Lab Soft News

More on Creating Customized Lab Decision Support Rules

In a recent blog note (see: Static Versus Dynamic Lab Test Selection Algorithms), I mentioned a product that could be used to create customized dynamic decision-support rules. Dr. Mammen, whose original comment had promoted this blog note, submitted the following additional comment:

The tool I have been using is called the mTuitive xPert (mTuitive, MA.). It is designed to be used by non-programmer/domain knowledge content manager (as I see myself). One can acquire content from the company or build it in to suit local practices. It allows building of rules, using logic to control list items and also accept formulae to determine staging/grading. We are also exploring the potential for other domains - non-cancer related.

I was negligent in not having mentioned the mTuitive xPert product and providing a link to the company web site, an error which I have now corrected.

Health Insurance Companies to Reimburse for "On-Line" Physician Services

It now appears that some health insurance companies have decided to reimburse physicians for the delivery of online professional services, which is to say for virtual office visits (see: Online house calls click with doctors). Below is an excerpt from the article:

Since the dawn of e-mail, patients have been pleading for more doctors to offer medical advice online....There was always one major roadblock: Most health insurers wouldn't pay for it....In recent weeks, Aetna Inc., the nation's largest insurer, and Cigna Corp. have agreed to reimburse doctors for online visits. Other large insurers are expected to follow, experts say. These new online services, which typically cost the same as a regular office visit, are aimed primarily at those who already have a doctor. The virtual visits are considered best for follow-up consultations and treatment for minor ailments such as colds and sore throats. But some specialists, including cardiologists and gynecologists, also see these e-mail tete-a-tetes as ideal for periodic checkups that don't require in-person visits....As more doctors move online, others are looking further ahead and adding webcams to their online arsenal, even if the video quality remains spotty.

I have blogged previously about a service called TelaDoc (see: Crafting an "Improved" Medical Practice Model) whereby a rotating panel of physicians can be contacted by telephone by subscribers of the service, frequently for emergency medical advice. Obviously, a virtual visit with a physician with whom one has a continuing relationship provides a higher level of service. On-line interactions can also be improved by the use of a webcam but this requires the deployment of more technology. I have also previously commented about how some physicians are using video emails to discuss lab test results with patients (see: Physician Uses Video Emails to Discuss Lab Results with Patients). It thus appears that the style and substance of e-interactions between physicians and patients are now undergoing major changes, stimulated in part by changes in reimbursement.

I suspect, given this new reimbursement opportunity, that the definition and parameters of an on-line physician visit will require greater clarity.  By this I mean that some common ground will be reached whereby both physicians, patients, and payers arrive at an understanding of what is meant by a virtual physician interaction, the processes and technology involved, and the expected outcomes.

An article that I encountered on this topic (see: e-Visits: The Tipping Point - Are We There Yet?) makes a number of interesting points, but the one that I found to be the most interesting is the following:

In 2004 the AMA established the CPT code 0074T for online medical evaluations between doctors and current patients. Insurers that recognize the code, and reimburse for associated claims require that the e-Visit produce a documentable action such as a medication change, ordering of a diagnostic test or a referral to a specialist.

It seems reasonable that payers will require that an e-visit will result in a documentable action such as the ordering/interpretation of a lab test, a new or modified drug prescription, or a referral to a specialist. For this reason, greater attention to e-visits cannot help but stimulate interest in office EMRs, particularly those with the integrated management of lab test orders/results and e-prescribing features. For additional comments about this same topic, you may want to refer to this blog note: The Painfully Slow Move to Online Services.

Static Versus Dynamic Lab Test Selection Algorithms

In a recent blog note (see: The Right Test for the Right Patient at the Right Time), I discussed lab test selection algorithms and their potential integration into LISs in order to provide informed guidance to test-ordering clinicians. I fully recognize that half the challenge of deploying such algorithms is first getting clinicians accustomed to using them without being too directive and also making them readily available during the test ordering process.

A reader of this blog, Dr. Joy Mammen, has submitted the following comment that draws the distinction between what I will call static algorithms, such as those available at ARUP Consult, and dynamic algorithms:

A logical extension would be a decision support tool that allows you to put in the parameters that you know or have results for and:

1. Have the pathways ahead listed...: Suggest test X, Y or Z.
2. Ask for ancillary test results and factor them into the decision tree: What is the serum LDH level?
3. Remind the operator of other clinical entities that impact on this decision: What is the patient's blood group?
4. Display relevant references simultaneously.

