Institute of Medicine Agrees Health It Poses Risks to Safety. Here Are Their Recommendations

A CAT scan. Where's that big spleen?

Years ago, the Disease Management Care Blog ordered an abdominal CAT scan for one of its patients. The study was reassuringly normal except for the unexpected - and most likely benign - finding of an enlarged spleen. In the absence of any other worrisome findings, the interpreting radiologist recommended that another CAT scan should be done in six months to make sure it wasn't getting bigger. The patient agreed. The DMCB then fired up the electronic health record (EHR), scheduled the CAT scan order for six months, closed the encounter and moved onto the next patient.

Approximately one year later the patient returned for an appointment. That's when it was discovered that the CAT scan had never been scheduled. The DMCB was later told that the EHR wasn't configured to schedule tests 6 months in advance.*

It's on behalf of patients like this and their physicians that the DMCB welcomes this Institute of Medicine report on Health IT and Patient Safety. It recognizes what most front-line EHR-using physicians have known for years:

"The evidence in the literature about the impact of health information technology (IT) on patient safety, as opposed to quality, is mixed but shows that the challenges... involve people and clinical implementation as much as the technology. The literature describes significant improvements in some aspects of care in health care institutions with mature health IT.  For example, the use of computerized prescribing and bar-coding systems has been shown to improve medication safety. But the generalizability of the literature across the health care system may be limited. While some studies suggest improvements in patient safety can be made, others have found no effect. Instances of health IT–associated harm have been reported. However, little published evidence could be found quantifying the magnitude of the risk." (bolding from the DMCB)

In other words, selective reporting involving best case scenarios fail to account for the reality that the old safety issues of a paper-based system are being displaced by new safety issues of an IT system and, what's worse, we don't know the extent of the problem.

Here are the IOM's (paraphrased) recommendations to Health and Human Services (HHS):

1. HHS, working with the EHR vendors, should develop a health IT surveillance plan.

2. HHS should foster the free exchange of information and address the "legal clauses" in contracts that shift liability from the vendors to the doctors (The AMA has referred to these as "hold harmless" clauses).

3. Data is needed that allows health IT users to publicly compare and share experiences among multiple vendors.

4. A Health IT Safety Counsel that assesses and monitor health IT safety should be established.

5. A public register of all health IT vendors should be established.

6. It's time to specify those processes that reduce risk and make the vendors adopt them.

7. Health IT related deaths, serious injuries and unsafe conditions need to be centrally reported.

8. An independent federal entity should be established that is empowered to any investigate deaths, injuries or unsafe conditions.

9. HHS should report annually on its progress.

10 More cross-disciplinary research is needed to improve the design, testing and use of health IT.

Bravo says the DMCB, bravo!



*many of the facts are changed, but you get the gist

A Thousand Dollars Says Dr. Ezekiel Emanuel Is Wrong About The Long Term Prospects Of ACOs

An inflation-protected bet

Approximately 15 years ago, the Disease Management Care Blog was a speaker at a conference with an audience mostly made up of managed care leaders. It boldly argued the nation's disease management vendors were going to help put the nation's health insurers out of business by simultaneously assuming risk and lowering costs.

Hows that for chutzpah. The DMCB was never invited back, but not because it isn't an outstanding conference speaker who deserves fat fees.

It was because it was utterly wrong.

And so is this online commentary on accountable care organizations (ACOs) courtesy of the The New York Times. In it, Dr. Ezekiel Emanuel boldly predicts that by 2020, ACOs will drive health insurance companies out of business. They'll do that by assuming full risk, dropping patient barriers to care, coordinating services, fostering communication, promoting health, banning fee-for-service, increasing efficiency, relying on evidence-based care, being locally responsive and competing against other ACOs on cost and quality

Dr Emanuel is being astonishingly overconfident for four reasons.

In fact, the DMCB, thanks to years of published evidence, is so sure of the odds that he's wrong, that it is willing to bet Dr. Emanuel $1000 dollars that employers, health insurers and government programs and not providers will control the majority of the U.S. health care budget in 2020. 

