Errors in Google Health

Google Health, the Personal Health Record (PHR), has quite a nasty problem. The Boston Globe reported that at least one patient had major errors in his record. The errors were quickly traced to Google's use of billing codes to drive its diagnosis and problem lists.

The good news is that Google 'fessed up and says it has learned from the episode. The bad news is that they may not have learned very well. Neil Versel reports that Google managers now say that "...Google Health would now allow free-text diagnoses that didn’t have to correspond to a billing code."

The underlying problem is not just that billing codes may be inaccurate, but that all data may be inaccurate, out-of-date, or misleading. Free text is notoriously difficult to work with and certainly prone to its own problems and errors. The problem of automatically constructing a patient profile from warehoused data is analogous to the problem of diagnosing a patient complaint. Very rarely does any one bit of data from history, physical exam, laboratory or imaging unambiguously make the diagnosis. Conflicting possibilities need to be resolved by assembling a (sometimes large) list of data points and finding the underlying pattern that explains them.

Likewise, the role for PHRs is not just to assemble the data about a patient. Making a diagnosis from those data requires very sophisticated algorithms that take into account the inherent quality of each data point, the process that produced it, how it relates to other data, when and where it was collected, and so forth. When you are all done, even the best systems are still going to have both false-positives (listing something not really true, like what happened to Dave deBronkart on Google) and false-negatives (missing something important).

How do we handle this on the Vermedx Diabetes Information System? The system provides very actionable advice to both providers and patients, so we need it to be very, very accurate and reliable. Before we do any contact with the patient, we do something a bit like what Google and other PHRs do: we automatically scour computerized records. In our case, the data are primarily clinical laboratory results which are generally much more reliable than billing codes. Then, we check with the primary care provider and ask them two critical questions: Is this patient truly diabetic? Are they really your patient? Only when we have two "yes" answers do we even offer to enroll the patient.

Even with this level of feedback and control, we still have false positive results. So, when we do contact the patient, we ask them to tell us if the information is incorrect. Some patients have moved, switched doctors or even died without the primary care practice knowing. And, sometimes the provider told us "yes" when they should have said "no."

The point is, we don't trust just data source. Even in this seemingly simple domain of a single diagnosis with good laboratory tests that are reliably reported, we need to combine multiple data sources to achieve high quality information. PHRs need to do more than copy a list of apparent diagnoses from one database to another. They need to explain the data's provenance, strengths and weaknesses, and help the user decide whether to accept, reject, or modify the tentative conclusions reached by the system. This is not easy, but when you do it well, there are big pay-offs!

How much is an e-mail worth?

According to the Santa Cruz Sentinel, doctors in Palo Alto, California are offering a new service to their patients: e-mail with their doctors. The fee is $5 per month. The e-mail option is apparently too new to tell how patients feel about it.

I've been exchanging e-mails with my patients for over 15 years, although I've never charged for it. It is much easier than playing phone tag and cheaper for the practice than involving a secretary and a nurse in message handling. (The toughest part right now is attaching the patient's birth date or record number to the message and getting it filed in our (still paper-based) record system. A dedicated secure e-mail portal could take care of that easily.) The volumes have never been very large and I have no fear of being overwhelmed.

How does it effect income? In my experience it almost never replaces a visit. Rather, it substitutes for one or more phone calls. And, like phone calls, it sometimes ends in "You had better be seen."

I'm not tempted to charge for the service, but it does seem like a smart thing to include in a capitation fee or medical home charge. Many insurance companies encourage patients to call or e-mail a health plan nurse for general advice. It seems logical that they should support the patients contacting someone who can give them even better advice: their doctor!

Health Technology News

Thanks to Rich Elmore at Heath Technology News for his recent coverage of Vermedx® and the recent studies showing its impact.

Rich does a great job covering both the business and health sides of HIT and his column is a must-read, always.