Joy appropriately takes the idea of lab test selection algorithms in another important and related direction, which is to deploy more flexible decision-support tools that can provide advice to clinicians for broader or more complex clinical scenarios by first inputting a known abnormal test result. There is one product that I know about (I am sure that there must be others) that enables the development of such dynamic algorithms, They can be created on-the-fly and de novo by lab medicine experts. It's produced by an Australian company, Pacific Knowledge Systems (PKS), and called LabWizard.

There are obvious differences between the algorithms provided by ARUP Consult and those that can be developed with the LabWizard. In the first instance, the clinician has a specific disease in mind that he or she is pursuing and the test-ordering pathway can be clearly mapped out graphically by the algorithm. In the second instance, the starting point may be perhaps a common test abnormality that has already been measured such as hypercalcemia or hyperkalemia and which leads to a broad differential diagnosis. The decision-support tool guides the clinician through a complex decision tree to [hopefully] the correct diagnosis. Both approaches can be useful and should be made available from the test-ordering screen of the LIS.

"Cancer Coaches" Start to Play Significant Role in Cancer Care

Cancer patients need all of the help that they can get, particularly from empathetic and knowledgeable guides who are familiar with the ups and downs of the treatment process. But the kind of "help" that I am discussing here is not necessarily a shoulder to cry on -- it's assistance in making informed decisions when and where they count most. I did not have a suitable name for this type of person until now -- I have discovered that they are now referred to as cancer coaches or patient navigators and below is an excerpt from an article (see: Cancer 'coaches' sway treatment choices) that describes the role in detail (boldface emphasis mine):

Bad advice, or just too much of it, can compound the trauma and damage done by the disease itself, cancer patients often find. Friends and relatives are important for support, but when these untrained people act as cancer coaches, they can sway people to make poor decisions about their care....Many advocacy groups and hospitals are using "professional" coaches — trained volunteers or paid workers who can objectively help new patients navigate the maze of information and options. The American Cancer Society started a patient navigator program a few years ago that now operates in 87 locations and is planning to expand.....

Whether amateur or professional, a good cancer coach should offer these things, experts say:

• Support: an ear to listen, a shoulder to cry on, a hand to hold.
• Resources: reliable information or help getting it, and only if the patient wants it.
• Objectivity: a willingness to help patients discover what is best for them, rather than to validate the coach's own cancer battle and choices.

Click here to view a video on YouTube produced by the American Cancer Society about their cancer navigator program.

One of the most important roles for a cancer coach is to provide a second set of ears during important physician and nurse visits relating to treatment. Let's be realistic here. The patient may be lucky to hear and retain half of what is being told to him or her during such visits for obvious reasons. The coach always needs to have a pad and pencil handy to take detailed notes including the exact names and dose schedules for all drugs that are being prescribed.

In my opinion, an ideal cancer coach will also be knowledgeable about web searches in order to acquire information about such matters as adverse drug effects, disease symptoms, and the availability of clinical trials. Clinicaltrials.gov is a good place to start to explore open drug trials. Although one's instincts would to be to turn to a relative to act as a cancer coach, my instincts tell me that friends or even trained previously unknown volunteers might serve  better in this capacity because they could bring more objectivity to the process.

Encouraging Results from LabCorp/Yale Multiplex Lab Test for Ovarian Cancer

We are starting to see some very encouraging results from clinical trials for multiplex biomarker testing for various types of cancer. A recent article discussed preliminary results from a clinical trial of a novel approach to the diagnosis of early stage ovarian cancer (Phase II Results For Yale University's Early Stage Ovarian Cancer Blood Test 99% Accurate). Also of interest is that this project is a collaboration between a large commercial reference lab, LabCorp, and a major research university, Yale. Below is an excerpt from the article (boldface emphasis mine):

Laboratory Corporation of America ... announced that Yale University has released the results of its Phase II study on a blood test to detect early stage ovarian cancer. Using six biomarkers, this blood test can discriminate between disease-free women and ovarian cancer patients (stage I-IV) with high specificity (99.4%) and sensitivity (95.3%). A Phase III trial is under way and should be completed in a few months. Initial studies are focused on women who have increased risk for the development of ovarian cancer. Ovarian cancer is the fifth leading cause of cancer-related deaths in women in the United States and is the leading cause of gynecologic cancer deaths. Despite being one-tenth as common as breast cancer, it is three times more lethal and the average woman carries a 1:70 lifetime risk of being diagnosed with ovarian cancer. This year, approximately 20,180 women will be diagnosed with ovarian cancer, and 15,310 will die from the disease.

