHRs will acquire a kind of organizational intelligence with more intuitive design.
One provider's CIO says he picked an EHR, from Medhost, with just such a design. "It is a picture: You look at it, you can figure it out," says Lee Powe, CIO of Hugh Chatham Memorial Hospital, which has 81 acute care beds in Elkin, N.C. Clinicians in the emergency department have lots of visual cues. "You see a test tube with blood in it, and it means something. Graphics and pictures are good for clinical people. They see it. They get it."
Other hospitals have been up to speed with CPOE for a number of years and took the meaningful use requirements more in stride.
At Albany (N.Y.) Medical Center, two electronic medical records have been live since 2005, with the early elements of CPOE beginning in 2007, says George Hickman, executive vice president and CIO of the 714-licensed-bed teaching hospital. CPOE "is not one of those the things that's a concern for us going forward in terms of the high level of physician adoption." Incoming medical residents "expect walking in the door to have all these capabilities in there waiting, so that's the way our new doctors are," he says.
Hickman, who also serves as the current chair of the College of Healthcare Information Management Executives, acknowledges that achieving CPOE compliance in meaningful use does not necessarily achieve the greater goal of leaving paper behind.
"What really starts taking away from paper is when all of the supporting clinical documentation becomes paperless and is available only in an online form," Hickman says. "We have something in the range of 50–60 of our documentation forms automated and CPOE online. We have not yet fully implemented all standardized and structured physician documentation."
Albany Medical Center still uses dictation as a form of getting documentation into the record, Hickman says. "It can be done in a very structured way or it can be done in a very unstructured way. We provide it both ways, especially with the physician clinic supporting the EHR," he says.
At Chatham Memorial, Powe employs printer control to tame the paper monster. Working with printer supplier Ricoh, Powe removed many printers from the hospital, and put technology in place to figure out what is still being printed and by whom.
"I have no idea what some of these documents are, but I got the person working for me who does, so I'll say, 'Hey, go find out where this is coming from and why is it coming out of the system instead of staying in the system,' " Powe says.
Despite best efforts, CIOs might feel a little like they're playing whack-a-mole as paper notes continue to pop up. But eventually, physicians will be in the habit of doing it electronically.
"Now in your beginning stages, no matter what software product you implement, everybody will whine, moan, and groan," Powe says. "Three months later, you turn it off for a software patch or something. Everybody will whine, moan, and groan, 'What do we do now? The system's never down.' "
But while that presents its own challenges, the fact that physicians now can't live without the EHR was exactly the point of meaningful use in the first place.
Reprint HLR1113-6
This article appears in the November issue of HealthLeaders magazine.
In our annual HealthLeaders 20, we profile individuals who are changing healthcare for the better. Some are longtime industry fixtures; others would clearly be considered outsiders. Some are revered; others would not win many popularity contests. All of them are playing a crucial role in making the healthcare industry better. This is the story of Fred Trotter.
This profile was published in the December, 2013 issue of HealthLeaders magazine.
"I always did everything I could do to get my work out there in front of people who I thought were smarter than I was."
At an impromptu lunchtime gathering in Silicon Valley's Computer History Museum, Fred Trotter is drawing lines and boxes on a whiteboard. From around the room, other programmers, healthcare IT experts, and a privacy expert or two toss in suggestions. Trotter excitedly adds to the drawing, challenging some suggestions as he goes, accepting others, and even raising a skeptical eyebrow at his own work.
It's all in a day's work—and play—for the author of Hacking Healthcare, who this May day was helping lay the groundwork for a crowdfunding project known as Hacking HIPAA, which tries to circumvent ponderous standards bodies, sluggish large healthcare giants, and secretive startups to bring HIPAA into the 21st century by clever coding and collaboration.
Trotter will be the first to tell you this work sure beats trying to change things by legislation. If anyone can pull it off through the wisdom of crowds, it's Fred.
"More and more I have been doing things where somebody decides to pay my bills for a little while, and I go and do what I think is the most interesting thing ever," Trotter says.
One of the latter inspirations got its initial crowdsourced funding of $45,000 in 2012. DocGraph, the result of a Freedom of Information Act request, was Trotter's first foray into crowdfunding liberated health data. It started with a set of data from CMS he liberated that contained 60 million referral relationships, nearly 80% of the referrals doctors make to each other in the United States.
He released the data set to attendees of Strata Rx, a healthcare data analytics conference, and reverted to a Creative Commons license. The goal: To create multiple doctor-rating algorithms that patients can use and doctors find fair. "The current credentialing system is abysmal," he says.
DocGraph didn't stop with referrals. Analysts could use Trotter's Freedom of Information Act–obtained data to see which hospitals have poor central line infection rates, and more. A few months later, without further FOIA prompts, CMS stepped up its own release of hospital quality and pricing data—and Trotter's early breakthrough probably shares credit for making that happen faster.
