Joel Diamond 4/22/09

Why We Need Natural Language Processing

Chief Complaint: “Are my testicles black?”
Operative Note: “The patient was brought to the operating suite. She was propped and raped in the usual fashion.”

Granted, these anecdotes came from colleagues using much earlier versions of voice processing software, but they show the inherent need for voice understanding.

Oh, I forgot to tell you the actual dictation:

Chief Complaint: “Are my tests all back?”
Operative Note: “She was prepped and draped in the usual fashion.”

Most long-time EMR users will agree. For documenting the ROS and physical exam, drop-down lists and templates, particularly when the users have customized them around their own workflow, have proven to be incredible time-savers. The problem is that History of Present Illness and Assessment/Plan are best expressed using free text. Moreover, these are the most critical parts of the record, as they  reflect nuances and thought process associated with the true art of medicine. 

Since my typing skills have improved drastically, I continue to express these areas with significant detail. To me, the thought of using voice recognition software just seems like a less efficient move from dictation. I know that there are many of you out there who have enjoyed tremendous cost savings and efficiencies, particularly with the vastly improved newer versions of this technology, but I’m holding out for something better.

Natural Language Processing (NLP) is a method that transforms text into structured data.In essence, it understands text. Clearly our growing requirements for reporting and analytics will make this technology essential, yet it continues to be discussed mostly in academic circles. Significant advances have been made in this technology and I believe that incorporation would enhance EMR adoption. Without it, I worry that we will continue to add unstructured (i.e. unusable) data into the collective medical record.

I would love for other readers to comment on this subject. (Equally important: my immature and irreverent side would love to hear more anecdotal voice-processing gaffes like the ones above).

 

Joel Diamond, MD is chief medical officer at dbMotion, adjunct associate professor at the Department of Biomedical Informatics at the University of Pittsburgh, and a practicing physician at UPMC and of the Handelsman Family Practice in Pittsburgh, PA. He also blogs on interoperability.

News 4/21/09

Walgreen pharmacies fill a record 3.1 million electronic prescriptions in March, which is a 211% increase over the previous year. Still, this number represents only about 15% of all eligible prescriptions.

A Colorado doctor is found guilty of practicing medicine in California without a state license after prescribing an antidepressant over the Internet to a patient who later committed suicide. The patient had completed an online questionnaire, but the two never talked directly. The doctor has been sentenced to nine months in jail. His attorney says that, “Telemedicine is now dead,” since it is not feasible for physicians to be licensed in every state. Nice grandstanding, Mr. Attorney. Perhaps the conviction was for practicing without a license, but isn’t the bigger issue that the doctor was providing inadequate patient care, presumably with limited oversight?

McKesson kicks off its 11th annual McKesson Community Days, a program to bring employees together to participate in large-scale volunteer projects. Over 14,000 McKesson volunteers are creating care packages for hospitalized veterans at VA medical centers. The packages included a hand-written note of thanks, along with a fleece blanket and personal care items. Well done.

The administrator of the 16-physician Florida Heart Associates claims its transition to EMR has been worth the effort and cost because doctors are spending less time at work and reimbursements are up. The group took 12 to 18 months to review EMR systems and spent about $600,000 initially. After a two and half year transition, the group is now paperless. The administrator estimates the group now spends $11,000 to $14,000 a year per physician for support and updates. No word on which software they use, though it was supposedly developed by cardiologists.

Lee Memorial Health System (FL) and area doctors discuss interoperability and the exchange of patient data, a conversation that will undoubtedly be repeated in medical communities across the country in coming months. Even with the potential for federal funds, some community physicians are hesitant to partner with the health system and utilize Lee Memorial’s Epic system. Others wonder how they’ll be able to tie their existing EMR with the hospital’s system and who will pay for the connection. Everyone sees the value of connecting patient data, but today most communities lack a clear plan to make it happen.

Since launching its e-Prescribe Tennessee effort in December, the Tennessee Office of e-Health Initiatives has awarded $14.6 million in grants to 1,830 Tennessee providers. The $2,500 to $3,500 grants subsidize the cost of the required e-prescribing hardware and software.

A jailed Arkansas physician accused of illegally possessing of hand grenades sues state medical board members for persecution. Dr. Randeep Mann, who names both present and former board members in the suit, claims he was persecuted over the past six years, at least in part because he is Hindu from India rather than white, Christian, and native-born. Mann has been in jail since March 6 when he was arrested on firearm charges.