Recall that test sensitivity is the ability of that test to detect a true positive test and specificity is the ability to detect a true negative (see: Defining Test Sensitivity and Specificity for the General Public). The test described above for ovarian cancer reportedly has very high specificity (i.e., 99.4%) which means that there are very few false positives -- normals who incorrectly have a positive test. This is ideal for a test used for large-scale screening. The sensitivity is also relatively high (i.e, 95.3%), which means that there are relatively few false negatives -- patients with the disease who do not test positively. You can read more about sensitivity and specificity at Lab Tests Online (see: How Reliable is Laboratory Testing?).

The lethality of ovarian cancer as compared to breast cancer is related, in part, to its late diagnosis in many cases. Early stage ovarian cancers are only rarely symptomatic. An accurate screening test for ovarian cancer, particularly for early stages of the disease, will thus be very welcome. For more information about biomarkers for the diagnosis of ovarian cancer, you may want to link to a PowerPoint presentation by Anna Lokshin, Ph.D., University of Pittsburgh Cancer Institute, entitled Serum biomarkers for early detection of ovarian cancer. It's located on the BioPhysical web site.

A few final words here about LabCorp's strategy of developing "captive" multiplex tests for cancer diagnosis and surveillance. On the one hand, I applaud their investment in clinical trials such as the one described that demonstrate the efficacy of panels of multiple biomarkers for the  diagnosis of ovarian cancer. On the other hand, I'm not sure that their investment will necessarily provide long-term benefits for the company nor a significant edge on its competitors. New biomarkers are being identified almost on a daily basis. Therefore, new combinations of them will continuously emerge that may be superior to the previous ones. Clinical trials to prove out new multiplex lab tests will also become less complex and expensive in this era of larger biorepositories. 

The Trend Toward New Subspecialties in Medical Training

The trend toward new subspecialties in medical training, often at the fellowship level, seems to be inexorable. In recent years, to mention a few, we have seen the emergence of interventional radiology, feto-maternal medicine, emergency medicine, and sports medicine. There are even ultrasound fellowships within emergency medicine (see: Additional Discussion About "Radiology Without Walls"). At the recent Molecular Summit conference that was held in Philadelphia, Dr. Aaron Baggish of the Division of Cardiology, Massachusetts General Hospital, delivered a lecture entitled Cardiovascular Medicine 2008. In his slide #7, he lists the following six categories of subspecialties in cardiovascular medicine in addition to the generalists:

• Electricians
• Imagers
• Basic scientists
• Interveners
• Heart failure specialists
• Miscellaneous (e.g., aortic disease, congenital, sports, genetics)

A cardiologist friend has suggested yet another emerging category to add to this list: preventive cardiology. In addition, the miscellaneous category above can be viewed as an incubation tank for new and emerging subspecialties in the field. Note the presence of cardiovascular sports medicine, presumably as a new branch of sports medicine. The aortic disease category category above undoubtedly encompasses the new treatment of descending thoracic aortic aneurysms with endovascular stents.

Given this background information, one might wonder why I would have the temerity, or simply bad judgement, to suggest in a series of previous blog notes that pathology, lab medicine, and pathology should consider merging. The trend toward new specialties seems to be moving in the opposite direction -- the splitting of major specialty groups into ever more specialized categories. The answer to this question is that most of the subspecialization activity cited above involves medical disciplines relating to therapeutic intervention. I believe that a merger of pathology, lab medicine, and pathology would cause a coalescence of the diagnostic specialties into a single top-level specialty -- diagnostic medicine with support from a new diagnostic information system (DIS) which would consist of an integrated LIS, RIS, and PACS. Such a merger then puts at the disposal of its practitioners all of the various diagnostic imaging modalities, surgical pathology techniques, and molecular diagnostic tests that are available.