If the DMCB is wrong, the money can go to any cause selected by Dr. Emanuel. If the DMCB is right, it gets to select one, like the 2020 presidential campaign of this person.

Here's why the DMCB is willing to put its money where its bloggery is:

1.Pundits have been incorrectly predicting the demise of U.S. health insurance for years. For all its problems, its ability to form huge risk pools, maintain a cushioning surplus and remain profitable has regretfully made U.S. style health insurance the least worst alternative.

Dr. Emanuel naively points to ACO risk pools made up of 15,000 members.  That's way too low and exposes the risk-bearing organization to wide swings in claims expense. As further testimony to the financial vulnerability of ACOs, California has required groups that take risk to maintain a surplus to mitigate the risk of bankruptcy.

2.  Managed care style health insurance and its utilization management approaches persist for one simple reason: it works and, what's more, even MedPAC knows it.

By the way, wasn't it the managed care insurers, not the providers, that invented the enlightened care management and coordination approaches that ACOs are supposed to adopt?  By the time ACOs get that up to speed, the insurers will be developing even newer innovations in areas like social media, gaming and wireless health management.

3. Dr. Emanuel embraces the replacement of fee-for-service by what he euphemistically describes as "a fixed amount per patient." At the height of managed care in the 1990s, that payment approach was called "capitation."

Patients distrust any form of capitation and when doctors aren't being screwed by "decapitation," the disincentives of a fixed payment stream can be problematic. What is it about ACOs that make their secret sauce better than anyone else's?

4. Not only do the low odds of success make the future of ACOs highly murky, many physician-led groups have simply been unable to simultaneously manage the competing business models of saving money and building clinical programs.

And what about quality?  While studies like this suggest integration leads to higher quality, others say the impact is more modest. And where's the proof that quality actually reduces costs?

While Dr. Emanuel may ultimately prove to be right and the DMCB will be out $1000, that doesn't necessarily mean ACO patients will be better off. Provider groups that are successful in managing capitation are just as prone to the bad behavior of insurers that they're supposed to replace.

(To be fair to the future beneficiaries of the DMCB's money, it recommends that it be parked in escrow in an inflation-protect asset class, like gold. That's because if ACOs fail and the U.S. health care budget balloons, the U.S. Treasury will be tempted to print its way out of trouble.)

The Latest Health Wonk Review Is Up!

'Tis the season for elections, so the DMCB votes for the Campaign 2012 Edition of the Health Wonk Review hosted by the Colorado Health Insurance Insider.  Enjoy!

Measuring Outcomes and Return on Investment (ROI) in Disease Management and Population Health

Changing opinion from right to left

Has marriage made the Disease Management Care Blog a better person?

While it and the spouse heartily agree on the ultimate answer, using hard numbers to prove it to a skeptical mother-in-law is a different matter. To definitively answer the question, the affable DMCB came up with some proposed measurement approaches, such as:

1. Pre vs. Post: comparing past beer guzzling to present-day chardonnay sipping;

2. Actuarial: actual vs. projected appreciation for the leather-clad vampire vixens of Underworld;

3. Comparison to a Control: the DMCB's willingness to take direction on shrubbery trimming, versus that of more docile hubbies;

4. Randomized and Prospective: (the DMCB has wisely opted to not go there).

While the DMCB continues to work on the complex methodology of marital outcomes, it is reminded of a key paradox: while we live in an "Information Age," other pressing questions - such as the extent of the Eurozone's influence on GDP, the merits of vouchers in public education and the link between the President's approval rating and his governing by remote memo - likewise defy conclusive measurement.

What's more, frustrated by our world's complexity, we ironically want fewer answers. The DMCB suggests this search for simplicity partially explains the luster of a balanced budget amendment, laws on minimum medical loss ratios, Newt's bombast, Obama's rhetoric, blanket coverage of birth control and, last but not least, single approaches to assessment of population-based programs.

Which brings the DMCB to Al Lewis' $10,000 challenge, in which he dares anyone to come up with a more accurate approach to measuring disease management return on investment. 

