Lab Soft News

Cancer Survivorship and the Role of PCPs in Continuing Care of Cancer Patients

I have posted a number of previous notes about cancer survivorship (see: Cancer Survivorship, an Emerging Subdiscipline in Oncology; "Chemo Brain" Can Persist for Three to Five Years; Exercise Can Help Reverse; New Research Casts Spotlight on "Chemo Brain"). However, I have not given much thought to exactly which physicians would administer long-term care to cancer survivors. A recent article addressed this topic (see: Study: Doctors differ in how best to care for America's 12 million cancer survivors). Below is an excerpt from it:

There are major differences between oncologists and primary care physicians regarding knowledge, attitudes, and practices required to care for American's 12 million cancer survivors. That is the key finding of the first nationally representative survey of doctors that reveals how these differences pose significant barriers to effective communication and coordination of care following initial cancer treatment. The study...included results of a survey of 1,072 primary cancer physicians (PCP) and 1,130 oncologists randomly sampled from among all U.S. practicing physicians during 2009. The authors say the findings are of considerable concern, given the rapidly increasing cancer survivor population and looming shortages of both oncology specialists and primary care physicians who will be needed to provide care to this population. Because many individuals stop seeing their oncologist after the first few years following their initial cancer treatment, primary care doctors often assume a large share of providing care to survivors. Optimal cancer survivorship care includes not only surveillance for recurrence or second cancers, but also addressing the long-term and late medical effects of cancer or its treatment, providing psychosocial support, and managing other diseases or conditions. Several factors may present challenges to ensuring a smooth transition for survivors as they move from initial acute care to post-treatment care...."What the survey tells us is that many doctors, particularly primary care doctors don't have a high level of confidence in their own knowledge of some aspects of survivorship care, and many oncologists believe that PCPs are not adequately prepared to provide such care. We also see some evidence of knowledge deficits in both physician groups in terms of guideline-based care for survivors," [the study author] says.

First of all, this article succinctly highlights the key elements of the care of cancer survivors that I now repeat for emphasis:

• Surveillance for recurrence or second cancers
• Addressing the long-term and late medical effects of cancer or its treatment
• Providing psychosocial support for the patient
• Managing other diseases or conditions of the patient

I don't think that it's a good idea to disrupt the relationship of cancer survivors with their PCPs. Hence, this last item, managing other diseases or conditions of the patient, should obviously  be the responsibility of this set of physicians. However, I think that the first three items in this list would be best managed within cancer centers. I understand that they may not be viewed as goals currently embraced by such centers. Cancer care has increasingly become more and more specialized. Hence, the observation of the study regarding primary care physicians: [they] don't have a high level of confidence in their own knowledge of some aspects of survivorship care. It also seems to me that survivorship research would be greatly enhanced if carried out to some extent in cancer centers.

Higher Quality of Services When Physician Executives Run Hospitals?

I have most commonly worked in hospitals where the CEO's major strengths were in financial management -- they were not MDs. Once, in the Army, I worked in a hospital where both the commander and deputy commander were physicians. My general impression has been that the former executives tended to stress the need for quality of care but often had little real understanding about the processes for achieving quality or the true cost of these processes. A recent article caught my attention and asks the question whether hospitals should be run by physicians (see: Should Hospitals Be Run by Doctors?), Below is an excerpt from it:

The conventional wisdom is that doctors should focus on patient care, and managers with a business or administrative background are better suited to running the day-to-day operations of a hospital. Among the nearly 6,500 hospitals in the United States, only 235 are run by physician administrators, according to a 2009 study in the journal Academic Medicine. But now new research suggests that having a doctor in charge at the top is connected to overall better patient care and a better hospital. The findings, published in the journal Social Science & Medicine, are based on a review of 300 top-ranked American hospitals in the specialties of cancer, digestive disorders and heart surgery. [The study author] tracked the professional background of each hospital’s chief executive and then compared the performance of physician-run hospitals with that of hospitals overseen by someone with a nonmedical background. The study found that overall hospital quality scores were about 25 percent higher when doctors ran the hospital, compared with other hospitals. For cancer care, doctor-run hospitals posted scores 33 percent higher....The research doesn’t offer any evidence as to why doctor-run hospitals appear to post better results, but [the author] said it may be because doctors truly understand “the core business of health.”....[The author] also noted that spending time caring for patients puts doctors in a better position to make the hard decisions that a hospital administrator faces. “M.D. C.E.O.’s are more likely to prioritize patients because patient care is at the heart of their education and working life as a physician,’’ she said. “When it comes to making hard budgetary decisions or rationing choices, M.D. C.E.O.’s may be able to make more informed decisions.”

