Wednesday, December 31, 2008

Don't Forget the Business Case

John Halamka, the omnipresent CIO for half the health care organizations in Massachusetts, has been every engaged in the problems of sharing data among institutions for some time. In his end-of-the-year summary blog, he distills his 2008 RHIO experience (he's the CEO of MA-SHARE, the Regional Health Information Organization) down to this lesson learned:
Health information exchange for technology's sake is not sustainable. There must be a business value case.
So, as we build RHIOs and shared data systems of all kinds, let's look for solid data that there is a reward to whomever is paying the freight. Likewise, let's make sure that whomever gets the rewards pays to develop and support the system.

Thanks John, and Happy New Year!

Tuesday, December 30, 2008

The Pitfalls of Computerized Order Entry

CPOE (Computerized Provider Order Entry) is one of the center-pieces of the move to get health care into the digital age. Lots of folks are putting lots of time, money and effort into installing CPOE in hospitals, clinics, nursing homes and everywhere else doctors, nurses and patients interact. There are great expectations for big improvements in quality, costs and work flow.

Years ago, I managed a General Internal Medicine clinic that installed a new Electronic Medical Record and we had some pretty bad outcomes. The amount of work that doctors had to do to get all the data into the EMR was unreasonable. Unfortunately, a few providers tried anyway. They worked long hours, got exhausted and frustrated, and in once case, actually wound up in the hospital as a patient due to the strain of trying to keep up with a badly designed system.

I was hoping that more recent EMRs are more thoughtfully designed, but a recent study by Emily M Campbell, Kenneth P Guappone, Dean F Sittig, Richard H Dykstra and Joan S Ash in The Journal of General Internal Medicine paints a pretty dark picture. They reviewed the experience of five different hospitals that installed CPOE systems and collected a long list of the ways that the systems made things worse, not better, for the users and their patients. Problems included:
  • Inadequate desk space to accommodate charts near the computers
  • Long sequences of mouse clicks needed to do simple tasks
  • Order sets that force the users to change the way they work
  • Problems integrating clinical data and orders
  • Systems that force the nurses to give medications at specified times rather than the best time for the patient
  • Systems that force users to re-enter the orders when the patient is moved to another unit, even when this is not clinically necessary
  • Multiple providers entering conflicting orders simoultaneously
  • Decision support alerts that don't know enough about the clinical situation to give good advice, but can't be avoided
and many others.

All of these problems have been recognized for a long time. It is disturbing that they continue to appear in the most recent implementations of the systems. At least thoughtful observers like Campbell et al. are providing advice on how to do it better. Can our vendors and institutions learn from them?

Saturday, December 27, 2008

Decision Support at Marshfield Clinic

The New York Times published a glowing review today (available here) of the Marshfield Clinic's use of its Electronic Medical Record (EMR) and registry to improve chronic disease care. It also discusses the integration of the EMR into clinical research and other health functions.

You don't actually need a full-service EMR to do all the things they describe. Fletcher Allen Health Care in Burlington, Vermont has been slowly moving towards electronic systems for some time. The Primary Care providers in their eight clinics get reminders to do vaccinations, screening and other health maintenance maneuvers using a home made system that culls the information from the billing data base.

Diabetes care at Fletcher Allen is supported by the Vermedx® Diabetes Information System, which works with or without an EMR.

So, you can achieve much of what Marshfield has without such extensive infrastructure. You don't really need a full-service EMR - but it sure helps!

Friday, December 26, 2008

Decision Suuport and Drug Choice

A recent article in the Archives of Internal Medicine by Michael A. Fischer, Christine Vogeli, Margaret Stedman, Timothy Ferris, M. Alan Brookhart and Joel S. Weissman (all at either the Brigham or Mass General Hospitals in Boston) looked at the effect of simple clinical decision support on prescribing habits. The intervention was simply providing the prescriber with information about the patient's medication insurance coverage at the time of prescribing. After the formulary decision support was turned on, low cost prescriptions (generics or "Tier 1" drugs) went up 3.3%. This resulted in an estimated savings of $8.45 per patient per year! The remarkable and optimistic finding is that they saw these savings with only 12% of the prescriptions getting the decision support.

Caveats - this was not a randomized trial, so it is hard to be sure that the savings were not due to some other factor. Also, the costs of the decision support system may not have been included in the savings calculation.