Cost savings from Vermedx® decision support

The American Journal of Managed Care has their web site back up, so here is the link to our new article:

The Vermedx® Diabetes Information System Reduces Healthcare Utilization
Benjamin Littenberg, MD; Charles D. MacLean, MDCM; Karl Zygarowski, BS; Barbara H. Drapola, RN; James A. Duncan, MD; and Clifford R. Frank, MHSA
Am J Manag Care. 2009;15(3):166-170
Published Online: March 16, 2009 - 12:00:07 AM (CDT)

It shows that the savings estimated in the NIH clinical trial of Vermedx® are confirmed in an analysis of managed care claims paid. I posted the abstract and a key figure Sunday. Here is the table showing the savings that are generated when patients are enrolled in the Vermedx® Diabetes Information System.

Table 2: Net savings per patient as a function of duration of the VDIS program

Duration (months)	Monthly savings	Annual Savings	Cumulative savings
12	$80.96	$504.24	$504.24
24	$165.92	$1,523.76	$2,028.00
36	$250.88	$2,543.28	$4,571.28
48	$335.84	$3,562.80	$8,134.08

Savings are calculated net of the costs of the program.

For more information, please see www.Vermedx.com.

The Vermedx® Diabetes Information System Reduces Healthcare Utilization

The American Journal of Managed Care published our article last week. Unfortunately, their web site got hacked and the PDF is not yet available. However, here is the abstract and main figure to tide you over till the whole opus is downloadable.

The Vermedx® Diabetes Information System Reduces Healthcare Utilization
Benjamin Littenberg, MD; Charles D. MacLean, MDCM; Karl Zygarowski, BS; Barbara H. Drapola, RN, CCM, CPHQ; James A. Duncan, MD; and Clifford R. Frank, MHSA
Am J Manag Care March 2009;15(3)
Objective: To confirm the cost savings in a randomized clinical trial of the Vermedx Diabetes Information System (hereafter referred to as the Diabetes Information System [DIS]) in independently collected data using claims paid by a managed care insurer for patients with and without DIS participation.
Study Design: Longitudinal analysis of paid claims with concurrent and historical controls from October 2002 through October 2007.
Methods: Using locally weighted smoothing functions and linear regression analysis before and after commencement of the DIS, we compared the total claims paid per member per month for 153 patients using the DIS versus 870 control patients.
Results: For DIS patients, paid claims increased at a rate of $8.30 (95% confidence interval [CI], $1.12-$15.48) per month before the DIS started compared with −$3.92 (95% CI, −$9.50 to $1.67) after commencement of the DIS (P = .008). For control patients, the slope changed from $6.80 (95% CI, $3.78-$9.82) to $3.16 (95% CI, −$1.06 to $7.38) (P = .17). After commencement of the DIS, the slope of the claims in the DIS group is significantly lower than that of the control group (−$3.92 vs $3.16, P = .046). The mean estimated savings range from $504 per patient in year 1 of operations to $3563 in year 4. The cumulative net savings reach $8134 in 4 years.
Conclusions: Participation in the DIS is associated with substantial reductions in claims paid, net of the costs of the intervention. The cost savings reported in the randomized clinical trial of the DIS are reproduced in an independent data set.

Figure 1: Claims paid per member per month estimated by non-parametric locally weighted smoothing. The vertical line represents the start date for VDIS patients and a randomly chosen date for control patients.

IBM's "Google Earth for the Body"

IBM has a new interface idea for the medical record: a humanoid avatar. It's a 3D image of a human body, with the patient's data linked to the various body parts. Users click on the heart to get cardiac reports, the kidney to see renal information, etc. They tried it out in a Danish hospital and report good results.

On the face of it (actually, the pictures I saw had no face), its a really cool technology, bit its hard to see what problem its solving. Do doctors and nurses need that level of help organizing information? In my experience, anatomic thinking is not where we fall down. How can this technology help us see the systems and connections among the organs? The out-of-body factors (environment and interpersonal relationships) and microscopic forces (genes and proteins) that drive so much of health don't have an obvious place in this model.

One potential upside: it could be a great way to educate patients about their health.

What do you think about using this technology to improve care?