"We are not just building better data systems. We are building a new kind of mind, and that is precisely the place where you need to have ethics deeply embedded," Trotter said at his O'Reilly's Strata Rx 2012 talk.
For all his iconoclastic tilting at medicine's windmills—Hacking HIPAA failed to receive its needed crowdsource funding by its initial August 2013 deadline—it might surprise you to hear that healthcare IT is the Trotter family business, if you will. Trotter's grandfather owned a franchise for Medical Manager, which Trotter says was a very popular practice management system in the 1980s. His aunt and uncle also worked on Medical Manager as well.
But Trotter took a side trip or two before settling into healthcare IT. His first job was doing Internet security for the Air Force Information Warfare Center as a contractor. During a subsequent set of lucrative Internet jobs, he met current collaborator Ashish Patel, an expert at security networking infrastructure of large hospitals.
The return to healthcare was prompted by Trotter's mother being diagnosed with ovarian cancer. "It's almost always progressed too far by the time you even know it's there, which is exactly what happened to my mother," he says. In short order, Trotter made major life changes, getting married, joining the Marine Corps Reserve, and taking over research and development at his family's business.
"The first question I asked of health IT was, what kind of system would need to exist that would be able to make a better guess about my mom's symptoms than what the doctor did?" Trotter says. Now he realizes he was imagining a system similar to IBM's Watson, emerging computer technology that understands natural language and reads vast amounts of healthcare data to generate hypotheses and different probabilities of various outcomes.
"I realized that there were just so many Lego pieces beneath Watson that needed to be built, that I was probably never going to get to that in my career, and I haven't," Trotter says. "I haven't even come close to those kinds of aspirations, but I had a significant impact on all these lower-level Lego blocks."
Then there was the process by which Trotter and coauthor David Uhlman arrived at the title for their book, Hacking Healthcare: A Guide to Standards, Workflows, and Meaningful Use.
"We debated that issue, who was going to be pissed off by this book title," Trotter says. "We spent a lot of time thinking about that, and then we came to the conclusion that all the right people would be pissed off by that title. So we decided that we would move forward under that rubric. So yes, there are people who don't like the title and don't like the thesis, and in general, I find that completely and utterly okay."
Trotter is bullish on the future of crowdsourcing healthcare IT. "I just realized, way back when I was doing my very first systems, that I wasn't the smartest person I knew who was working on the problem I was working on, and that I wanted to collaborate and be ruthless about collaboration," Trotter says. "I always did everything I could do to get my work out there in front of people who I thought were smarter than I was."
Like a lot of open-source advocates, Trotter struggles with IT seesawing between setting standards to promote interoperability and promoting innovation through open-source projects. "It's a curve ball for the whole healthcare system," he says. "MITRE [a not-for-profit organization that operates research and development centers sponsored by the federal government] did an analysis which showed a fairly reasonable correlation between the complexity of the clinical quality measures and the rate at which they were adopted in the EHR developer community. More studies need to be done on that.
"We should be using a method of evaluating quality measures; I became a lot less confident in our system of quality measures because I don't see the evaluation happening."
To understand how Trotter thinks, you have to immerse yourself in what has made open source, with its meritocratic ways, successful in IT. "Collaboration without meritocracy is communism,"
Trotter says. "You have to have a meritocracy, so if you and I were working together, and we get to have equal votes on what ideas get implemented, then we have a problem, because you get a hundred people in a room, the lowest common denominator will always get the vote. If you look at HIMSS and the American Medical Association, any of the big associations, unfortunately, the lowest common denominator is what those organizations very frequently spew out. The exact opposite happens in the open source community."
With all the challenges still ahead for healthcare IT, count on Trotter to be raising those skeptical eyebrows a bunch more, and then getting back to work and urging others to join him.
Surgeons are justifiably excited by the possibilities of using Google Glass in the operating room. Despite Wi-Fi hurdles, privacy concerns, and speech recognition issues, this device will inevitably become another indispensable tool in healthcare.
Forget the new iPhone. The gift every healthcare technologist wants under his or her tree next month is Google Glass.
Suddenly, talk of Glass in healthcare is snowballing. It's the newest bright shiny object in tech, and we're hearing it could be a panacea for all that ails healthcare—unusable software, patients irritated by doctors glancing at their screens or iPhones, even infections spread by lack of handwashing and further spread by the vector of the keyboard.
Whoa. Not so fast.
Google Glass is undoubtedly a cool step forward in healthcare technology. For one thing, it's the latest marketing tool for the very modern hospital. Second, it's from Google – a company that has already changed the practice of medicine forever. Show me a physician who hasn't Googled something as part of their day's work, and I'll show you an out-of-touch physician.