Nonetheless, this report suggests that even systems that are not very sophisticated (this one doesn't seem to have any clinical acumen - it just looks up the formulary) can have a beneficial effect.

Monday, December 15, 2008

Joint Commission Issues Safety Alert on Health Information Technology

Last week, The Joint Commission, the major regulator and standards-setter for hospitals and other health care providers, issued a "Sentinel Event Alert" on Safely implementing health information and converging technologies. Although the Commission has issued technology alerts before, this one is remarkable in how broad it is, covering the entire domain of HIT and the devices that are attached to the HIT system, including Clinical Decision Support systems.

In it, they noted that 25% of the medication error reports submitted to the US Pharmacopeia involved some kind of computer technology. (It is not clear if this is higher than expected given the prevalence of computerized medication systems in hospitals. Nonetheless, it is a big number.) They also provided a high-level list of 13 suggested actions for implementation and use of HIT that represent a kind of best practices. This list ought to be brought to the attention of everyone involved in these systems - especially the institutional leadership who need to find the funds for these 13 safety steps.

The good news is that Clinical Decision Support was not singled out as a particularly dangerous aspect of HIT (pharmacy took the brunt of the data-driven bad news). Nonetheless, all of us developing Decision Support Systems should keep these issues top-of-mind.

Thursday, December 11, 2008

The Right Decisions at the Right Time Saves Lives

(Thanks to guest writer Sarah Scrafford for this post. - Ben)

Medicine and medical treatments have improved and advanced by leaps and bounds over the last 50 years or so, and we have cures for almost all diseases known to man. Of course, newer ones are emerging from the woodwork and researchers and scientists are working tirelessly to mitigate the suffering and extend the lives of people with terminal diseases like cancer and AIDS. But no matter how many cures are discovered, no matter how innovative medical technology becomes, there’s always one aspect that leaves the result of any test or treatment in doubt – the decision of the patient.

There was a time when doctors didn’t involve patients in the decision-making process at all – they just informed them that they needed so-and-so procedure done and that it would cost such-and-such amount. Only the most educated and mature of patients and/or their family asked questions and sought second opinions before going with what their doctor thought. Today though, the scenario is changing; with medical mistakes being made and malpractice suits more in vogue, patients are taking a more active role in their treatment.

Family doctors help a great deal in the decision-making process; when people need surgeries or further treatment for a disease or illness, they are the ones who tell them of the options open to them and ask them to see the concerned specialists. Decision-making involves a series of questions that the patient and his/her family must ask themselves and their doctors. If it involves a surgical procedure or further treatment, the questions normally asked are:

· Should it be done?

· When must I do it?

· Do I stand to lose if I choose not to do it immediately?

· What are my options if I choose not to do it?

· How much will it cost me?

· Is it covered by my insurance policy?

· Who is the best doctor (or which is the best hospital) to handle this procedure?

· How soon can I expect to recover?

· How soon can I expect to get back to work?

· Do I have to undergo physiotherapy after the procedure?

· What are the dos and don’ts to be followed before and after the procedure?

The questions may vary depending on the particular patient and the procedure involved. Some decisions are more difficult than others – like the legal ending of a life in the case of terminal diseases. Washington and Oregon are the only states that allow physician assisted suicide in the USA, and when you’re a terminally ill patient and have the option to take your own life rather than face the suffering ahead which will eventually lead to death, it’s not that easy to choose. On the one hand, there’s the fact that you want to end it all; on the other, you really don’t want to die.

Decisions are difficult, more so when they’re related to your body and may be the difference between life and death, health and illness. So take them wisely, without letting emotions rule you.

(This article is contributed by Sarah Scrafford, who regularly writes on the topic of Radiology Technician Schools. She invites your questions, comments and freelancing job inquiries at her email address: Thanks, Sarah!)

Wednesday, November 26, 2008

IT Heavyweights Place Big Bets on Medical Informatics

Both Google and Microsoft are investing heavily in PHRs. The companies had senior-level managers at the Health Care IT Summit in San Diego, Nov. 2008. The event was engineered by The Everything Channel.

Grad Conn is Microsoft’s Senior Director of Global Consumer Health Strategy in the Health Solutions Group.

Alfred Spector’s Google title is Vice President of Research and Special Initiatives. PHRs are one of those special initiatives, apparently.

