Lab Soft News

Some Lessons to Be Learned from the Dr. Oz Flap

I have been following the Dr. Oz flap closely but, with all the charges and countercharges, I am unsure where the truth lies (see: Dr. Oz Is No Wizard, but No Quack, Either). Oz has been accused of medical quackery in a letter to the dean of Columbia Medical School but it turns out that some of his accusers may not be so squeaky clean either and have their own agendas (see: The Attack On Dr. Oz Just Backfired In A Big Way). I have been trying to extract some sort of lesson from all of this and so an article by Paul Keckley of Navigant Healthcare caught my attention (see: Dr. Oz’s Odyssey). Here is an excerpt from it that made sense:

The odyssey of Dr. Oz seems a challenge to this entire body of unwritten rules. But it’s unlikely the 3 issues raised will be resolved in this high profile media flap.Dr. Oz is to Millennials what Marcus Welby is to Boomers: a credible clinician with a large, loyal fan-base.  But in the end, it falls to consumers to decide who’s right on these tricky issues. Health reform lacking an engaged consumer is likely to fall short. And engaged consumers are paying attention to Dr. Oz and online resources that augment what their clinicians say about not only their diagnoses and treatments but their supplements and food supply as well. They prefer it from their clinicians as part of their coordinated care plan, but in most communities, care coordination is inadequate or reserved only for those in certain health plans that benefit from engaged consumerism across the full continuum of health. Dr. Oz is likely to play and bigger role in how consumers navigate their health and decide who’s right on issues like these. The Oz odyssey fills a void between the information consumers want and what their clinicians choose to share.  My hunch is consumers will figure it out for themselves.

if physicians are dubious about some of Dr. Oz's pitches and feel that many of his claims are spurious, they need to get behind their own charismatic spokespeople who can communicate with the broad healthcare consumer audience like him. It's not enough to just take potshots at him from on high. Physicians need to present an alternate set of appealing ideas and proposals. They also need to remember that many people are suspicious  about any claims of organized, mainstream medicine. Here is a key paragraph from an article discussing how misguided the letter to the Columbia medical school was and how it has now backfired (see: A publicity stunt against Dr. Oz threatens to backfire spectacularly):

...I’ll go even further and suggest that Miller’s letter [attacking Dr. Oz), after the initial embarrassment it caused Dr. Oz, is probably now seen by him and his producers as a godsend that gives them the pretext to counterattack and to tar all the physicians—not just Dr. Miller and company, but other bloggers, me, and all the rest of us who have been criticizing Dr. Oz for the last five years over his promotion of quackery and pseudoscience—as being industry shills of some kind and to make it stick in the public mind. I’m sure he’ll find a way to go after Julia Belluz over at Vox.com as well, given that her excellent article on the making of Dr. Oz as a quack (my interpretation) was published the same week as Miller’s letter. That’s the narrative Dr. Miller has handed to Dr. Oz on a silver platter.

Columbia University can't be enjoying any of this. Tenured professor or not, the school needs to find some way to put Dr. Oz on the shelf and disassociate him from the school. This should have been done long before now and perhaps it's now too late.

LabCorp to Offer Direct Access Lab Testing (DAT) to Consumers

I began posting blog notes about direct access testing (DAT) some ten years ago (see, for example: Reinventing Direct Access Testing (DAT)) DAT, also called consumer directed testing, enables consumers to order lab tests for themselves via web sites in those states where it is legal. Even in states where it is prohibited, there were work-arounds using physician networks employed by the web site and providing MD test approval in the background. Quest even had a DAT subsidiary called QuesTest but exited this market in 2006 (see: QuesTest Bails Out of the DAT Market).

Although it has not emphasized its participation, LabCorp seems to have served as the back-end lab for most if not all DAT web sites. These web sites developed computer interfaces to LabCorp with PDF reports then sent to the web sites for customer retrieval. This LabCorp participation became obvious to consumers ordering tests because they were instructed to report to LabCorp patient service centers for their blood draws. It was recently revealed that LabCorp is now entering the DAT market with its own branded product (see: The Doctor Is Out: LabCorp to Let Consumers Order Own Tests). Below is an excerpt from the article:

Consumers will soon be able to bypass their doctors by going online to order cholesterol readings, thyroid tests and other bloodwork from the biggest diagnostics company in the U.S. Laboratory Corp. of America Holdings will let customers go online to pay for tests, visit a service center to get blood drawn, then view the results on the Web. The company has already been doing back-office lab work for a number of Internet firms that let people order up tests without a doctor. Rapid and at-home diagnostics are a growing corner of the health-care market, with businesses like WellnessFX Inc. and Direct Laboratory Services LLC tapping into demand from patients who want to get sensitive results in private or seek to monitor their health outside of the traditional doctor’s office.