The hardware of Google Glass itself is a wonder of miniaturization : a forward-facing 5-megapixel camera, hands-free high-definition video control, and a small prism that presents the wearer with an effect of looking at a 25-inch monitor as seen from 7 or 8 feet away.
While it has already been used in orthopedic and gastrointestinal surgeries, the University of California at San Francisco is the first to receive IRB approval for use of Google Glass during cardiothoracic surgery. Already, Pierre Theodore, MD, a cardiothoracic surgeon at UCSF, has performed 10 of 15 planned surgeries using Google Glass.
Image Handling Such is the interest in Google Glass that last week Theodore was a star speaker at Salesforce.com's "Unusual Thinkers" annual conference track, and he spoke to a packed house of several hundred.
Theodore showed a photo of one of UCSF's most advanced operating rooms, and noted with some irony that the 48-inch monitor provided for surgeons to review radiographic images was in a corner of the operating room, behind two anesthesia monitors and a storage unit. "It's not pointing towards the surgeon, so its overall accessibility at the point of care is limited," Theodore explained.
To use Glass, Theodore's team transferred these images to a secure Web site, taking care to remove all personal health information from the images before transfer. At a surgeon's voice command, Google Glass displays the image.
"We did a series of dry lab experimentations to demonstrate that hand-eye coordination and the distraction using the Glass was minimal, and that the images could be readily brought up and removed from the vision of the surgeon," Theodore says. "We sought limited use in a single specialty for the first pilot trials, and we set certain metrics and aims."
Wi-Fi and Privacy Considerations Among the early findings: operating rooms contain lots of equipment that interferes with Wi-Fi. Google Glass depends upon Wi-Fi to receive data, so data transfer protocols may need to be more forgiving, and the need for a more robust Wi-Fi infrastructure is clear. (That's a subject I'll write in another column.)
It's clear that it's still early days for this technology. Technicians are required to magnify images. "Down the road what will be important is the hands-free manipulation, so no additional technician will be required to project the images," Theodore says.
Despite the initial security precautions already taken, Theodore is concerned about Glass data that lives in the cloud, and moves over the Internet. "A lot of this information is actually transferred over networks, of which certain portions of those networks can be less than perfectly secure," he says. "That of course becomes more and more important as we think about transferring information over a whole variety of wireless networks."
A Distraction? The potential for Google Glass to be a distraction in the OR came up during the audience Q&A afterward.
"It's a really important question, because, in fact, it can potentially be extraordinarily distracting," Theodore says. "The advantage of Google Glass, at least as we've designed it, and used it in our early implementation, is that it's a voluntary decision to have the image actually show up in the field of vision, which means that when the Glass is off, it's virtually as though you're just wearing your glasses or your contacts or what have you."
So, at the moment a surgeon activates Google Glass, "you have made a frank decision that you wish to cognitively turn your attention away from the patient in front of you to the image that's displayed within your field of view, without needing to leave the operating table or in some cases even leave the operating room to see the X-rays elsewhere."
Another questioner brought up privacy and security again. "Largely the data right now does have to transfer outside of the hospital before it ends up coming back to the Glass itself," Theodore says. "So the way that we've worked about this, and the way we introduced this to our IRB, was to anonymize completely all of the data, so in principle there's no patient identifying information – names, birthdates, what have you – that can be identified in the information that's transferred out onto the network. But your point is well taken.
"Certainly as you consider trying to expose the entirety of either the medical record or a digital radiographic archive, it will be all the more important that the security protocols be essentially watertight."
Speech Recognition Needs Work Lastly, Theodore was asked what was tops on his wish list for improvements.
"From our initial pilots what I would say is that the speech recognition element of the Google Glass as it stands is extraordinarily rudimentary," he says. "With all due respect of course to the amazing work that's been done, one could certainly imagine a far more robust speech recognition system that allows a clinician really to interface much more with the content of the Glass.
"I imagine that it's going to come down the line and we have coders at UCSF who are helping us develop that at present, but I'd say, improving speech recognition would help."
Whichever way it plays out, it seems that devices like Google Glass will inevitably become another indispensable tool in healthcare. Theodore noted that doctors already use head-worn lenses for image magnification, and that Google Glass is "not too big a leap" from that.
I predict it won't be that long before the use of Google Glass in medicine will no longer be news.
In our annual HealthLeaders 20, we profile individuals who are changing healthcare for the better. Some are longtime industry fixtures; others would clearly be considered outsiders. Some are revered; others would not win many popularity contests. All of them are playing a crucial role in making the healthcare industry better. This is the story of David Green.
This profile was published in the December, 2013 issue of HealthLeaders magazine.
"They're marching to your doorstep because of accessibility and quality, and again, that's what changed the competitive landscape."
David Green might be the most disruptive force in medical devices today.
Green's startup companies are disrupting the high-end hearing-aid market and the intraocular-lens market for correcting the vision of cataract patients.