The two IT giants approach PHRs differently.

Microsoft launched HealthVault in October 2007. It’s a PHR platform designed to store and access personal health and fitness information. It combines single HealthVault records—the storage piece (with access provided by Windows Live ID) —with Live Search Health, a search engine that allows HealthVault users to search for health and fitness information. HealthVault accomodates both the CCR and CCD standards, the latter of which was approved in 2007, and whose proponents say essentially harmonize with HL7s CDAs with CCRs.

Google Health is a PHR in which users voluntarily enter their health records and create one, centralized profile. Even though Google maintains it will not sell advertisin in Google Health, privacy concerns have dogged the concept since it was unveiled in May this year. Google is partnered with Walgreens, CVS, Quest Diagnostics (also a Vermedx® partner), Beth Israel Deaconess Medical Center, and other entities.

Its application programming interface is based on the CCR—an XML-based standard for health records developed to allow physicians to create and transfer electronic health records with ease. “We have no intention of [involving advertising] at this juncture,” said Google’s Spector. (The emphasis is mine.)

Spector explained that one of Google’s key motivations is “. . . the epidemiological research value in the data. It must somehow be valuable over the long term, where we knew drug ‘x’ affected person ‘y’ because of a certain set of factors, and that [a person] could check for not just correlation but cause and effect. Down the road I think that’d be an incredible contribution to the world.”

In contrast, HealthVault is a software and services platform. Conn explained that "like PayPal, we’re an intermediary." Microsoft’s Ballmer has been quoted as saying that health care is one of the top six bets the company is making.

The personal nature of health records, the understanding required to interpret them correctly, and the obvious need for security and privacy have been highhurdles for designers of health data systems. But it appears the design phase is nearly complete, and roll-out is beginning.

The Financial Crisis

Well, this is quite a bit off-topic, but I can't resist posting a link to this lovely slideshow by Chris Hulls on The Financial Crisis: an Historical Perspective. In just 32 beautifully crafted images, he provides a penetrating analysis of how the world's finances fell apart so suddenly. Highly recommended.

Wednesday, November 19, 2008

Medicare and Personal Health Records

Personal Health Records (PHRs), patient-managed repositories or health information, got a big boost this month when the feds announced that they will share Medicare claims data with 4 vendors. Patients using Google Health, HealthTrio,, and PassportM will be able to link their on-line accounts to their Medicare claims data showing many of the health services they use (doctor visits, hospitals stays, outpatient tests, etc.).

This is a big boost to the Health 2.0 concept and represents an endorsement of the idea that patient empowerment is good.

Tuesday, November 11, 2008

Writing letters to patients

Writing letters to patients is a key part of my clinical practice. Henry Tufo, the mastermind behind the innovative Given Comprehensive Care Clinics at the University of Vermont, made sharing medical records with patients a standard practice years ago. Activating the patient for self-management is the goal and my patients still love getting a copy of their clinical notes, including any recommendations we discussed. What happens when we provide more information to the patient?

An article in this month's American Journal of Managed Care expands our knowledge of the effect of clinical decision support when it is extended to patients as well as providers. A team from ActiveHealth Management in New York City (part of the insurance giant Aetna), evaluated a system which sends decision support messages based on lab and billing data. The alerts to the physicians are followed 10 days later by a letter in lay language to the patient. Using an observational (non-randomized) design and claims data, they evaluated how often the alerts generated the intended action (changing a medication, ordering a test, etc.).

Overall compliance with the alerts was only middling (~30%), but the rates went up when they started sending the letters to patients. The absolute gain of 3% was a statistically significant 12.5% relative gain when adjusted for age, sex, and the mix of alertable issues. In other words, they needed to mail about 33 letters to patients to generate one additional clinical action.

How could the system be improved? The article didn't say how good the data were that the alerts were based on. Billing data are notorious slow to arrive and full of errors. Basing alerts on near real-time information from laboratories and clinical records would help.

Activating the patient sooner rather than later might improve things, too. The Vermedx Diabetes Information System mails the patient letter the same day that it sends out the alert to the provider. This gives the provider only a few days of "grace" and encourages them to develop fast and effective office systems (so they get fewer alerts in the first place and deal with them before the patient calls).