Companies like LabCorp are tapping into demand from consumers who want to measure their bodies to monitor the effects of exercise and healthy living and to learn about their potential risks of disease....LabCorp is also facing competition as options emerge for consumers to get tests without visiting a service center at all. Startup Theranos Inc., founded in 2003, has developed a diagnostic kit available in some Walgreens locations that it says can provide a range of results, from lipid panels to the presence of HIV, with mere drops of blood. LabCorp’s direct-to-consumer business will initially be run online. The company is exploring a partnership with a drugstore chain as well -- an idea that rival Quest Diagnostics Inc. tried and scrapped....

WellnessFX, based in San Francisco, charges $988 for its most comprehensive package, which includes biomarkers for omega-3 fatty acids and fibrinogen, a protein produced by the liver. Customers can go to a Quest Diagnostics center to do their bloodwork, and they can add a 40-minute consultation with a physician to discuss the results. DirectLabs, based in Mandeville, Louisiana, offers more routine tests like a $29 metabolic panel -- glucose, kidney, fluids, electrolytes, calcium and liver -- and a $49 measure of Prostate Specific Antigen, which the company says can be used to detect cancer. Lab operators like LabCorp and Quest Diagnostics are looking for new sources of revenue as they contend with lower reimbursement from insurers and Medicare. The recent movement by hospital companies to buy up physician-owned medical practices has also siphoned off some of the lab work that the major providers had traditionally done.

I have covered Theranos in detail in the past which not only threatens the LabCorp and Quest business models but also those of the major in-vitro diagnostic companies (see: Finally, Some Important New Details about the Theranos Business Model). I also posted a note suggesting that the business models of Quest/LabCorp were suffering from the double whammy of lower reimbursements and the loss of private physician office business (see: Quest Diagnostics and Lab Corp Business Models Becoming Obsolete). I should emphasize, however, that Quest is still showing great quarterly profits. To offset a future anticipated loss of revenue, LabCorp now seems to be wisely expanding its business model to include DAT business.

We are now seeing some segmentation of the DAT market with WellnessFX offering a very expensive set of tests, including some esoteric tests, and an optional MD consultation. Quest provides the lab testing at the back-end. By way of contrast, DirectLabs offering inexpensive testing which has historically been the niche of most of the web-based DAT sites. This allowed them to appeal to cost-sensitive consumers, some of which were uninsured or had high-deductible health plans. 

My thanks to David Maizenberg (@biologypartners; @unpatients) who called my attention to the referenced article.

Reducing EHR "Wrong Orders" by Limiting the Number of "Open Charts"

A major medical center has been experiencing a spate of "wrong orders" with its EHR. After studying the problem, officials have determined that the best solution involves decreasing the number of open charts allowable from the current four to a maximum of two. With its prior legacy EHR and with the capability of opening only one e-chart, the wrong order problem had not been previously broadly encountered. Hospital officials surveyed other academic health systems across the country and found that this problem was not unique to it. Most of the other hospitals had also found it necessary to decrease the maximum number of allowable open charts from the default setting of four.  

I performed a Google search for "wrong orders" associated with EHRs and discovered that the acronym COWPIE (Charting On Wrong Patient In EHR) is now being used to describe this problem (see: Wrong-Patient Orders). Here's a quote from this article about the extent of the COWPIE problem in a pediatric hospital:

In the children’s hospital studied, automated surveillance identified 644 probable CPOE COWPIEs. Only four had been reported to risk management, presumably events in which the wrong patient received the medication. The great majority of these events appeared to qualify as near misses, with the errant order rapidly cancelled and replaced with the correct order for the correct patient.