If and when such a merger were to take place, it would be impossible for any single individual to be knowledgeable about the diagnosis of all diseases across all organ systems. I therefore believe that the specialty of diagnostic medicine would immediately cleave into a set of diagnostic subspecialities related to individual organs, organ systems, or general disease categories such as cancer or autoimmune disease.
 

Seeking a More Accurate Lab Diagnosis for Non-Alcoholic Fatty Liver Disease

A recent article about non-alcoholic fatty liver disease (NAFLD)/non-alcoholic steatohepatitis (NASH) caught my attention (see: Geisinger And Celera To Collaborate On Liver Disease Test) for several reasons. Here's an excerpt from it with boldface emphasis mine:

Geisinger Health System and Celera...[have] announced that they have entered into a research collaboration with the aim of developing a diagnostic assay for the increased risk of non-alcoholic steatohepatitis (NASH). The collaboration will evaluate Celera's numerous genetic findings in liver diseases...in the Geisinger bank of more than 600 liver tissue and blood case-control samples donated from patients who have undergone bariatric weight loss surgery....Non-alcoholic fatty liver disease (NAFLD) is a condition where there is accumulation of excess fat in the liver. NAFLD is usually asymptomatic and, as a result, many people don't know they have the condition until substantial liver damage has occurred. NASH represents an advanced stage of the disease when the accumulated fat results in significant inflammation and scarring. This occurs in approximately a quarter of the patients and can lead to cirrhosis....NAFLD is linked to metabolic syndrome characterized by diabetes, obesity, and cardiovascular disease. The American College of Gastroenterology estimates that NAFLD affects 2 to 5 percent of Americans and 10 to 20 percent of those who are obese...."Understanding the natural history of NAFLD and NASH, especially identification of those patients with or at risk for NASH represents a critical unmet diagnostic need in liver disease," said [a medical specialist in the field]...."Currently a needle liver biopsy is the only way to definitively diagnose the disease so the development of a blood test to detect NAFLD risk would be a tremendous benefit to patients," said [another physician].

I have posted a number of previous notes about the obesity epidemic that is facing the U.S. population as well as the metabolic syndrome. These two problems can result in the accumulation of lipid in hepatocytes with inflammation and scarring in the more advanced cases. The fact that the liver can bear the brunt of these diseases should come as no surprise given the criticality of this organ in digestion as well as for a host of metabolic processes. There are so many important ideas reverberating through the article above that I don't know where to start when discussing them. Below are just a few of my initial reactions:

• Geisinger Health System, discussed in previous notes, has a history of innovation in healthcare delivery processes but is not known as as major research center. However, here is an instance where Geisinger is capitalizing on its development of a tissue/blood biorepository, a byproduct of its active bariatric surgery practice, to pursue worthwhile research goals in collaboration with Celera. This approach can serve as a financial and research model for other similar health systems.
• There are obviously many facets of the morbidity/mortality associated with obesity, the metabolic syndrome, and diabetes. Teasing out one disease continuum, fatty liver disease that progresses to NASH, and improving the diagnostic approach to these problems can only help to clarify their clinical management.
• A recurring theme in Lab Soft News has been the increasing use of molecular diagnostics and imaging as a partial substitute for biopsies and surgical pathology examination in the diagnosis of disease. A liver biopsy is not a trivial procedure and new diagnostic approaches are welcome. Moreover, a non-invasive blood test (described above as a "diagnostic array") for the diagnosis and monitoring of NAFLD/NASH can be repeated periodically and will lead to a better understanding of the natural history of this disease complex. Such a test will be even more valuable if it can serve to identify a patient's predisposition to the disease.

Reserved Hotel Suites for Lab InfoTech Summit Going Fast

If you are interested in attending the Lab InfoTech Summit on April 9-11 in Las Vegas, Nevada, you might consider booking a suite at the Venetian Hotel relatively soon. A total of 175 of them have been set aside for the conference at the special group rate of $239 per night and about half of this allocation has already been spoken for. The hotel reservation cut-off date is March 10, 2008. Reservations made after this date are still possible on a space-available basis but frequently at a higher rate. Of course, there are many other hotel properties on the Las Vegas strip and not everyone will want to stay at the Venetian. You may want to take a look at the conference brochure to view lecture titles, faculty names, and the workshop descriptions. On-line registration for the conference is available at the conference web site.