Although the authors did not present any financial analysis, my back-of-the-envelope calculations are promising: If a letter costs $1.50 to generate and mail, the cost of moving one patient to a better state of care would be about $50. This is likely a very good bargain for many conditions. What's missing, though, is convincing evidence that the system really improves outcomes and saves money. For that, we will need a large scale randomized trial.

Sunday, November 2, 2008

Diabetes Incidence Going Up

The CDC just released more data on the incidence of diabetes by state. The rates have almost doubled over the last 10 years and now stand at about 9 new cases per 1,000 persons (age adjusted). That means 1.3 million new diabetic patients every year!

Minnesota has the lowest rate of new cases at 5/1000; Puerto Rico topped the list at 12.8.

We're going to need much better systems of chronic care management to handle the influx of millions of new patients. Or else, we can expect millions of premature heart attacks, strokes, kidney failures, amputations, and deaths. The costs will be overwhelming unless we get these patients insured, deploy smart management techniques, and pay very close attention to them over the years.

Monday, October 20, 2008

Is the Revolution Upon Us?

Amar Gupta, the Thomas R. Brown professor of management and technology at the University of Arizona, writes in the Wall Street Journal that the revolution in IT that transformed banking, manufacturing, media and many other aspects of American business is finally about to descend upon the health care industry. Setting aside that this is the same promise I heard from my medical school graduation speaker in 1979, and that Prof. Gupta doesn't provide a time line, he does provide a nice analysis of three potential modes of health care:
In the future, there will be three often overlapping modes of delivering health-care services: services performed in person by humans, services that can be performed by people at a remote location, and services performed by computers without direct human involvement.
Services at remote locations are a no-brainer and already here: mail order pharmacy, off-shore transcription, tele-radiology, tele-medicine, tele-surgery. Computer-only services are going to be the real revolution. Automating the clinical laboratory made it one of the most reliable aspects of health care. Electrocardiograms that come with their own interpretation have long made cardiac care better everywhere. The Vermedx® Diabetes Information System currently uses IT to replace the failing human reminder and decision making systems in one small aspect of primary care.

Perhaps what's notable about these examples is not the pure processing aspects (although interpreting an electrocardiogram is pretty remarkable), but the connections to the human systems. Figuring out how to send a computer-generated message to a human doctor, nurse, or patient is not easy. It required literally years of tinkering and adjustment to get Vermedx® to work. Although our experience is valuable, there are no rules for how to do this in the next setting.

So, yes, I think the revolution is coming, but maybe not as fast as the Professor suggests.

Sunday, October 19, 2008

Twitter for health?

PF Anderson from the University of Michigan recently posted a slideshow that suggests (among a zillion other interesting ideas), that micoblogging with Twitter can be used to do personal health behavior logs. For instance, a patient with diabetes, might send a text message from their phone every time they check their blood sugar.

It's a very interesting idea. I was involved a few years back with Interactive Voice Response Telephony (IVR "Press 1 if you want accounts, press 2 if you....") for depression and asthma. Patients would enter their symptoms, medication adherence, and other pertinent events. The system would summarize the patient's course and periodically send an up date to the patient and the provider. Some patents loved it, some loathed it. John Helzer and Magdalena Naylor at The University of Vermont have had very good results using it for problem drinking and chronic pain.

Can we provide decision support to patients this way? The biggest problem with all health diaries is patient adherence. The good thing is that Twitter might make it easier for some patients to make entries. The bad thing is that incomplete data is sometimes worse than no data at all. Of course, Twitter can "nudge" a user to make an entry, so maybe that will help.

Anybody have any experience with Twitter or other such systems for communication with patients?


Saturday, October 11, 2008

AHRQ Web M&M on Electronic Health Records

Every month, the federal government's Agency for Healthcare Research and Quality (AHRQ) publishes a free on-line journal about patient safety or some other aspect of quality. Edited by Bob Wachter, MD, these articles take the form of a case report with discussion by an expert. (There is also a quiz and free CME for physicians and others.) This month, the spotlight article focuses on the safety impacts possible by linking together medication information and laboratory information across a variety of settings - primary care, Emergency Room, and inpatient hospital. The discussant, Ted Eytan, MD, MS, MPH, argues that an Electronic Health Record could have prevented the repeated medication errors described.