Al is a luminary in the disease management firmament who leads the Disease Management Purchasing Consortium. He was there at the founding of the Disease Management Association of America, led the attack on the vendors' past lazy outcomes reporting and has been instrumental in questioning the conclusions about North Carolina's Medical Home Program. He now claims to offer the only approach to accurately measuring the financial impact of disease and population health management.

Maybe, but the DMCB would like to humbly offer up an alternate perspective.

Check out this DMCB paper that simultaneously deployed three uncomplicated methodologies to assess the claims expense impact of a chronic heart failure disease management program. While all three gave different answers, they all pointed in the same direction.

That was enough for the DMCB boss to continue the program's funding.

This same overlapping and multi-layed approach also underlies the Care Continuum Alliance's Outcomes Guidelines Reports, which recommend a suite of measurement approaches that pivot on important determinants such as population characteristics, the influence of confounders as well as bias and the resources available to answer the question.

None of this should be any surprise to seasoned and prudent health administrators, physician leaders, clinical program architects or DMCB readers. They know that good actuaries use complimentary and overlapping approaches to come up with the right premium. They understand that good medical researchers demand caution, skepticism and multiple research studies before reaching any conclusions.

In other words, there is no one-time and one-size fits all approach in outcomes assessment.

All this adds up to the fact that optimum outcomes measurement triangulates on the truth. The measurement approaches advocated by the DMCB, Care Continuum Alliance members and health system leaders have been around for years, are within the reach of standard statistical software, are familiar to researchers and are highly adaptable to the circumstances of 99.99% of disease management programs, not to mention the medical home and accountable care organizations.

The DMCB says that when multiple, competing, overlapping, repeated and adaptable measurement methodologies point in the same direction, that's when regulators, consumers, purchasers, buyers, providers and patients and mothers-in-law can be really confident that they have the answer they need.

Image from Wikipedia

More On The "Death" Of Disease Management

At the email prodding of several colleagues, the Disease Management Care Blog next turns its attention to a blog posting by Archell Georgiou MD provocatively titled "The Death Of Disease Management (Finally!).  The Archelle on Health Blog contrasts the industry's early promises of evidence-based medicine plus patient self-care with the bitter fruits of non-existent savings, the disappointing Medicare Health Support (MHS) demo and a just-published anti-disease management New England Journal article. 

Using that last Journal article as the final word, Dr. Georgiou provides her readers with a graveside eulogy of what went wrong:

1. The NCQA promoted an "enroll and counsel all patients at risk" approach, when what was really needed was a focus on the patients at greatest risk.

2. The disease management industry relied on gimmicky marketing instead of reducing costs, and

3. Inflexible adherence to evidence-based medicine failed to account for individual patient circumstances.

Enter Al Lewis of the Disease Management Purchasing Consortium with a rebuttal, but, um, sort of.  Al points out that the DM industry still clings to life and, even worse, uses spurious financials.  Yet, thanks to ICD-10, inclusion of labs in predictive modeling, lower program costs, physician involvement and the marvelous insights of his soon-to-be-published book, he confidently predicts that the DM vendors will emerge from what is better described as a near-death experience.

Where to start?

Medicare Health Support's Lessons

That New England Journal article used by Dr. Georgiou to  nail DM's coffin was a really retread of Medicare Health Support's (MSH) failings.  Look deeper and what MHS really demonstrates is CMS' ineptness in demo design, program support and data interpretation.  Toss in this CBO Report and it's clear that the issue isn't whether DM "works" in in FFS Medicare, but that the administrators running FFS Medicare don't know how to implement DM. 

Disease Management Is In Use In Most Commercial Settings

CMS' bumbling stands in contrast the long term and continued reliance by commercial insurers on disease management in the commercial and self-insured sectors.  While nattering nabobs may suggest that the commercial sector isn't paying attention to the "evidence" of studies like MHS, a cursory search of the published literature can find studies like this, this, this, this and this that suggest that the commercial sector knows what it's doing and that patients and their premiums are better off for it.