The Right Test for the Right Patient at the Right Time

My recent note about the lab test selection algorithms that are available at ARUP Consult (see: ARUP Offers Lab Algorithms for Disease Diagnosis Support) has caused me to think about the appropriate use of lab tests by clinicians. The title of this note -- the right test for the right patient at the right time -- reflects this theme. It's a mantra that is often quoted but often neglected. Lab test algorithms such as those published by ARUP are a tool that can promote appropriate test selection. Put another way, lab professionals should be in the business of actively promoting cost-effective healthcare delivery of which lab testing is an important component.

There is a group of laboratorians who place more emphasis on increasing the lab test volume than the effective use of lab resources. To defend their position, they will often say that it is not our place to "dictate" test-ordering strategies to clinicians. I respectfully disagree with this idea. For hospital inpatients with fixed DRG reimbursement, lab testing is a cost and anything we can do to hammer down the total cost of an inpatient stay will be beneficial. However and In the long run, the lack of initiatives to reduce inappropriate test ordering may cause a backlash against lab professionals. Radiologists are now learning this lesson the hard way (see: Stemming the craze on CT scans). Below is a quote from this article (boldface emphasis mine):

Now comes a scientific backlash to the imaging boom. A recent study by two Columbia University researchers in the New England Journal of Medicine argued that the overuse of CT scanning is adding up to dangerous levels of radiation exposure, especially in kids. Another risk is needless panic. A study from the National Institutes of Health found that 17% of patients getting tested for cancer had at least one false positive chest X ray over a four-year period, and 8% of women had at least one false positive ultrasound for ovarian cancer.

But let's get right down to the nitty-gritty here. In my opinion, clinical lab testing will continue to flourish even if lab professionals take more decisive measure to discourage the relatively small amount of inappropriate test ordering that is now occurring. The reason for this is that the clinical labs are about to experience a boom of unprecedented proportions ushered in by the era of molecular diagnostics, molecular medicine, and the use of large panels of biomarkers for wellness and disease monitoring. A large part of the early health and early detection/treatment models are based on this approach.

UnitedHealth Draws Criticism for Its Out-of-Network Reimbursement Policies

I have posted a number of previous notes about UnitedHealth, particularly with regard to its punitive policies toward physicians for out-of-network lab testing. The company now finds itself once more in the news (see: Health Plans Put Onus on Insured). The Attorney General of New York, Andrew Cuomo, is investigating the company's practices (Cuomo Announces Industry-Wide Investigation into Health Insurers' Fraudulent Reimbursement Scheme). Here's a quote from a recent  press release from Cuomo's office (boldface emphasis mine):

Attorney General Andrew M. Cuomo...announced that he is conducting an industry-wide investigation into a scheme by health insurers to defraud consumers by manipulating reimbursement rates.   At the center of the scheme is Ingenix...the nation’s largest provider of healthcare billing information, which serves as a conduit for rigged data to the largest insurers in the country.

When choosing health insurance plans, many of us elect a more expensive version that allows the use of "out of network" services. We elect this option obviously to broaden our care choices when we are not satisfied with those in the network. Such policies frequently stipulate that the policy holder will be reimbursed at 80% of the reasonable-and-customary charges for that service in the community. With increasing frequency, policy holders are now discovering that the reasonable-and-customary amount that is assigned to a particular service by the insurance company is very far below the amount actually billed. The 80% calculation then results in a very large out-of-pocket expenditure.

You have three guesses about who owns Ingenix, described in the quote above as the "nation’s largest provider of healthcare billing information" and as "a conduit for rigged data to the largest insurers in the country." Yes, you are correct. Ingenix is part of the UnitedHealth Group. I can't state my case regarding the impropriety of this situation more clearly than the New York Times in an editorial quoted below (A Rip-Off by Health Insurers?). Boldface emphasis is mine.

This system is an invitation for abuse. UnitedHealth owns the company whose database will affect its costs and profitability, so both have a strong financial interest in keeping reimbursement rates low. Even Ingenix seems unwilling to stand behind its numbers. In licensing its database to insurers, it stresses that the data is “for informational purposes only” and does not imply anything about “reasonable and customary” charges. Yet that is precisely what the health insurers use the data for, as Ingenix knows, according to investigators.

My question is: why is all of this only now coming to the attention of the public? Perhaps, though, it's a case of better late than never.