In particular, Dr. Eytan argues that routine discharge summaries are ineffective in getting the right information to the primary care provider at the right time. Although he advocates for full-scale cross-institutional EHR solutions, I feel there is plenty of room for specific focused solutions to the problems of communicating with us primary care physicians. As much as I would love to have a full-service EHR, we can't afford to wait! We need to use the solutions that are on hand now and integrate them into the EHR when it is finally ready.

As always, AHRQ's Web M&M is always well-written, thoughtful, and worth checking out.


Wednesday, October 8, 2008

Are EMRs Ready for Primary Care?

I thought I was the only one thinking like this, but I'm happy to see that I'm not so alone. Shahid N. Shah at the The Healthcare IT Guy found this article in HealthcareITNews:
"When you put an EMR into a primary care practice, your life is hell for the next year," said L. Gordon Moore, MD. "EMR vendors aren't really giving us what we need. We have to make a distinction between a robust EMR with decision support tools, and one that is just being marketed as a way to improve coding. And we really need to get out of the E&M coding game."
I couldn't agree more completely with this sentiment. Getting IT support for delivering care rather than just for generating bills is the critical missing link in most EMRs. Even the ones with "decision-support modules" don't really do all that can be done to make life better for physicians (as Moore emphasizes) and (even more important) patients.


Tuesday, September 16, 2008

Light Registries: All the taste -- none of the data entry

Recently Healthcare IT News published a story on the positive effects of a diabetes registry on the quality of care at Duke. They report that Duke has used DocSite, a registry product, for about a year and has improvements in the achievement of “perfect care”(in which the patient is on time or on target for every guideline). They report a "doubling" of this rate (although they don't report the rate itself, which is notoriously low - often well below 5%.)

DocSite has been around a long time in various forms and has suffered in the past from a major limitation: somebody has to type in all the data: labs, visits, vitals, smoking status, you-name-it. This limitation has caused many practices to give up on chronic disease registries. However, there are two strategies that can help: one is to use the registry as a back end to an EMR that already has all the interesting data in machine-readable form. I suspect this what Duke is doing. Essentially, the DocSite decision support tools are bolted on to the EMR.

The alternative is to develop a lighter registry that contains only those data that are easily accessible. For instance, Vermedx depends primarily on laboratory data because they are almost always electronic from the moment they are generated. As other data types become available (vaccine usage, medications, vitals, etc.), they can be added to the system with a concomitant expansion of decision support capability. Until then (which still looks like a long way off for most Americans), a registry like Vermedx can provide huge benefits without requiring a tremendous effort in data entry. It can:
  • allow population views at various levels (practice, provider, geographic, age-based, etc.)
  • drive communications to the practice about exactly what to do next
  • drive communications to the patient about what to expect and how to get it
  • stimulate improvements in the essential office micro-systems that are at the root of Primary Care quality
  • identify patients who need extra attention from specialists, case-managers or other resources
  • connect the practice to lab results from outside institutions - even if the two institutions don't share any infrastructure
These advantages, which translate into huge savings for the system as well as improved quality of life for the patient, are how, as Blackford Middleton of the Center for Information Technology Leadership puts it,
“...diabetes registries are the only form of information technology-enabled diabetes management we found to be cost-beneficial when adopted for patients with Type 2 diabetes.”
The challenge now is how to get light registries in use now while everyone is so busy installing (or planning to begin to think about installing) large, complex EMRs. We need the savings and the quality improvement now! This is a case of the perfect being the enemy of the good.


Saturday, September 13, 2008

Health Informatics Discussion Forum

You can't do it alone and the web gives us some great opportunities to get help from the pros around the world! Many thanks to Abbas Shojaee and Dr. Gil (Gunter) Pollanz from Chris Paton's Health Informatics Discussion Forum for the excellent advice and insight they provided on Vermedx. You can see the whole thread here.

If you have ideas on how to make personalized clinical decision support better, faster, cheaper, higher, louder or just plain cool, please join the discussion either here or at the Health Informatics Discussion Forum.



Saturday, September 6, 2008

More Diabetes Maps

We love graphics of all kinds and can use Vermedx® to make some pretty cool maps. Now the folks at The Center for Public Health Informatics at the University of Washington in Seattle have developed an open-source GIS tool for public health that takes up your data set and makes a Google Map out of it. They call it EpiVue and it has a web page and has been published in The International Journal of Health Geographics (Hey, I'm a subscriber!).