Disease Management Ver. 1.0 Deserved To Die, But Didn't.  It Changed

Archelle On Health has it only partially right. When industry's own studies proved that a) an all-patient approach didn't work, b) it had to get serious about documenting outcomes and c) patients had to be "engaged" on their own terms, it didn't crawl into a hole and die.  It changed to what the DMCB has characterized in past postings as "disease management Ver. 2.0," otherwise known as "population health."

And What About Al Lewis' Insights?

The DMCB agrees with four out of five.  If his new book is anything like his last one (by the way, it included the DMCB as an author), make that five out of five.

But Wait, There's More!

In the original "Death" post, Dr. Georgiou points to six innovations that promise to further shake up the landscape of population health: social networking, gaming, remote biometric monitoring, wireless health management apps and "passive" environmental changes that lead denizens to lead more healthy lives.  The DMCB says she's right and finds the topic endlessly fascinating.  While the population health and disease management industry is using monitoring and social media, it remains to be seen how it will embrace the other innovations. 

That being said, the DMCB hopes that Medicare doesn't run any demos on any innovations like gaming, apps or environmental changes, lest future bloggers end up pronouncing their death too.

Image from Wikipedia

Congressional Budget Office Concludes Medicare Remains Incapable of Implementing Disease Management and Care Coordination

By now, many readers may be familiar with the recently released Congressional Budget Office (CBO) report Lessons from Medicare's Demonstration Projects on Disease Management and Care Coordination. 

The Disease Management Care Blog finally caught up with it. 

Here's its take:

It's a well-written and traditionally crafted summary of CMS' six demonstrations that had the participation of 34 programs:

Care Management for High Cost Beneficiaries
Medicare Coordinated Care
Disease Management for Dual Eligibles
Informatics for Diabetes Education and Telemedicine
Disease Management for Severely Chronically Ill Beneficiaries
Medicare Health Support

Why Demos?

Demos are used by CMS to test innovative care strategies before they are formally adopted as a covered benefit in the fee-for-service program. These demos were funded because CMS knows that half of all Medicare beneficiaries have five or more chronic conditions and that they account for 75% of its costs. In addition, up to 50% of beneficiaries are not receiving recommended care, such as vaccinations and cancer screening.

How Did the Demo's Work?

The programs enrolled persons with a variety of chronic conditions such as diabetes and heart failure. Kick off dates ranged from 2000 to 2005. "Hierarchical condition categories" or high baseline expenditures were used to identify and target the persons at high risk. All used care coordination nurses who were typically - but not always - remote and disconnected from the physicians. There was little information on admission criteria to the programs, on the training or credentials of the care managers, the approaches used in to manage transitions across care settings or improving medication compliance.

What Was the Bottom Line?

On average, the programs were unable to achieve savings in excess of Medicare's fees.

The programs reduced hospitalizations by an average of 7%. In order for Medicare to have saved money based on the disease management fees, the hospitalizations would have had to be reduced by 11%. Programs that involved greater in-person interactions between nurses and patients appeared to have greater success in reducing admissions rates.

The evaluation was also statistically stymied by relatively low numbers of patients in some of the programs. It is possible that there were savings, but if they exist, they were too small to be detected by widely accepted statistical methods.

So What DMCB Conclusions Can Be Drawn and Why?

There are two:

1.The "versions" of disease management used by Medicare from 2000-2005 relied on approaches used in commercial settings from 1995-2000. These deployed remotely stationed nurses calling patients with worrisome cost patterns. From 2000-2005, commercial insurers and employers knew that wasn't working and transitioned to more sophisticated engagement protocols as well as in-person care programs tailored for patients with high as well as modifiable risk.

In other words, the impact of later versions - like this, this and this - of disease management in fee-for-service Medicare remains largely unknown. That's why the CBO report ends with the conclusion that....

"...the demonstration results indicate that additional experimentation, evaluation, and refinement over a period of years will probably be needed to identify policies involving disease management and care coordination that can significantly reduce Medicare expenditures."

Of course, by the time CMS gets done with more years of "experiments," "evaluation" and "refinement," those results will likewise be obsolete.