It's of some interest in this cites study how the COWPIEs were identified ex post facto. Here's a description of the method:

if a provider 1) ordered a drug on a patient, 2) cancelled the order within 120 minutes, and 3) then reordered the same drug on a different patient within 5 minutes of cancellation, it is presumed to be an error. When the authors performed chart reviews on a subset of these automatically identified “errors”, they found that at least 60 percent and perhaps as many as 100 percent of the charts confirmed the error (documentation of the reasons for the provider’s actions often was ambiguous).

It's clear, even from these brief reports, that "wrong order" errors with EHRs are both very common and underreported. I raised this issue in a blog note about two years ago: Physician & Nurse Involvement in EHR Design; Patient Safety and EHR Gag Clauses. In that note, I linked to an article on EHR design (see: Clinician Involvement Critical in EHR Design, Improvement) and here is a quote from that article:

Some vendors prohibit users from sharing hazards even in the academic literature.  Although EHR systems that are marketed and sold to hospitals and healthcare providers are certified by the Certification Commission for Health Information Technology, that process is not focused on safety issues....and no mechanism exists to allow or encourage system users to address ongoing safety issues or provide feedback.

As far as I can tell and as illustrated by the case described above at a major medical center, EHR wrong order remediation issues are being resolved at the local level with other hospital experiences confirmed by telephone surveys. This kind of remediation based on local detection and effort does not lend itself to national organized efforts to reduce errors and improve patient safety on a broader basis.

Do Patients Need to Learn "Medicalese" to Converse with their Doctors?

I came across a rather innocuous article recently posted in the University of Michigan Health Blogs concerning medicalese, the language used by physicians, and the need for patients in learn it (see: Mastering Medicalese). Below is an excerpt from it:

Unless we happened to choose a career in the medical field, most of us gave up learning medical terminology in high school health or college biology. Then we get cancer and come face to face with big, incomprehensible words that we’ve done just fine not knowing our whole lives. It’s like going to a foreign land without speaking the language and being unable ask for directions. Since healing is our destination when we detour to Cancer Land, it helps to understand how to get there. And that means learning at least some Medicalese. I know it’s not easy. Like most people, I was thoroughly dazed and confused when I was first diagnosed. Not only was I facing a life-threatening disease, but suddenly I was hearing and reading huge words that I didn’t understand. No wonder I felt unequipped to make informed decisions that needed to be made....I was frustrated that I had to spend hours poring through medical dictionaries just to understand what was happening to my body and where, and I complained to my husband that Medicalese might as well be Greek....The National Institutes of Health has an excellent online medical dictionary with easy-to-understand definitions of most medical terms. Additionally, our doctors know that we’re not fluent in their language. I’ve found that some are better than others at translating Medicalese into lay language, but most will do their best if we ask.

Although the author of the article was well meaning, I found that I disagreed with her basic premise. I think that she is saying the following: many physicians will try their best to use a vocabulary that patients understand but patients would be well advised to learn medicalese. As part of the patient-centered approach to healthcare, this equation needs to be turned around. Physicians should be trained (and yes, obligated) to use a vocabulary that the majority of patients will understand. If the patient has a knowledge of some medical terms, so much the better, but this should not be an obligation on his or her part.

All individuals with some form of specialized training from auto mechanics to financial planners have their own "job" vocabularies but they understand that they need to communicate with their customers or they will lose business. We tend to shy away from labeling patients as "customers" but we might be farther ahead if we did so in terms of the expected behavior of the physicians caring for them.

My first reaction to the article quoted above was to question the value of the NIH online medical dictionary but then thought better of it. Such a dictionary could be essential in enabling patients to use the web to seek information about their diseases which I think is both necessary and appropriate. I have referred to this phenomenon in past notes as consulting Dr. Google (see: Paging Dr. Google! We Are Waiting for a Second Opinion; Teaching Consumers to Say "No" to Physicians' Recommendations; Unreported Drug Side Effects Discovered by Analysis of Google Big Data). 

Hospital EHR Interoperability; Are Vendor Fees Necessary or Even Appropriate?