I loaded up the CDC Diabetes Prevalence Data and gave it a whirl. I had some trouble getting the data formatted just right. And, it wouldn't take the whole dataset at once. But, I did get a pretty nice map of Texas showing the prevalence of diabetes by county. However, I couldn't for the life of me figure out how to save it so I could post it here!

Overall, it's a pretty grand idea - make a free tool to advance public health epidemiology using off-the-shelf parts (Java, Google Maps, JFreeChart, R, etc.). However, its going to need a better user interface to make it a real winner. I hope they keep going with it!


Wednesday, September 3, 2008

CDC Assessment Initiative Conference

Dr. Littenberg organized a panel discussion at the annual Centers for Disease Control and Prevention Assessment Initiative Conference. This conference brings together public health officials from around the country to share ideas and accomplishments regarding the assessment of public health. There are a variety of tools for this purpose including the Behavioral Risk Factor Surveillance System (BRFSS), which is the world’s largest ongoing telephone health survey. It has been continuously administered in every state since 1994, and in some states as far back as 1984. States use BRFSS data to identify emerging health problems, establish and track health objectives, and develop and evaluate public health policies and programs. Many states also use BRFSS data to support health-related legislative efforts. Presenters at the Assessment Initiative Conference discussed ways to make this information more available to policymakers, researchers and the public. Other conference presenters discussed topics such as novel ways to assess hard to reach populations, the role of assessment in promoting community action, and ways to link data from various sources in order to paint a more complete picture of the health of the population.

Dr. Littenberg’s panel was entitled Using the Clinical Laboratory for Public Health Surveillance of Chronic Disease. Dr. MacLean discussed the Vermont Diabetes Information Study and described the collection of lab data from disparate sources, creating a chronic disease registry, and providing decision support. Dr. Fernando Guerra, Directory of Public Health, San Antonio Metro Health District discussed the use of the Vermedx technology to create an A1C registry for San Antonio. Dr. MacLean then discussed the ethical, legal, and social implications of chronic disease surveillance, highlighting the experience in Vermont, New York City and San Antonio. Edward W. Gregg, PhD, acting chief, Epidemiology and Statistics Branch of the Division of Diabetes Translation at the CDC presented on the CDC’s diabetes surveillance system and the potential for adding laboratory based surveillance to this system.

This was a great opportunity to think about the bigger picture of how we assess the health of our various communities, ranging from the local to the national level. What would a network of labs contributing results to a central database be able to tell about regional variation and quality of care for chronic disease such as diabetes?

Tuesday, September 2, 2008

Looking at the outliers

This graph is a little unfamiliar, but it has lots of power to show what's going on behind the data. 71 small community practices are laid out on the horizontal in order of their average LDL-cholesterol. The vertical axis show just what that level was.

The red lines at 100 and 130 mark the guideline-recomended targets. Although most diabetic patients should be below 100 mg/dl, the average patient in most practices is getting there.

Pay extra attention to the right hand end of the line. The last 4 practices have very high levels. In fact, the highest 11 practices seem to have a bit of space between them and the regular run of practices. Perhaps these are the practices that need some extra help in taking care of cholesterol.

We something similar on the left hand side: the lowest ("best") 4 practices seem to be standouts. Maybe we should visit those places and see what they've got going on - so we can bottle it for others to use!

Of course, the graph is just suggestive. There may be other, less interesting, reasons for outliers such as small sample sizes, data errors, miscalibrated lab results, etc. But, the picture does seem to tell us where to look next.

One more example of what you can do from a population view that you can't do one doctor at a time!


Thursday, August 28, 2008

The Missed Patient with Diabetes

Xuanping Zhang and colleagues from the CDC have just published an interesting paper on the prevalence of undiagnosed diabetes. Using data from NHANES (a nationally representative survey), they found that adults without insurance are more likely to have undiagnosed diabetes than those with insurance. Being uninsured for more than a year was associated with 2.5-fold increased risk of having diabetes without a dignosis.

The main message, of course, is that we need much broader (dare I say "universal?") health coverage so we can intervene early and prevent the nasty complications of diabetes. However, Charlie MacLean (my nearly silent partner) brought up an interesting issue: What about insured patients? How often do we miss them? Can we use the laboratory registry to find folks with incidental blood sugar results that warrant a diagnostic work-up?