2. This begs the question: if Medicare can't achieve savings, why do practically all insurers and most self-insured employers still include disease management programs in their benefit designs?

The DMCB suggests that in addition to a growing body of peer-reviewed literature showing that more modern (and still untested by Medicare) disease management is "efficacious," non-governmental insurers have some additional advantages that increase real world effectiveness:

1.Using disease management as only one component in an overlapping suite of other mutually supportive programs, including consumer-based incentives to rely on primary care, value-based insurance designs, intelligently restricted networks and concurrent wellness and prevention programs.

2. Access to internal assessments of program success that will never be submitted for publication in the peer reviewed literature.

3. Flexible program adaptation to the myriad local cultural and community factors that determine success. After all, what works in a motorcycle factory in the Midwest may not work in a cubicle farm in Silicon Valley.

The DMCB Bottom Line:

While the CBO report is technically correct, it promotes the tired canard is that modern disease management consists of naive nurses making blind calls and exhorting baffled consumers to take better care of themselves. In addition, the idea that single standard can be implemented everywhere in a stand-alone fashion is a pipe dream. No wonder Medicare's disease management demos have failed.

The Feds need to figure out how to implement disease management, not what doesn't work.

Spin and Innovation vs. Savings and Disease Management

Spinning is hard work!

With the 2012 political theater season underway with a faux budget rejection and a re-energized President, The Disease Management Care Blog is reminded of an inconvenient truth: the nation's unsustainable health care cost trajectory continues and is the single biggest threat to a bipartisan "grand bargain." 

So our pols' response?  Spin.  For a hot-off-the-presses example, check out this press release from CMS' Innovations Center lauding the creation of quality-boosting and cost-cutting health care "models." There's also a "one year of innovation" monograph describing hundreds of millions of dollars worth of demonstrations, initiatives, programs, partnerships and sessions.

Impressive says the DMCB, but the bottom line is that these are ongoing and innovative experiments. There are no outcomes data, cost savings are far from certain and the mainstream FFS Medicare program in place today hasn't really changed its stultifying and high cost ways.

Sooner or later in the coming months, Congress will have to agree on some sort of budget. As that moves forward, the DMCB offers up some hard "un"spun facts for consideration:

• Just as effective governing is not a matter of selecting from a series of policy options, effective doctoring is not a matter of selecting from comparatively effective treatment options.

• There is no proof that the versions of the electronic health record being currently adopted nationwide save money. The savings argument remains anecdotal or theoretical with considerable room for doubt. What is clear is that installing an EHR costs money.

• The Patient Centered Medical Home is still a work in progress. The cost savings widely reported here did not achieve statistical significance and the real truth underlying North Carolina's medical home data is mired in actuarial debates outside the peer review process.

• While ACOs are ultimately modeled on the success of large integrated systems, we know that bigger is not necessarily better.  As this multi-year experiment gets ready to set sail into politically stormy seas and if (and that's a big "if") they are proven to save money, it'll take years to expand them. Any real savings are more than a decade away.

• Commercial insurers' profits, while high in absolute terms, have a relatively low return on investment and, compared to their administrative burdens, are not one of the major drivers of health care costs.

So where do the real cost savings lie? Former White House advisor Ezekiel Emanuel, in this just published JAMA Viewpoint, points out that the best answer is not any of the notions above but but controlling chronic illness with what essentially can be described in two words: disease management.

Here's the quote:

Successful efforts seem to entail instituting at least 4 common changes: (1) installing electronic health records and using them to track patients' health status and physician performance, as well as using decision supports to increase adherence to treatment pathways; (2) using the information for more intensive interactions between patients, caregivers, and clinic staff, including use of care coordinators, 24/7 access, interventions to increase medication adherence, specialized clinic services for recurrent problems of patients with chronic disease such as anticoagulation clinics; (3) reducing use of specialists, and when specialists are involved using those who are more efficient; and (4) providing services not traditionally covered by fee-for-service reimbursement, such as e-mail, wireless monitoring to increase medication adherence, home evaluations to minimize falls, lifestyle interventions to improve nutrition and exercise, and transportation services for office visits. Cumulatively, the savings appear to occur through fewer hospitalizations, emergency department visits, and lower use of specialist services (bolding from the DCMB)

The DMCB couldn't have said it better itself.  Dr. Emanuel describes the ingredients of successful commercial disease and population health management programs in place today today: a later generation EHR that is coupled with decision support and registries, risk stratification to identify the patients at great risk, care coordination with expedited access to specialized services and support for preventive care. 