In response to my recent note commenting on the EHR vendor and hospital approach to EHR interoperability (see: NYT Op-Ed on EHR Interoperability Blames Vendors and Greedy Hospitals), Grahame Grieve submitted the following comment:

it actually costs money to share data. So surely it's reasonable to pay the people who do the work to do it. Maybe the problem is that the benefits of sharing the data don't accrue to the people who actually share it. It's the recipients who do - but there's no obvious channel for them to pay. So, a broken market. With all the consequential behaviour.

Apropos of the topic of interoperability and the fees, or lack thereof, for data sharing, a recent article on this topic provides a reasonable response to Grahame's comment (see: As HIMSS15 Ends, Health Data Interoperability is Just Starting). Here is what I consider the key from the article:

In a keynote address, Humana CEO Bruce Broussard urged all players in the industry to embrace health data interoperability as the foundation for a new perspective on chronic disease management, population health, and overall wellness for patients. “We need to look at information as a shared asset, not a proprietary one,” Broussard said. “It’s an asset that should be flowing through our system freely, as opposed to being taxed and tolled along the way. We all can benefit from interoperability.  As you think about the silos in the industry and you think about value-based reimbursement and you think about customers, interoperability is really the intersection of that.  It’s the ability for us to start acting like a team in healthcare, as opposed to everyone thinking they’re a track star.”

I think that it's both appropriate and gratifying for Bruce Broussard, the Humana CEO, to emphasize the need to start thinking about healthcare as a team effort rather than the sum of the efforts of individual "track stars." One could reasonably assume that Mr. Broussard would have more of a track star mentality, given his yearly  compensation, reported by Bloomberg, of $10,155,401. However, I think that his idea is worthy of consideration. He says: information should flow freely and not be taxed or tolled. His quote does beg the question, however, as to what are the "team effort" characteristics required of healthcare professionals.

One of the qualities that may have been lost in recent years by some hospital executives, in my opinion, is the notion that healthcare is a "higher calling" and that success can't always be measured by P&L statements. I believe that the stratospheric compensation packages provided to some hospital executives can easily distort this higher calling. Hospital executives seem to me to be on an endless quest for higher profits to fuel their ever-increasing compensation packages (see: Hospital Entrepreneurship and the Salaries of Non-Profit Hospital Executives). I'll let you decide whether you believe that the executives managing hospitals should continue to pursue their entrepreneurial instincts at the risk of decreasing the quality of care.

NYT Op-Ed on EHR Interoperability Blames Vendors and Greedy Hospitals

The major barriers to the easy exchange of health data among hospitals, EHR interoperability, have been erected by EHR vendors and hospital executives under a tacit collaboration agreement (see: Hospitals as a Major Barrier to the Pursuit of EHR Interoperability). A recent op-ed piece in the New York Times casts a harsh glare on who's behind this problem (see: Roadblocks to Sharing Medical Records). Below is an excerpt from it:

The ability to transfer electronic medical records from one doctor or hospital to another is essential to the smooth functioning of the health care system and to providing the best possible care to patients. Yet all too often these transfers are being blocked by developers of health information technology or greedy medical centers that refuse to send records to rival providers.

The tide has now turned with Congress attempting to enforce EHR interoperability but with industry observers commenting that one new legislative approach, not based on standards, is doomed to failure (see: Congress Can’t Solve the EHR Interoperability Problem). All of this attention on the need for clinical data exchange is beginning to have the desirable effect of shaming vendors into taking some minor steps in the right direction. For example, Epic is removing  the fee for data exchange between its client and non-client hospitals (see: Epic to remove data sharing fee). Below is an excerpt from the article reporting this change:

Epic Systems has revealed that it will no longer charge a fee for data transfers between Epic customers and non-Epic customers through its Care Elsewhere module. "We're not going to charge for Care Everywhere for at least until 2020," CEO Judy Faulkner said. Up until now, clinical messages sent to an information exchange were priced at 20 cents apiece and incoming messages from a non-Epic system cost $2.35 for that patient for a year for unlimited messages, according to Faulkner. Faulkner said the company "thought it was cheaper" than its competitors, but found the fees irritated customers.

....Realistically, those fees only account for a tiny, tiny portion of Epic's revenue, so it seems smart to simply eliminate them, given what looks like increased federal heat. The ONC filed a Congressional report last week accusing "some healthcare providers and health IT developers" for "knowingly interfering with the exchange or use of electronic health information in ways that limit its availability."Plus, the company's competitors are already moving in that direction (Modern Healthcare's Joseph Conn called this an "interoperability price war".... Cerner CEO Neal Patterson said last year that the company would cover its customers' costs for participation in CommonWell Health Alliance through 2017, and athenahealth said it would do so indefinitely through the Alliance. 