Federal agents raid two Indiana medical offices and three physician homes in connection with a Medicare fraud investigation. The raids follow a lengthy HHS investigation of two doctors affiliated with Doctor’s Hospital in Bremen. The physicians are suspected of billing fraud, improper referrals, and illegal kickbacks to physicians.

Sharp Community Medical Group (SCMG) in California announces that its first physician is live on the SCMG CONNECT EHR system. SCMG is working to establish a shared, single patient record for its 600 independent primary care physicians.

Top military doctors in Iraq aren’t happy with its new tracking system for wounded soldiers. The new system, Theater Medical Data Store (TMDS) replaced Joint Patient Tracking Application (JPAT). One of the main complaints involves data loading time (a minimum of three hours to load from the field and into servers in Virginia, and then to the physicians at medical centers in Germany.) Also a concern are the multiple information links for each patient, which make easy interpretation difficult. An army vascular surgeon sounds like his civilian counterparts: “I know JPTA is dead, but our current system is not functional. As we do more with less putting the administrative burdens on the doctors is ludicrous.”

A University of Texas School of Public Health study finds that the uninsured in Harris County are visiting hospital EDs for minor illnesses less often because of improved access to primary care programs and clinics. Over a three-year period, Harris County hospitals saw a 5.5% decline in the percentage of primary care-related visits to EDs by the uninsured. Diverting these patients to clinics saves an estimated $10 million per year.

Cardiac Science Corporation completes certification to connect its HeartCentrix ECG connectivity software to the Sage Intergy EHR.

The FTC issues a proposed rule requiring PHR vendors and related entities to notify consumers when their identifiable health data has been breached. Public comment is open until June 1.

The Vermont legislature explores the use of smart cards for health insurance, medical billing, and EMRs.

A Chinese man accidentally swallows a pair of nail scissors he was using as a toothpick. Doctors surgically removed the scissors, which had entered the man’s esophagus. If this had happened in America, you know this guy would have been asked to be the next spokesperson for Oral-B toothbrushes.

E-mail Inga.

Intelligent Healthcare Information Integration 4/17/09

Two Thirds of the NHIN by 2010 (or, Not Your Daddy’s CHIN)

The Nationwide Health Information Network. By 2014. That’s the timeframe we’ve all heard about since GWB, Michael Leavitt, David Brailer, et al, began the ball rolling from a governmental initiative perspective about five years ago. So, half the time has passed and we’re halfway there, right?

OK, it seems Sisyphus had an easier time reaching the mountaintop with his boulder than we’re having with this NHIN rock. It has been more than 25 years since some pretty smart people saw the advantages of using electronic brains to assist with the collection, manipulation, and dissemination of healthcare information. Yes, we have seen the pioneers and the early adopters join in the uphill shove, but we are currently so far away from any sort of national healthcare data sharing that it is almost comical. (Have you ever tried to aggregate old records for a patient who has seen three doctors, two ERs, an urgent care, and a couple of health departments in the past few years?)

So, on to Step 3 of the “Official Grunt-in-the-Trenches Complete U.S. Healthcare System Overhaul and National Health Information Network in Five Easy Steps Disruptive Innovation Package.”

Is it really possible to achieve 70% of the NHIN by 2010? (Well, maybe 2011, but NHIN and 2010 have a better rhythm!)

Yes. Period.

I know, I know. Many with far bigger brains than my little peanut tell us the interoperability difficulties, the interfaces required, the rules, and the regulations are almost insurmountable with our current systems. And, maybe … maybe they are correct. With “our current systems” maybe there is no hope. But, have you ever noticed how many times throughout history that the prominent intelligentsia got it wrong? Have you noticed how many times those who discuss why things can’t be achieved are eclipsed in the historical record by those who say, “Why not?”

Well, here goes my next shot at “Why not?” and – hey, it could happen – a historical footnote. (For the first shot, see Step 2.)

We have been focused upon building these big center systems – the RHIOs, HIEs, and their ilk – and then figuring out how to connect all these silos, these disparate giant Sequoias of information. Well, first, did I miss something? Don’t we have connectivity? Doesn’t this massive information exchange system already exist as a little thing called the Internet? (Internets, according to some.) So, if you and I can exchange information – as we are right now, regardless of where you are – how is it that we need these massive health information exchange projects to send data across town? Is health information somehow different? Does the exchange of health data involve some unique form of electrons or a specialized set of zeroes and ones compared to lay data? Doesn’t seem to be a problem for first cousin Finance.