I think that it is highly unlikely that there will be any significant movement in the direction of having more physician CEOs in U.S. hospitals. As noted in the article above, only 3.6% of hospitals are currently run by physicians although the appointment of physician CEOs may skew toward larger hospitals.The reasons for this conclusion is simple. First, there will be a limited supply of qualified physician candidates for such positions. Because the field of medicine is very broad, physicians who don't like direct, or even indirect, patient care can generally find other well-compensated positions that satisfy them such as in pharmaceutical companies. Moreover, many hospitals are being consolidated into large health systems for which skilled financial management is becoming even more critical. Most health system CEO searches will therefore focus on candidates with ever-increasing financial management expertise rather than a deep knowledge of clinical affairs.

In my opinion, one of the greatest challenges for finance-oriented, non-MD hospital CEOs is cost-cutting. In response to a call for budget cuts in hospital clinical units, the physicians and nurses in such units will frequently assert that the cuts will put patient lives at stake. Because of their lack of deep knowledge of clinical issues, these CEOs are often unable to refute such claims in any substantive way. Their only recourse, then, will often be to make budgetary cuts across all clinical units. A similar problem occurs for these CEOs when they want to invest in technology. They tend to focus on new tests, procedures, and operations with pent-up demand and high profit margins. They also may be reluctant to invest in the service infrastructure with quality implications such as IT without financial incentives. I have often had to struggle for additional capital investment in lab software such as the LIS.

Highest Risk for Anesthesia Mortality among Older and Multimorbid Patitents

Is general itself anesthesia dangerous? It depends, in part, on your age and co-morbidities when you undergo an operation (see: Is anesthesia dangerous?). I have always been interested in the mortality rate of general anesthesia but the numbers have been hard for me to find until now. Below is an excerpt from an article on this topic:

In pure numerical terms, anesthesia-associated mortality has risen again. The reasons for this are the disproportionate increase in the numbers of older and multimorbid patients and surgical procedures that would have been unthinkable in the past. This is the result of a selective literature review of André Gottschalk's working group at the Bochum University Hospital....In the 1940s, anesthesia-related mortality was 6.4/10,000. By introducing safety standards such as pulse oximetry and capnometry, the rate was reduced to 0.4/100,000 by the late 1980s. This value still applies for patients without relevant systemic disease. However, mortality has risen in patients with relevant comorbidities (0.69/100,000). Such comorbidities include heart failure, angina pectoris, chronic renal failure, or severe malignant hypertension. Because of improved safety standards such patients can have surgery nowadays—something that would have been unthinkable in the past owing to their multimorbidity. Another factor explaining anesthesia-related mortality is the fact that the proportion of patients who are older than 65 rose in Germany from 28.8% in 2005 to 40.9% in 2009.

Here are a couple of definitions that may help you understand this article:

• Pulse oximeter: a medical device that indirectly monitors the oxygen saturation of a patient's blood and changes in blood volume in the skin, producing a photoplethysmograph.
• Capnometry: the monitoring of the concentration or partial pressure of carbon dioxide (CO2) in the respiratory gases. Its main development has been as a monitoring tool for use during anaesthesia and intensive care.

The challenge for anesthesiologists is not going to get any easier in the future. The U.S. population is certainly getting older and the rate of obesity and other chronic health diseases is rising (see: Breast Surgery Proves Riskier for Obese Women). However, anthesiologists have risen to similar challenges over the years. I can only assume that they will continuously improve their record of keeping obese, older, and multimorbid patients breathing under general anesthesia as their experience with such patients broadens.