The data exchange fees required of  non-Epic clients have been the subject of previous articles (see: Epic Systems feeling heat over interoperability). Here is a quote from one of them:

...[An] Epic spokesman said those fees are more than offset by the cost savings achieved from electronic records interchange...[An industry observer was skeptical about this.] “While I don't know for certain, I would be shocked if the average interface is only a few thousand dollars,” she said. She suspects that hospitals have to pay more to share data with its peers using different EHR systems. While interface fees are common across the EHR industry, some observers say Epic's leading role in the EHR market means it has a disproportionate negative effect on interoperability.

The fact that Epic has charged non-client hospitals for data exchanges illustrate a key fact about the attitude and culture of the company. It has previously viewed the ability to provide free interoperability as a "feature" for its clients rather than as a service for all patients. It has now reversed this policy in response to governmental pressure. For me, it's also refreshing that the NYT recognizes that hospital executives have been complicit with EHR vendors in this lack of easy sharing of hospital data. The availability of such data allows patients to shop hospitals for the lowest price and highest quality providers.

GE Medical Exits Hospital EHR Business, Concentrates on Ambulatory Software

This will probably come as no surprise to veteran observers of healthcare IT but GE Healthcare is exiting the hospital EHR business. Details were presented in a recent article (see: GE Healthcare: We’re leaving the hospital EMR business) with an excerpt below:

GE Healthcare confirmed at HIMSS 2015 that it is getting out of the hospital electronic medical records business. The company acknowledged...that three years ago it decided to phase out its Centricity Enterprise product. It’s now helping customers with a “graceful transition over a number of years,” said...[the] general manager of clinical business solutions at GE Healthcare.....[He] said that GE Healthcare is instead “doubling down” on its ambulatory EMR products, Centricity EMR for large medical groups and its integrated Centricity Practice Solution EMR and practice management system, which is meant for smaller practices. The hospital-focused Centricity Enterprise product generated about 5 percent of total electronic medical records revenue....In 2013, Centricity products were thought to generate close to $300 million, making GE Healthcare a distant sixth in EMR vendors. That was more than $100 million behind fifth-place athenahealth that year....But the gap doubtlessly grew since then. Industry insiders gossiped endlessly about GE Healthcare’s struggles with Utah’s Intermountain Healthcare ....Intermountain left GE for Cerner in 2013, prematurely ending a planned 10-year, $100 million collaboration that began in 2008. Intermountain first installed Centricity in 2006, the rebranded name of what used to be called IDX Carecast, shortly after GE acquired IDX Systems.

I don't think that there is such a thing as a "graceful exit" from the hospital EHR business, at least from the perspective of the hospitals running the product in question. They may sometimes view their current EHR as inadequate but it's always painful to rip out a complex system and replace it with another. The purchase price of a new system can also be challenging in terms of acquiring new capital. I think that, in this context, GE is suggesting that "graceful" means that current Centricity customers are given an adequate period of time to search for a replacement system and install it.

I have serious doubts about the notion that GE will be successful in "doubling down" on its ambulatory care EHR product line. Integration is one of the most important drivers in healthcare IT. The term can be used to apply to interoperability between hospitals and also between hospitals as well as with the physician practices and clinics from which hospital patients are referred. Hospitals are also continuing to purchase local physician clinics and practices. All of this adds up to the fact that hospitals often seek ambulatory EHRs from the same vendor as their hospital EHR. This can achieved with what I call "shrink wrap" integration which is to say that system integration is provided by the vendor. In the light of this preference, I don't think that GE's focus solely on ambulatory care products is destined for success.

Are any of the readers of this blog old enough to remember the GE sloan of the past? Roughly paraphrased, it was: we will be first in a market or we won't stay in it. I think that this focus solely on ambulatory care EHR products will be short-lived and is intended mainly as a face-saving gesture for the company.