Why not…consider a different approach to both linking healthcare professionals and motivating participation in the Grand Healthcare Digitization Project, participation of both providers and the general population? Does the fact that we have already invested oodles of money in current projects preclude the thought trail which veers sharply away from the current, overly beaten track?

Why not … convince health systems and providers to understand that patient health data is not “owned” by anyone except the patient, that sharing that info is not counterproductive to corporate profits but rather contributive, once all are duly linked?

Why not … start building a system less filled with silos, less federated, and begin endorsing truly integrative designs, designs which take advantage of the connectivity we already have and utilize that connectivity to motivate usage?

What I’m saying is, there are certain systems and/or design elements already available which could integrate and inspire usage and adoption of both providers and public. We are not focusing upon them, at least in part, because of earlier (and some current) failures of integrative-type systems. (CHINs, RHIOs, etc.) Regrettably, those systems had some grossly neglectful design flaws: they were built upon financially unsustainable models and/or they did not take basic human motivations into sufficient consideration.

Here’s the skinny on attaining 60-70% of the NHIN (OK, it’s a somewhat chubby skinny):

a) Small communities and their associated small community hospitals serve an estimated 60-70% of the U.S. population. (See where this is going?)

b) Let’s utilize the power of “community,” a strong force within small towns, to help inspire adoption and conquer the learning curve necessary for this “paradigm shift.” (Sorry…I hate such corporate-speak clichés, too.) Enabled consumers and ancillary medical services providers – EMS, police, fire, sheriffs, home health, hospice, health departments, school nurses, etc. – will entice physicians to adoption by their creation of a need to which the docs can respond, this response being something of a forté for most clinicians. (Commercials with the local fire chief and Little Timmy or Grandma Gertie, who were saved because emergency responders had access to important health data at critical moments will drive digital participation far more than, “Look, see how many demographic or health history boxes we have for you to fill out.” Enabled patients will begin to ply their caregivers for participation with community-based digital care basics.)

c) The “Promised Land” of EHR/PHR use involves all of the tremendous information, subsequent insights, availability, and integration said info will provide. But, we delay our journey by focusing upon all of the wonders to come. Small bites, small sips – that is what consumers and physicians need to start along the path out of the Paper-filled Desert. Don’t blind them with the overwhelming roar of the “Voice of the ‘Almighty’ EHR.”Rather, allow all us Moseses to hear the Word and feed the people those portions which they can tolerate as we attempt to lead them out of the paper-filled desert. (Read this as: Institute small, valuable tools in modular form that can provide immediate, real life value and then build upon this foundation with additional modules providing increasing value in tolerable, stepped progressions.)

d) Using the tools mentioned in Step 2 – an end user-friendly system with simple, familiar tools and goops of support and education, along with an open, but secure, design that is inclusive (read: non-proprietary or, at the very least, one that works and plays well with others) – we establish a local non-profit governance organization to oversee and insure local concerns. (A for-profit corporation to help establish and support these local 501(c)3s would likely be more fleet-footed than a governmental oversight org.

e) Within each community, while we enable bridging the digital divide in these typically less technologically advanced populations, we would create a minimum of 10 to 20 new jobs (20 X 2,000 = 40,000 new jobs) for support and education services. We would also enable small hospitals to retain employees, a current, major challenge. (Our local hospital recently had to let 50 people go: 50 X 2,000 = a possible 100,000 jobs saved.)

f) Utilizing a combination of tools, such as the Health Record Bank, so pleasantly detailed at HIMSS by Drs. Deborah Peel and Bill Yasnoff, along with the Integrated Health Record (IHR,) as discussed in Step 2, we could create a patient-centric, patient-controlled, community-driven healthcare model which, with one of the available integration/exchange engines – for instance Medicity’s Care Collaboration Platform – would allow for a truly inclusive, integrated healthcare community system. All could participate, including new and legacy systems.

g) Such community systems, based upon designs specifically for small communities, would not provide Mayo-esque, Rolls Royce-type digital healthcare magic. They would, nonetheless, provide what small communities actually need: the initial Honda Civic version that may have few bells or whistles but can get us from paper (here) to electronic (there) healthcare – or at least down the first part of the journey. Once small communities have given up their horse-and-buggy (paper full) systems, they will be able to actually share basic healthcare data seamlessly, as their systems’ similarities will allow such sharing, and we can add new modular enhancements as people become more and more comfortable with their new mode of healthcare data transportation (less paper.)

h) Small communities rarely have competing hospitals. Sharing data between such communities would not engage the “it’s my data, you can’t have it” seen within larger centers. With appropriate consent and protections, sharing across the biggest HIE pipeline (the Internet) between small communities would only boost information access, enable less costly care provision, and promote small center usage by consumers who can receive higher quality care closer to home.