There are (at least) a few problems: What thresholds should we use? Probably 126 mg/dl for fasting and 200 mg/dl for non-fasting samples to start. Eventually, we could probably work out the risk of diabetes at lower levels of glucose.

How many are there? Damon Lease, our colleague who has done quality management at a small community hospital guestimates that he finds a few dozen a month. A regional registry might find many, many!

Whom should we notify? A blood sugar ordered as part of a chemistry profile in, say, the ER, might get overlooked in the hurly-burly of that setting. The best person to judge its significance and orchestrate follow-up is the Primary Care Provider. However, figuring out who the PCP is from administrative data (and even an EHR) is fraught with difficulty.

So, there are some challenges, but we may be able to take a bite out of the missed diabetes problem.


The Missed Patient With Diabetes: How access to health care affects the
detection of diabetes
Xuanping Zhang, Linda S. Geiss, Yiling J. Cheng, Gloria L. Beckles, Edward
W. Gregg, and Henry S. Kahn
Diabetes Care 2008;31 1748-1753

Wednesday, August 27, 2008

Is Vermedx® "Health Information Exchange?"

First, what is "Health Information Exchange?"

“The movement of health-related data–clinical and/or administrative –according to an agreed-upon set of interoperable standards, processes and activities across independently operating organizations in a manner that protects the privacy and security of an individual’s information.” (“Consensus Conventions for the Use of Key HIT Terms” Project – ONC/HHS, 2008)

Well, let's see:
  • move health-related data (mostly clinical) - check
  • agreed-upon interoperable standards (among the various labs and practices) - check
  • independently operating organizations (dozens of labs, hundreds of practices, few of them even talking to one another, let alone operating together) - check
  • privacy and security - check
So, it is HIE, but it doesn't feel like it. Why? Because VDIS doesn't require a lot of new infrastructure or capital outlay by the participating organizations (especially the practices). It also doesn't have a big national consensus standard or a far-reaching scope that seeks to do all of the possibly conceivable things you might want to do with health information. VDIS is a small, single-purpose system designed to do one type of HIE fast, easy and inexpensively. As it gets good at one important task, we extend it and add other tasks and features. Happily, the underlying architecture is very flexible, so we'll be able to do this for lots and lots of cool new services.

But it is open for business now.Thousands of patients are getting reminders and alerts when their diabetes needs attention. Hundreds of PCPs are getting decision support and advice tailored to their patients' current clinical situation.

I'm very excited about RHIOs and the wonderful potential to make wholesale changes in broad chunks of health care. But, I'm also impatient to start making things better now. So, I'm glad that providers and insurers in my part of Vermont are using Vermedx® to start bringing the information age to us today, while the state-wide infrastructure for a fully-functional RHIO gets put together.

Tuesday, August 26, 2008

Diabetes Maps for Public Health

Here is an example of what you can do once you have the registry up and running. These data are a few years old, but they show the variation in average A1C across a large swath of New England and upstate New York.

Many of us were surprised that the best control is in a county with major social and economic challenges, while some of the worst control is in some of the wealthiest and best-served areas. One possibility is that the care is not so good near the big academic medical center, but another is that folks with worse diabetes aren't even getting measured at all!

What do you think?


Monday, August 25, 2008

Video: Automated Use of Clinical Laboratory Results in Adults

I just found out that the Jones Lecture from last May is on the web as a video. The talk covers the issues of using laboratory data in both clinical outreach to individual patients and for public health. If you want to see it, you can find it here.

Any feedback on the talk would be very welcome.



Friday, August 22, 2008

Architecture for Clinical Decision Support AND Public Health

Because of the clever way Mike Gagnon designed the Vermedx® Diabetes Information System, we got an unanticipated "twofer." In addition to supporting the robust clinical decision support activities described in this blog, it also provides a unique resource for population management beyond the practice level. The database in the middle of the whole thing can be used to provide a snapshot of the current level of care for diabetes across a wide population.

Here is the layout:

So, although we started out to get a better view of the patient's situation, we wound up with a way to see the whole population. More on this soon....



Thursday, August 21, 2008

Automated Use of Clinical Laboratory Results

I was very honored to give the Jones Lecture at Dartmouth College's Thayer School of Engineering last May. The talk covered the design and evaluation of the Vermedx® Diabetes Information System. It is now available on slideshare.