Xifin's ProNet Expands Business Opportunities in Telepathology

In a previous post (see: XIFIN and vRad Partner to Pursue an Integrated Diagnostics Business Model), I discussed the new partnership between Xifin and vRad. Xifin had previously purchased PathCentral with its cloud-based AP-LIS and telepathology consultation network (see: XIFIN acquires pathologist software developer PathCentral). This latter product is now called ProNet. I had the opportunity recently to view a demo of ProNet with Joe Nollar, director of product development for Xifin. This provided me with an opportunity to think about some of the business opportunities that such a network would make available for surgical pathologists, some of which Xifin is already discussing with current and potential clients. This idea will be the main thrust of this note.

It occurred to me that one could roughly divide the telepathology consultations into three segments: (1) esoteric surgical pathology consultations for which the requesting pathologist is seeking the interpretation of a "marquee" consultant with an established record of expertise or, alternatively, the services of a prestigious pathology department; (2) routine histopathology work generated by a smaller hospital temporarily or permanently lacking a pathologist with work sent either to an intra-system pathologist(s) or to some outside consultant for a standard fee; (3) philanthropic surgical pathology services such as those rendered to hospitals in developing countries with a lack of pathologists and lab services and presumably without any, or very modest, reimbursement.

As I understand the current telepathology environment, a number of a academic pathology departments such as UPMC have developed one-to-one telepathology services, often with a single hospital abroad or perhaps linked to a for-profit health system in a country like China. Such relationships are designed to serve the more affluent healthcare consumers who are seeking U.S. expertise for complex cases. Xifin will make their ProNet network available to any qualified individual or pathology department in the U.S. Joe Nollar of Xifin mentioned to me that ProNet can also accommodate branded telepathology services whereby a prestigious academic department would market a wide range of telepathology services and then triage cases, as received, to the most appropriate pathologist on the staff. By using ProNet, the department avoids the capital expenditure of developing its own one-off network and takes advantage of the experience of the company and its software. Xifin may also be able, at some point, to provide marketing services to its departmental clients.

I need to think more about the opportunities that ProNet offers to a variety of pathologists. Generally speaking, it would seem to offer a way to optimize the efficiency of the allocation of pathology services on a global basis. Most importantly, it also provides a powerful financial incentive to both established surgical pathology consultants and also to younger practitioners to become skilled in digital pathology in order to both expand their career opportunities as well as to generate revenue for their home departments. I have been waiting for such a development whereby junior faculty members demand the deployment of digital pathology in their departments over the objections of senior members who may be lukewarm about the technology.

Can Information Technology Be Used to Shrink Staff and Healthcare Costs?

I learned early in my career in pathology informatics that it was unwise to promise cost savings or staff reductions in exchange for capital investments in IT. Requests for such savings were often the response to my requests for new software. I just came across an article suggesting that the federal government could shrink itself by grater investments in IT which runs counter to my experience, at least in healthcare (see: The Government Found A Way To Significantly Shrink Itself: Invest In Tech). Below is an excerpt from the article:

Techies in Silicon Valley often tell me that if they do their job right, it should cease to exist. The goal of a lot of technology is to reduce the amount of human labor necessary to get something done. A new study finds that this Silicon Valley maxim holds also true for the government — in a massive way. ...[The author of the study] found that for every $1 governments invested in technology, spending decreased by $3.50.....Between 2000 and 2008, [he]...analyzed government spending and the portion dedicated to information technology. After controlling for population, debt, grants and other institutions that tend to bloat government, there was a clear relationship between states that invested in tech and lower-than-expected public expenditures....Indeed, this is why the government is proposing a significant increase in budget for the newly created United States Digital Service, an internal IT consultancy group. The USDS was created out of the Healthcare.gov meltdown fiasco, and pays sizable wages to attract top talent from the likes of Google and Facebook, who build out government web services more efficiently. The government can shrink itself, if it invests wisely. Read a free version of the study here.