To recap (i.e., the skinny skinny):

• Most folks get care outside of giant centers in small communities and their associated community hospitals (60-70% of NHIN)
• Design specifically for small communities (i.e., no ill-fitting trickle down)
• Graduated, modular development (no “drink from the fire hose”)
• Immediate, real world value to inspire participation (school & home health & nursing home & hospice telemedicine services, care reminder services, real “saved lives” via first responders now)
• Patient-centric, community-driven (LOADS of support/education, opt-in only, local governance)
• Community to community connectivity with Aspen-like interrelatedness (compared to our current siloed Sequoia-like systems)
• User-friendly, customizable, intuitive, simple, familiar tools (think cell phone apps, JAVA, Flash, video, “fun” elements, community kiosks, local education sessions by locals, door-to-door help and support)
• Utilize “natural human motivations” to inspire co-adoption and conquer learning curves (en masse, in community, we are more inspired, more driven)

I promised at least a Tweet about funding and sustainability. So…

Funding? – Obama.

Sustainability? – A permanent Obama aristocracy.

Seriously, yes, we could certainly receive a tremendous bang for a chunk of the Obama bucks. (60-70% of the NHIN using less than 1/3 of the ~$38 billion already planned – e.g., projected costs are far less than 5 million per community X 2,000 communities = <10 billion) Anybody else know of a system or design which could attain two thirds of the NHIN for such a bargain basement, fire sale price?

Sustainability must be derived from a variety, a community, of sources. Just as the community can help drive adoption, so, too, can the community model drive a diversity of ongoing income sources. Besides having EHR vendors pony up some of the development costs for products destined to enhance their ongoing revenue streams, others should participate in supporting this communal sandbox: local employers, local hospitals and physicians via savings and enhanced earnings, HRB service fees, community-based grants via the aforementioned local non-profit, insurance company rebates and support, local levies, and, of course, the traditional small town bake sales and fish fries. Yes, yes, just kidding on those last two, but the point is, with a coordinated effort and a seriously detailed plan, sustainability can be obtained using the same community-based mindset. (It is difficult to detail this adequately within the constraints of a blog…er…”News & Opinion” site. No dis intended, Mr. H!)

Finally, the runway model skinny:

• Start with the basics in small test tube environs, provide adequate growth media, let the system grow along natural growth lines, allow Aspen-like spread and interrelatedness (maybe we should call the system, Pando?) deriving scale via reproducibility, and “allow” all of the beneficiaries to feed and water it ongoing.

As I said, I know there are many really smart people out there who are going to point out the bazillion reasons why this can’t work. Personally, I’ve never found intelligence to be a substitute for initiative. Those who depend upon nay-saying and “why you can’ts” tend only to limit; I more prefer to heed the call of the “Why Nots.”

Still to come:

Step 4: Equalizing the Playing Field (“Open” is not a Four Letter Word; Systems That’d Suit)
Step 5: Verdant Health (Lush, Full, Eco-friendly, Yet More Jobs – “Green” in Every Sense)

Dr. Gregg Alexander is a grunt-in-the-trenches pediatrician and geek. His personal manifesto home page…er..blog…yeh, that’s it, his blog – and he – can be reached through http://madisonpediatric.com or doc@madisonpediatric.com.

News 4/16/09

From AF: “Re: In-office medication dispensing. A-S Medication Solutions, a division of Allscripts, is trying to sell doctors in-office medication dispensing so they can make money on it like a pharmacy would. Has anyone had success with this?” I have to say I personally knew nothing about this subject, but have now completed “In-office Medication Dispensing 101”, with the help of the A-S website. A-S claims the average profit is $3.75 to $5.25 per prescription, depending on the medication and insurance. The Stark regulations say this is all legal as long as the patients belong to the physician. And, the physician can direct staff to dispense medications on his/her behalf. Readers will have to tell us if it is worth it. By the way, Allscripts is actually selling off this division, though they will continue to sell the solution through a co-marketing agreement.

SRSsoft announces that Valley Oak Orthopaedics (CA) de-installed a CCHIT-certified EMR and replaced it with the non-CCHIT-certified SRS hybrid EMR. SRS is clearly making a statement that CCHIT certification alone does not guarantee the product will be efficient solution. The administrator of the three-doctor group is quoted as saying,”We chose the SRS hybrid after the existing traditional EMR in our practice drained our productivity and became unusable.”