Trends in Pathology Fellowship Training; Speculation about the Job Market

A recent article on fellowship training in pathology got me thinking about the number of years of postgraduate training that is required for training in pathology and also the job market (see: 2011 Fellowship & Job Market Surveys), Below is an excerpt from the article:

Each year the ASCP Resident Council directs [a] survey on fellowships and the job market for pathologists in training, both residents and fellows....This year, 2,591 residents participated in the survey. Competition is tight for fellowships. Similar to both the 2009 and 2010 results, slightly more than half of residents (54 percent) received one fellowship offer; 21 percent received two fellowship offers. But 12 percent of applicants received no offers (up from 10 percent last year). While 59 percent of residents intend to complete one fellowship, 40 percent of residents intend to complete at least two fellowships, up from 38 and 36 percent in 2009 and 2010 respectively.  Surgical pathology, hematopathology, cytopathology, gastrointestinal/hepatic pathology and dermatopathology ranked as the top five most popular fellowships. Most pathology residents in their third and fourth year do not apply to pathology jobs immediately after training. Of those who did, 33 percent sought a community group practice, 24 percent sought an academic role, and another 24 percent had no preference. Starting annual salaries are between $100,000 to $250,000 for residents taking pathology jobs immediately after residency, most (37 percent) landing in the $150,000 to $200,000 range for both academic and community group practice. Forty-two percent received one offer and 39 percent received no offers....Those applying for jobs after completing one fellowship also ranked job security as a top reason for applying to and accepting positions. Interestingly, the majority of those in forensics and transfusion medicine sought additional fellowships because it fit future plans, whereas those in hematopathology cited enhancement of skills or employment opportunities as the reason for additional fellowships. This trend is reflected in the number of jobs applied to after fellowships. Thirty-seven percent of hematopathology fellows applied to more than 10 jobs. Those in forensics overwhelmingly applied to between one and three jobs.

Here's a quote from a web page about the number of postgraduate years required for pathology training (see: Training Requirements for Pathology)

Training consists of a minimum of four years of postgraduate education for anatomic and clinical pathology. There are also training programs in anatomic pathology only or clinical pathology only which require three years of postgraduate education. There are 149 pathology/anatomic and clinical residency training programs accredited by the ACGME for 2009/2010 offering over 500 categorical positions available to U.S. seniors. [Subspecialty fellowships require an additional one to two years of training.]

Here's some of the ideas that I have derived from these materials about current trends, and the future, of pathology training:

• The majority of pathology residents these days will complete one and often two fellowships.
• The job market in pathology does not seem to be too robust: forty-two percent of trainees received one job offer and 39 percent received no job offers.
• This idea about a weak market, or perhaps a weak applicant pool, is reinforced by another finding in this survey: thirty-seven percent of hematopathology fellows applied to more than ten jobs.
• It may be the case that part of the demand for fellowship positions can be accounted for by trainees without any job offers who are marking-time and hoping that the additional training will make them more competitive later.

I have a hunch that the job market in pathology will not improve soon. My prediction is based on two assumptions. The first is that many older pathologist are hanging on beyond the normal retirement age because the recent economic downturn has whittled away at their savings. Secondly, I think that digital pathology will soon make major inroads into pathology departments. If this happens, it may replicate in pathology what is happening in the radiology job market (see: Radiology Jobs Trend Downward; Blame Technology and Reimbursement). Despite the grousing about digital pathology, I believe that it will eventually enable surgical pathologists and hematopathologists to work more efficiently. This will take five or more years and require a young cohort of pathologists for whom manipulating digital images will be second nature. More mature digital technology will also contribute to this enhanced work efficiency.

Early Detection of Alzheimer's Disease: Mutations of Three Genes Studied

From a diagnostic perspective, one of the hottest current issues is the early detection of Alzheimer's disease. The volume of testing for Alzheimer's will undoubtedly be large so that the commercial market for a lab test will be attractive (see: Diagnosing Alzheimer's Disease with Imaging and Biomarkers). Also, obtaining a population of patients with documented early disease is critical in the development of clinical trials for early drug treatment. Below is an excerpt from an article about early testing for three gene mutations (see: Clinical trials to detect Alzheimer’s 20 years before dementia onset planned):

Inherited forms of Alzheimer’s disease may be detectable as many as 20 years before problems with memory and thinking develop....Identifying Alzheimer’s in its earliest stages is a top priority for researchers. Many think that by the time symptoms become apparent, Alzheimer’s disease has already damaged the brain extensively, making it difficult or impossible to restore memory and other mental abilities....Initial...results [from an international study network] confirm and expand upon earlier insights from studies of the more common sporadic forms of Alzheimer’s, including data suggesting that changes in the levels of biological markers in the spinal fluid can be detected years before dementia....[R]esearchers are studying members of families who have mutations in one of three genes: amyloid precursor protein, presenilin 1 or presenilin 2. Participants with these mutations are certain to develop Alzheimer’s disease early, with symptoms beginning in their 50s, 40s, or, in some rare cases, 30s....By looking at the age of symptom onset in a parent who passed an Alzheimer’s mutation to a...participant [in the study network], scientists can establish an estimated age of onset for a study participant. If a parent developed dementia at the age of 50, they would expect a child who inherited the mutation to develop dementia at roughly the same age. As a result, scientists can start amassing a detailed chronology of disease progression that covers the many years Alzheimer’s is active in the brain but still before the onset of dementia....Washington University researchers....will report initial results from [the study network], including confirmation of the value of disease indicators from cerebrospinal fluid analyses. Participants who carry the mutations but are still asymptomatic have significantly lower levels of amyloid beta and higher levels of tau protein in their cerebrospinal fluid than participants without the mutations. 