The truest form of Green's disruptive thinking is the way he proved it could be done profitably with broad implications for making medical devices more affordable for people who cannot afford them.
Green's achievements, featured on National Public Radio and elsewhere, originated in developing countries but are now coming to his native United States.
"I'm helping to set up an eye hospital in San Francisco that will take what we've learned in emerging markets and transfer it here to create a revenue model where the enterprise uses its profit and production capacity to serve uninsured Medi-Cal dual-eligible and undocumented" patients, Green says.
The story begins in 1992 when Green started Aurolab, a company dedicated to making price-busting intraocular lenses, sutures, and pharmaceuticals. The artificial lenses implanted during cataract surgery then cost $300. Countries such as India performed only 800,000 cataract surgeries annually. Working with Aravind Hospital in India, Aurolab initially offered a $10 artificial lens, and India's implant volume skyrocketed to 5 million in 2001. Today, the average price of the same implant is $3 to $4 and can be as low as $2.
"It created a whole price-competitive ophthalmic industry, not just for intraocular lenses, but for all the consumables and equipment used in ophthalmology," Green says. For instance, Aurolab's suture manufacturing lowered the cost of a box of ophthalmic suture from a "monopoly competitor" price of approximately $240 to Aurolab's $23.
Green's next startup, Sound World Solutions, has begun marketing an innovative, aggressively priced hearing aid developed by partner Stavros Basseas. The device will cost a couple hundred dollars, far less than comparable high-end hearing aids.
Now Green is about to market a high-end retinal imaging system, which usually costs around $22,000, for only $5,000 to serve low-income populations. This third company, Brien Holden Vision Diagnostics, will also bring to market a highly affordable device to detect and measure brain concussion, glaucoma, and various neurological disorders.
Although Green's innovations require plenty of inventions and patents, his real strength is in redesigning the supply chains that often add excessive margin at every step of bringing a medical device to market. By analyzing and reimagining these supply chains, Green targets ways to remove "non-value-added-margin" to keep products affordable.
"If you reach the low-income strata in a meaningful way, it increases the volume of people coming who pay the higher-margin prices, and that's what makes these programs viable and sustainable," Green says. "That's what enables them to generate the margin to be profitable and to be able to do cross-subsidization, so it's all kind of like a whole virtuous ecosystem, which is in marked contrast to the present non-virtuous ecosystems that you find in most medical ecosystems where basically you're trying to maximize return on investment at all costs."
Device makers are paying attention. In India, Medtronic is working with consulting firm Innosight to market lower-cost pacemakers through the Healthy Heart for all initiative. "I'm sort of the architect behind that," Green says. "I helped them figure out how to take a Medtronic product with different branding, and how to create a different supply chain pathway to reach lower-income patients who were qualified to be below a certain income with an affordably priced pacemaker."
The lesson Medtronic and others are learning from Green: If you reach the lower economic strata in a meaningful way, it will increase the device maker's volume of higher-margin patients as well.
Unlike the world view of various nonprofit foundations and non-governmental organizations, Green believes in the free market's ability to innovate even among vast populations afflicted by poverty.
"Where I intervene is I address the pricing disparity," Green says "How do you make sight or hearing or life itself affordable and accessible to low-income people? Everything I do, it's using the tools of capitalism. I've learned just by trial and error that the tools of capitalism work to convert need into demand and to be market driving, whereas charity doesn't work."
Green draws upon a degree in health behavior, health education, and a master's degree in public health from the University of Michigan. He has no formal engineering or economics background. "I find people with extreme technical competence in whatever the arena is, and then I work with them," he says. "I assemble teams and craft business models."
Lately, Green also arranges a fair amount of financing for startups following his startup methodologies, financing such as helping Deutsche Bank create a fund that provides debt financing for eye care programs. "The whole thing about self-financing, if you're generating more money than you're spending with the social twist, that means that you're being very efficient and effective in what you're doing," he says. "You're reaching the market with something that they want. You're creating demand. They're marching to your doorstep because of accessibility and quality, and again, that's what changed the competitive landscape."
Both Sound World Solutions and Brien Holden Vision Diagnostics were capitalized in the $10 million to $12.5 million range. "A lot of the social investing funds out there are risk-averse and only put money into revenue-generating entities," he says. "It was a lot of work to capitalize those companies, and I'd like to see funds that are more into early stage risk capital for pre-revenue for companies that can be transformative in the healthcare space. That's a really big issue."
Meanwhile, Green's innovations have touched the lives of millions of people already. "Aurolab has helped 18 million people see, and that's just with the lenses," he says. "The eye care programs that I helped develop, collectively they do something like 800,000 surgeries a year, and treat 6 million people a year."
But Green hastens to add: "It's not just me, but it's all the different groups that I've helped develop and help direct how they scale, which is really an economic model, rather than a donation model."