A local paper examines the move to electronic health records across the Sarasota, FL community. Included is the story of a two-doctor pediatric and internal medicine group that moved to EHR nine years ago. Despite the $70,000 per physician up-front cost, the physicians believe the system has improved their record keeping, facilitates the transfer of records to other physicians and patients, and provides excellent tools for proactive and preventative care. We continually hear stories about how EHRs are not yet doctor-friendly, so it’s nice to hear a few success stories now and then.

We’d like to welcome Eclipsys, our newest HIStalk Practice Platinum sponsor. Eclipsys has been providing information solutions to healthcare systems and hospitals for many years and just last year purchased Medinotes (which previously acquired the highly regarded Bond CLINICIAN software, now called “PeakPractice”). Eclipsys also offers their physician offices the Sunrise Ambulatory solution. We appreciate their support of our new site!

Tell me if I am wrong to be mildly offended by the title of this article recently published in American Medical News newsletter (produced by the AMA): “How to handle patients who are always late?” I am an on-time person and I hate tardiness, too. I want to know where the article is entitled, “How to handle doctors who are always late?” Maybe it is just my doctors, but I always end up waiting at least 15 minutes any time I have an appointment. While I appreciate that each patient gets personalized attention, I don’t appreciate that unstated message that my time less important than the doctor’s. There. It’s off my chest.

Set your DVR: Jonathan Bush, athenahealth’s chairman and CEO, will preside over the NASDAQ closing bell Thursday April 16th at 3:45 pm ET.

MGMA finds that compensation for medical directorships in non-hospital-owned groups is an average of 69% higher than in hospital-owned groups for all specialties except primary care. Across all specialties, the recruitment and physician education responsibilities yielded the highest compensation. Physicians accepting medical director roles were able to increase their compensation between 80 and 100% or more, depending on specialty.

I got my taxes done at the last minute. Sure hope the government spends my money wisely.

E-mail Inga.

Mark Anderson 4/15/09

My HIMSS Thoughts

Traditional buzz around vendors’ products with not much substance behind their claims. Most vendors were talking about how great sales were going and that the HITECH portion of the ARRA bill was going to really help explode sales. However, when I asked them what was in the HITECH bill, they had no idea of the details.

So the big buzz was HITECH, but …

Many intellectuals and government officials have been convinced that technology cost is the major factor for slow adoption of EHR technology. As we read in the August 2008 article in the New England Journal of Medicine, only 4% of physicians are fully utilizing EHRs in their practices today, with an additional 13% using parts of an EHR. In the hospital setting, HIMSS Analytics estimates that less than 2% of hospitals are using an EMR based on the seven levels of hospital technology adoption.

In reality, cost is a factor, but maybe a minor factor. With over 400 vendors in the marketplace, physicians have numerous opportunities to adoption EHR applications that cost less than $1,000 per year. This equates to less than 0.00033% of a physician’s annual gross income. We believe the real barrier to adoption has been twofold:

1. Physician data entry time increases by 7X over the paper based system, and
2. Physicians are not paid for data entry time.

Therefore, if we cannot decrease the physician’s data entry time, EHR adoption will never take off.

But wait — that’s where HITECH saves the day. The HITECH Act requires data sharing and interoperability between all care providers, thus potentially reducing physician data entry time by up to 75%. Finally, someone in the government figured out that the value of the EHR is in the data sharing between the primary care physician and the specialist, and between the specialist and the hospital, and even more importantly, between the patient and their care providers.

We predict that actual sales will not begin until after the economy turns around, and maybe not until January 2010 when "meaningful use" is is clearly defined.

When the bill was first announced, many organizations were excited to hear that the government was going to help fund EHR adoption. At first glance, most healthcare providers believed they were going to receive funding to purchase an EHR. They were wrong. Physicians who have already adopted EHRs were excited that they were going to receive funding to help reimburse them for their EHR. They were wrong. 

Funding is going to providers who meet “meaningful use” criteria, can report quality indicators to the government, and most important, can exchange patient-specific clinical data with other providers in the community.  Funding will not go to providers that have pre-existing EHRs unless they are connecting to a community HIE. One of the government’s primary goals is to eliminate the silos of patient information within an individual provider organization.

Therefore, the vast majority of the funds within the HITECH Act are assigned to payments that will reward physicians and hospitals for effectively using a robust, connected EHR system.

Mark Anderson is CEO and healthcare IT futurist with AC Group, Inc.