I think that most patients will readily agree to undergo biomarker testing to detect early stages of cancer. This is because early detection of most types of the disease can effect a cure or amelioration of it. However, I suspect that some, or even most, patients might not be as eager to undergo testing to diagnose early stage Alzheimer's. This is a consequence of not having effective drug therapy for the disease at this time. There's a chicken-and-egg problem here -- we need the patients to run the clinical trials of various drug candidates to find a cure.

It also occurred to me as I was writing this note that I was focusing mainly on the lack of drug therapy for Alzheimer's, particularly in those genetically predisposed to developing it. It did not immediately occur to me that there were potentially other ways to forestall or prevent the disease such as changes in lifestyle (Exercise and a Mediterranean Diet Lower the Risk of Alzheimer's Disease). This thought was reinforced when I discovered another article that suggested that exercise can stave off dementia in the elderly but this was primarily cognitive problems resulting from vascular disease (see: More, better evidence that exercise staves off dementia: two new studies).

The iPhone Effect: Smartphones and Their App Ecosystems Have Changed Everything

I firmly believe that the iPhone "changed everything" by introducing us to what is now referred to as the smartphone app ecosystem (see: New Definition for "Apps": The Smartphone Market for Medical Software). Think about the rules of this ecosystem and how they differ from what we were accustomed to when using only PCs: (1) we shop for new apps online and many of the best ones are free; (2) we are promptly notified about operating system and app upgrades when they become available and can make these changes quickly and online; (3) the device easily fits in our pocket, giving us ready access to all of our apps including a rapidly improving camera and video recorder; and (4) most of the available apps are easy to learn and useful. A recent article discussed this iPhone effect and made a number of salient points (see: The iPhone Effect: How Apple’s phone changed everything), Below is a listing of these key points:

The rise of smartphones

• Smartphone adoption in first quarter 2008 was 10 percent according to Nielsen. Nielsen predicts smartphones will outnumber feature phones by the end of this year.
• Worldwide smartphone sales will hit 468 million this year and reach 1.1 billion by 2015, according to Gartner.

Smartphones open the door for tablets

•  Apple has sold about 19.5 million iPad through the first quarter of this year.
•  Gartner estimates there will be 294 million tablets in 2015.

App ecosystems thrive

• Apple’s App Store now boasts 425,000 apps, with 14 billion app downloads and $2.5 billion paid to developers to date.
• The Android Market has 200,000 apps and has had 4.5 billion downloads as of May.
• IDC now expects 182.7 billion mobile app downloads by 2015 across all platforms.
• Canalys estimates mobile app revenue will hit $14.1 billion next year and rise to $36.7 billion by 2015.

I really have nothing to add to this list other than to say that I am currently personally using an HTC Evo 4G android phone that continues to delight me on a daily basis. The battery life of the phone is a little disappointing but it's constantly working to update my Gmail and Google Calendar. I am therefore willing to accept this downside on the basis of the useful services that it provides on a continuing basis. I am now also using a Bluetooth headset so that my phone can stay plugged to an electrical outlet during my weekly longish conference calls for which I use Google Voice so the time does not count against my cellphone minutes.

Orchard Promotes Its CP/AP LIS as an Integrated Diagnostics Solution

Integrated diagnostics can be achieved by breaking down the sub-specialty silos in the diagnostic specialties like pathology, laboratory medicine, and radiology. The current standard of practice is to present the procedure and test-ordering clinicians with individualized reports from the various specialty labs (e.g., surgical pathology, immunology, microbiology) and radiology units like MRI and CT. This approach results from the super-specialization and subdivision of the diagnostics specialties. Under the integrated diagnostics mantra, the goal of integrating the diverse diagnostic reports is moved upstream and becomes the responsibility of the diagnosticians themselves rather than the clinicians.