Let me know what you think of it.



Sunday, August 17, 2008

Population Reports

One of the things that's particularly difficult for a PCP to do is keep a population view of all the patients under their care. We're able to use the Vermedx® database to generate just such a view: a roster of all the provider's diabetic patients.

The first page has a summary of the panel with statistics about how many patients are included, how many are under control, and how the panel compares to the whole database and to the best practices.

Th following pages list the patients with their latest results.

A full size version is here.

Providers can use these reports to assess their performance, monitor quality improvement efforts, and identify patients who need extra attention.

We send them out to the practices every three months. One of the nicest calls we get is "Can I have a copy of my Pop Report? I want to do some quality improvement work." That really makes my day!

Monday, August 4, 2008

Patient Reminders

After we worked out the kinks in communicating with the providers, we turned our attention to the patients. One of the big problems in diabetes care is that the patients frequently don't get their tests on time. We stimulate the practices by sending Practice Reminders when the patient is overdue for tests. We also stimulate the patient by mailing them a letter at the same time. The letter is produced by our computer, but it has the name and address of the practice and is sent on behalf of the patient's personal provider. (It can even have the logo of the sponsor such as the insurer or the practice group or a state agency.) It lists all the diabetes tests and when they are due and asks the patient to call the practice to set up testing.

A full size version is here.

These letters stimulate the patient to call the office and get back in care. They inform and activate the patient while the Practice Reminders prepare and activate the practice team, just like the Chronic Care Model says we should.

Direct outreach in this way has also turned out to be a huge patient satisfier. They just love getting these reminders and we have gotten many calls and notes from patients thanking us for the service.

Thursday, July 31, 2008

Practice Reminders

The flowsheets from the Vermedx® Diabetes Information System (VDIS) give providers information when a lab test is done. However, one of the big problems in diabetes (and other chronic diseases) is that many patients don't get the tests they need on time. Part of the reason for this is that the schedule for when tests are needed is complicated. Each of the four basic tests (A1C, cholesterol, serum creatinine, and urine protein) have a different due date that depends on the date and the level of the previous result. The rules are hard to remember and doctors and nurses don't have a good way to keep track of them.

So, we programmed the VDIS to scan the database of results looking for patients who are overdue for one of the tests. When it finds one, it produces a simple report to the practice that says, basically, "Hey! This patient is getting lost! Get a hold of them and bring them back under care!"

A full size sample of the report can be seen here.

Practice Reminders arrive at the practice 30 days after the test was due. If the practice has a protocol or standing orders, the office staff can call the patient, order the lab tests, and schedule an office visit for a few days after the tests. At the visit, the provider can review the results with the patient face-to-face and together they can make a plan based on current information. This is often much better than getting the tests after the visit and writing a letter or playing phone tag.

Practice Reminders are a Value Added service that a Health Information Exchange can provide at low cost. Stay tuned for some other examples of ways to use HIE to improve care.



Wednesday, July 30, 2008


Flowsheets, the systematic recording in chronologic order of findings and facts about a particular patient, are a great way to maintain and improve the quality of care - especially for patients with chronic disease. Doctors (and other providers) love to have flow sheets - but they hate to make them. Personally, I spend a huge fraction of every day in the clinic updating the lists with medications, vital signs, lab results, etc.

One of the main stimuli for developing the Vermont Diabetes Information System was to lessen the burden of keeping the flowsheet. It started back in the early '90s when Charlie asked the lab at our hospital to make flowsheets of the key lab results for patients with diabetes. That simple idea evolved over more than 10 years and two NIH grants into an automatic system that includes the patient's test results no matter which laboratory they use, and provides guideline based advice to the primary provider to help them stay organized. Here is the current version:

(A full size pdf file is here.)

It arrives in the office the morning after the lab test is done - often before the official lab report. And, it has more than just the latest result for a single test - it has the last 4 results for all of the major diabetes tests plus advice based on the American Diabetes Association guidelines. The doctors and nurses liked these "value-added" laboratory reports and they became the basis for other HIE-based products and services which I will write about soon.



Monday, July 21, 2008

Top US Counties for Diabetes

Because so much of laboratory-based HIE comes out of the world of diabetes, these data are very interesting. They come from the CDC's county-level prevalence estimates here.