He's right, and that's no spin.

Image from Wikipedia

The Latest Cavalcade of Risk Is Up!

Choose the best answer:

You should head on over and read the latest Cavalcade of Risk at the Nothwithstanding Blog because:

1. It's a collection of the latest blog postings on a spectrum of business risk topics;
2. It affirms that no risk professional can go without regularly reading the Disease Management Care Blog;
3. It's a demonstration of the "chose the best answer" tests prevalent in medical undergraduate and graduate education programs;
4. After reading it, you'll once again be the smartest person in the universe and your co-workers and competitors will know it;
5. Nothing else is going on right now.

The good news is that if you chose ANY answer, you got it right!

Enjoy.

The Non-Surprise of Increased Physician Referrals

Big deal!

News flash! 

In the last century, American derrieres have grown by 5%! 

In the last 20 years, rates of professional athlete misbehavior have skyrocketed! 

And in the last 10 years, there's been a near doubling of physician referrals!  

If these trends surprise and amaze you, you may want to check out the up-to-the-minute news outlets like The New York Times' "Prescriptions Blog."  On the other hand, regular readers of the Disease Management Care Blog may not need to go to such extremes. They are astute observers of backsides, know how to access ESPN and can confidently predict trends in health care utilization.

That being said, the DMCB thinks a recently published article on physician referrals may be worth a look.

Michale Barnett and colleagues extracted physician survey data on a sample of clinic visits from 1999 to 2009 that had been collected by the National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey.  Physicians were asked about their type of practice setting (employed vs. physician owned), their patients' demographic variables, reasons for the visit, diagnosis, treatment and disposition, including whether or not there was a referral to another physician.

The study found that the rate of referrals increased from 4.5% of visits in 1999 to 9.3% in 2009.  This increase included Medicare patients, who went from 4.2% to 9.7%.  Physicians who owned their practices were less likely to refer (from 4.2% to 7.5%) and physicians with more than 50% managed care experienced an increase from 5.5% to 8.7%.

In the Discussion section of the manuscript, the authors point out that two possible explanations for the increase in referrals include 1) the increasing complexity of health care and 2) the benefit of off-loading some of the work of patient care onto another colleague.

The DMCB agrees and would add two more explanations:

1. The increase in "PPO" vs. "HMO" style managed care during the period of study, and

2. An increase in patient consumerism.

Can recent health care reforms, like ACOs, reverse this referral trend?  While the DMCB awaits the answer to that question, it looks forward to the continuing roominess of its movie theater seats and the next juicy sports scandal.

Stay tuned!

Image from Wikipedia

A Work-Around for an Electronic Health Record Shortcoming Has Implications for Population Health Management

An EHR screenshot... but where
is that diagnosis of diabetes?

Welcome KevinMD readers.  The DMCB  blogs frequently on HIT, primary care, health reform and everything in between.

Years ago, if you were elderly, had diabetes, high blood pressure, low back pain, needed a yearly flu shot and came to see this electronic health record-enabled physician (now with the nom de plume "Disease Management Care Blog"), you would have had your diabetes, high blood pressure and low back pain reassessed, you would have been given a flu shot and, for good measure, the DMCB would have tossed in a discussion about the unpleasantness of getting screened for cancer. After the indignity of your physical exam, the DMCB would have typed its clinic notes into the EHR. Then the DMCB would have [click!] opened a new window,[click!] and "processed" your clinic encounter by selecting a [click!] "principal diagnosis" and some [click!] "secondary diagnoses," set a [click!] "level of care" and then [click!] "closed" the record.