On the face of it, it would seem logical that the deployment of IT should be able to shrink departmental budgets and the number of FTEs. After all, computer systems serve as a substitute for human labor and personnel costs account for a large percentage of the cost of healthcare. At least in pathology and for the more than three decades that I was involved with pathology informatics, the number of personnel in my unit was stable or increased across time despite nearly continuous investment in new software. Off the top of my head, here are a few reasons why I think that this was the case:

• Consider the S-shaped curve that governs the deployment of new technology (see: Organizational Innovation with a Political Overlay; Lessons from Microsoft). After new software is installed, productivity is flat or even declines due to the fact that the new product is not well understood. Productivity soon increases exponentially until it tails off again as the software becomes obsolete. It seems that my LIS group was always in the process of deploying new software so that we were not able to take advantage of the near vertical part of the S-curve. In other words, we never seemed to be able to take advantage of the "sweet spot" of the S-curve.
• Secondly and most importantly, the deployment of new software almost always provided new functionality for the department that was not previously available. This increased the demand for the services of my personnel and the need to hire more of them. In short, new and improved software demanded more personnel to install and maintain it.
• Finally and focusing on the cost of personnel, LIS improvements usually decreased the need for entry-level personnel such as those in the specimen processing or phlebotomy units with lower salaries than IT personnel. New software deployment might have been decreasing the head-count in other pathology units but not in pathology informatics.

All of this discussion then begs the question: what type of software applications allow an organization to reduce staff and thus reduce costs. In other words, how can IT help to shrink state and federal government for example? I think that the answer is probably in in various types of service applications such as ordering license plates or applying for health insurance. Needless to say, the latter is not a slam-dunk (see: The Health Insurance Marketplace: A Case Study of Incompetence). In other words, if and when a governmental agency deploys new IT, it needs to employ competent personnel to optimally perform the task of software development and management.

EHRs Pose a Major Threat to Fields of Pathology & Radiology Informatics

The deployment of enterprise-wide-solutions (EWSs) provided by vendors like Epic and Cerner poses a major threat to the future development of pathology and radiology informatics. The crux of the problem is that such EHR installation always move system control to the central IT departments and away from pathology and radiology. This is in contrast to the more than three decade history of pathology and radiology managing their own LISs, RISs, and PACSs. It turns out that that the radiology informaticians have been worrying about this problem in addition to the concern in pathology (see: Rethinking Radiology Informatics). A recent article voiced the concern of radiology informaticians about this turn of events and below is a quote from the article:

Radiologists are groundbreaking leaders in clinical information technology (IT), and often radiologists and imaging informaticists created, specified, and implemented these technologies, while also carrying the ongoing burdens of training, maintenance, support, and operation of these IT solutions. Being pioneers of clinical IT had advantages of local radiology control and radiology-centric products and services. As health care businesses become more clinically IT savvy, however, they are standardizing IT products and procedures across the enterprise, resulting in the loss of radiologists' local control and flexibility. Although this inevitable consequence may provide new opportunities in the long run, several questions arise.

Drs. Kohl, Dreyer, and Geis, the authors of this paper, emphasize the loss of "local control and flexibility" for radiology with the deployment of EHRs like Epic. With such enterprise wide solutions, the ordering and reporting of radiology procedures is transferred from the RIS to the EHR. In previous times, control of ordering and reporting rested with departmental systems. This problem is going to get even worse as vendors like Epic develop their departmental modules like Beaker and Radiant that completely replace the department-based LIS and RIS. This will result in only a minimal set of IT responsibilities for pathology and radiology informaticians.

There is no question that standardization is a major impetus on the part of hospital executives for choosing an enterprise-wide-solution. Another objective, one that is usually unstated, is the desire by hospital executives to exercise control over all clinical information. The downside of this need to control all hospital IT is the loss of the skill sets of pathology and radiology informaticians. Although the radiology article cited above lays out the current challenges to radiology informatics, I believe that it falls short in proposing solutions for the future for both it and pathology informatics.

My initial response to this problem was that pathology informaticians, when pressed by the EWS model, should cede control of basic lab IT functions to the hospital central IT group and concentrate on the more interesting and esoteric components of lab computing such as molecular diagnostics and cancer genomics. Unfortunately, the Epic EWS is architected with only a single huge database with inadequate tools to support analytics and research. I have therefore come to the conclusion that, in the long run, the only possible solution for pathology and radiology informaticians is to deploy and manage their own LISs, RISs, and PACSs in the cloud with a return to departmental control.

I want to thank Dr. Bob Miller from the pathology department at Johns Hopkins Medicine for calling the radiology article cited above to my attention.