In my mind, the first major movement toward integrated diagnostics on the manufacturing side occurred five years ago when Siemens purchased Bayer's diagnostic unit and DPC (see: Bayer Sells Lab Diagnostics Unit to Siemens; Rationale Behind Siemens Purchase of DPC), Recall that prior to this time, Siemens had a major presence in medical imaging but not in in-vitro diagnostics. I described this shift as the emergence of a molecular medicine strategy on the part of Siemens (see: Siemens Pursues Molecular Medicine Strategy; Siemens' Pursuit of an Integrated Diagnostics Portfolio).

In the intervening five years, there has been modest progress in the evolution of integrated diagnostics at the hospital level but nothing substantial among the major LIS vendors which generally sell an LIS that supports clinical pathology and an AP-LIS system for anatomic pathology. Recently, however, I received an email from Orchard promoting what the company describes as a single system that promotes integrated diagnostics  (Put all your results in a single basket). Below are the key bullet points from that email:

• Orchard Pathology is an all-inclusive CP/AP System that simplifies integrated diagnostics. 
• To meet the unique demands and ever-changing trends in pathology, we offer Orchard Pathology-the all-inclusive CP/AP pathology information system designed to handle the complex workflow and reporting of integrated diagnostics.
  
• Orchard Pathology was designed to handle clinical, molecular, cytology, and anatomic pathology testing and reporting.

Needless to say, I am pleased that Orchard has seen fit to use the term "integrated diagnostics" in their advertising materials. This suggests to me that the term is gaining some credibility in the market. However and since the term was coined roughly five years ago, the field of surgical pathology has been further fractionated by the emergence of digital pathology. The digital pathology vendors arose separately from the LIS vendors and with a special focus on scanner technology, massive data storage, and algorithms for the interpretation and quantification of special stains in surgical pathology. From the perspective of the trajectory for the integration of LISs with AP-LISs, we now then face another barrier -- the integration of digital pathology with AP-LISs.

Orchard® Pathology is a comprehensive pathology information system designed to meet the current and future trends in pathology and handle the complexities of clinical, molecular, and pathology testing and reporting. Orchard Pathology can be used as a stand-alone system for anatomic pathology testing or as a complete system that includes clinical and molecular testing. The shared database used by these various diagnostic disciplines provides access to the patient’s entire history and enables the consolidation of clinical results to the cytology or pathology report.

Image management tools make it easy to link digital images, annotated diagrams, and scanned documents to case worksheets and incorporate them onto patient reports. Information on each report is stored in discrete data fields that enhance EMR integration and simplify the process to mine data for evaluations, correlation studies, quality assurance, regulatory, statistical, and many other management reports.

A Different Paradigm for Analyzing the Competition between Cerner and Epic

In a recent note, I discussed the competition between the Cerner and Epic EMRs and quoted another blog, Chilmark Research, to the effect that Cerner was moving in new strategic direction, emphasizing a network of services to provide communities of care (see: Cerner Fights Back in the EMR Market: A Community Network of Services + PHR). Such an approach is sometimes referred to as the development of a care coordination platform. I concluded that Epic's approach closely coincided with the business model currently favored by hospital executives and would probably continue to succeed in the market. Vince Kuraitis responded to my note with a comment that included a link to a lecture he had recently delivered titled Platform Wars (see: Platform Wars). The presentation can also be accessed at Vince's blog (see: e-CareManagement blog).

Vince's lecture presents a different paradigm for framing the discussion about the Cerner versus Epic. In slide #31 of his presentation, he suggests that there are three types of EMR platforms: closed systems, walled gardens, and open platforms and apps. In his slide #35, he shows Epic as currently offering a closed system with the Cerner EMR, and other competing vendors, providing a walled garden. He further suggests that Cerner is now shifting to become a more open, care coordination platform. Given Vince's perspective about EMRs, a key question about hospital CEOs remains: To what extent are they interested and willing to move to an open, community-oriented, care coordination EMR platform from the closed platform Epic EMR? I hasten to add that I personally would favor an open approach because is would provide a richer functionality and would introduce more competition in the EMR market.  It would also provide, in the long run, better care and more care choices to patients.