And after all that, it's very possible that, thanks multiple diagnoses, entry fields and the press of time that the DMCB would have never selected [or clicked!] "diabetes." In other words, the diagnosis of diabetes would have been an electronic tree in the forest that no one would have heard falling.

It turns out that this is another shortcoming of the EHR that fails to get mentioned by its naive paladins.  While these weenies would have you believe that physicians can use this technology to magically retrieve any information on any sick patient around the world, the truth is that these systems are notoriously reliant on old fashioned human behavior.  Relying on standard diagnosis coding to identify every patient with every condition remains notoriously inaccurate. 

That's why work-arounds have become necessary.  An interesting one is the Centers for Disease Control's "BioSense." It uses the words or text that are entered into an electronic record's chief complaint "field" in emergency rooms and clinics.  Combinations of certain keywords are consistently associated, for example, with influenza and can even act as an early warning system that heralds an outbreak of the disease.  Using text is also the logic behind Google's ability to gauge the presence of disease among users of its search engine.

With that background on the EHR, its coding travails and influenza, members of the population health management community may want to pay attention to this Mayo Clinic Study was recently published in the Annals of Internal Medicine.

"Why?" you ask?  Read on.

Persons who had been laboratory tested for the presence of influenza comprised the study population. 1455 persons had the virus while 15,788 did not.  Of the 1455 persons with influenza, 1203 had an encounter recorded by a provider at the same time as the test.  The encounters were the typical "free text" typed notes that were composed by providers and included a history (including patient symptoms) and a physical examination (such as the presence of a fever or a red throat). 1455 patients with a "negative" test (i.e., no virus) were selected as control patients and of these, 905 had a recorded typed encounter. 

The providers' free text encounter records were then scanned using a "Multithreaded Clinical Vocabulary Server" (MCVS) system. As the DMCB understands it, this looks for certain key words, terms and phrases in the providers' notes as well as associated x-ray reports and lab tests that, based on prior studies, seem to be associated with the presence of the influenza virus. The extract (dubbed a "synthetic derivative") was then downloaded into "SNOMED CT," which acts as a medical dictionary of medical concepts that can be used in regression algorithms to predict the presence or absence of influenza.  That prediction ("the patient has influenza") was compared to the gold standard of the viral testing ("influenza virus really found").

How did MCVS do vs. the lab? By combining certain terms in the records ("fever" and "cough" for example) and excluding others (an abnormal chest x-ray, which is unusual in viral influenza) the authors found, compared to viral testing, a remarkable degree of accuracy: using a receiver-operator characteristic curve that reconciles testing accuracy over a range of assumptions, the rate for MCVS exceeded 90%.  In other words, the description of the illness spotted the disease prior to any lab testing.

While the Annals paper focused on the ability of MCVS to act as an early biosurveillance warning system that can spot infectious disease outbreaks early and and accurately, the DMCB was intrigued by the implications for the population health management community and the care of persons with chronic disease.

Here's why:

1. With an accuracy rate of more than 90% for spotting persons with influenza, it's possible that systems like MCVS will likewise be able to identify those with known diabetes, COPD or heart disease who fail to have their condition officially tagged and recorded as a diagnosis. With systems like MCVS, health management providers will  be able to overcome the shortcomings of the EHR and obtain a complete picture of their populations' disease burden
 
2. What's more, systems like MCVS may, on the basis of a scan of multiple encounter notes about risk factors (for example thirst, excessive weight and a suggestive family history), be also able to spot persons with undiagnosed conditions.  Furthermore, other factors could be used to prospectively identify those persons at greatest risk for future complications, such as an avoidable hospitalization. In other words, getting access to the medical records would be another step forward in the still evolving science of predictive modeling.

3.  The DMCB also wonders if systems like MCVS will eventually be used to define and submit the diagnosis codes for the provider.  Not only would this unburden the physicians from what is an administrative hassle, it'd probably be ultimately more accurate.  Combine that with the detail of ICD-10 coding (assuming we can actually implement it), and we may finally be on the cusp of fully understanding the care needs of populations and be that much closer to collecting on the still-unfulfilled promise of the EHR.