Here is a list of the issues that I think will be dominating the minds of most hospital executives regarding their choice of an EMR as we go forward:

• They want a reliable system with maximum functionality and minimal chance of failure; they will place great weight on the EMR choice of their peers in other hospitals.
• They will listen respectfully to recommendations from consultants, Medicare officials, and payers about the need for a "community network of healthcare services" and "coordination of community care."  However, they will not act on such recommendations until such time that there is a strong and irreversible link between such recommendations and reimbursement.
• If and when such a strong and irreversible link to reimbursement is established, they will turn first to the vendors of their "tried and true" legacy system vendors to provide them with the required, specified new functionality.
• In short, Cerner needs to first staunch the flow of its current EMR customers to Epic. Second, it needs to convince its (now stable) installed customer base that its proposed, and more open system, will benefit the hospital in the long run and support a robust business model.

EMR-Based Order Sets as a Locus of Control of Hospital-Based Physicians

In a recent note, I discussed the evidence-based physician order sets that were being deployed at the University of Kansas Hospital (see: Evidence-Based Order Sets Deployed at the University of Kansas Hospital). At the end of the piece, I commented on the very large number of them and speculated that many of them would probably never be used as a practical matter. Here is the specific quote:

Here's what I don't understand about the deployment of standardized order sets. Why develop 240 when most physicians will use only a small number of them? In the current case, I am sure that Zynx Health wants to prove the worth of its product to the University of Kansas Hospital and it provides bragging rights to have 240 of them. Even differentiating them by type of patient, my guess is that perhaps 20% of them will get used 80% of the time. In other words, the most useful of them will get used the most. The rest of them will lie dormant and forgotten but won't disturb anyone.

Shortly after this note was posted, I had a conversation with a new intern in medicine in a prominent teaching hospital who was field-testing a new order set designed for patients with delerium tremens (DTs). She pointed out to me, first, that many of the order sets she was accustomed to had embedded in them various types of complex calculations. Their presence in the order set assisted the house officers in arriving at the correct diagnosis and ordering the most appropriate lab tests, imaging studies, and drugs for a patient with that particular diagnosis. Fair enough. This made perfect sense to me -- embed rules into the order sets to ensure that the physician orders generated for a patient with the admitting diagnosis of the DTs, in fact, had such a diagnosis. Here's a quote from an article in the Wikipedia about CPOE (computerized physician order entry) and drug errors (see: Computerized physician order entry):

Prescribing errors are the largest identified source of preventable hospital medical error. A 2006 report by the Institute of Medicine estimated that a hospitalized patient is exposed to a medication error each day of his or her stay. Studies of computerized physician order entry (CPOE) has yielded evidence that suggests the medication error rate can be reduced by 80%, and errors that have potential for serious harm or death for patients can be reduced by 55%,and other studies have also suggested benefits. Further, In 2005, CMS and CDC released a report that showed only 41 percent of prophylactic antibacterials were correctly stopped within 24 hours of completed surgery.

It then dawned on me that CPOE and evidence-based order sets presented an irresistible and unique target for correcting errors in hospital practice. In essence, they provide a locus of control over the physicians "writing" patient orders. All of the energy and work effort of a company like Zynx and a multitude of hospital committees gets channeled onto the shoulders of a single house officer in a teaching hospital who is then required to train on the use of 240 order sets. Is there any chance that this number of computerized order sets will be reduced and simplified? Slim to none. If anything, the number of order sets will continue to increase in hospitals as the various medical specialists study the DTs order set, for example, and become convinced that it needs to be subdivided into more subcategories of the condition.

Here's the essence of the problem: The people who generate the order sets are not usually the ones in the trenches who are required to use them on a daily basis. And here's the dilemma that we will soon be facing: As our evidence-based order sets multiply and become increasingly complex, physicians will have less and less of an understanding of the logic behind the orders generated by the rules. Physician order generation will become a "black box" and physicians will assume that all orders are correct. After all, they are evidence-based. Here's a quote from the above Wikipedia article on errors introduced by CPOE:

CPOE presents several possible dangers by introducing new types of errors. Prescriber and staff inexperience may cause slower entry of orders at first, use more staff time, and is slower than person-to-person communication in an emergency situation. Physician to nurse communication can worsen if each group works alone at their workstations. Automation causes a false sense of security, a misconception that when technology suggests a course of action, errors are avoided.

Cerner Fights Back in the EMR Market: A Community Network of Services + PHRs

In my opinion, Cerner is facing a formidable competitor, Epic, in the high-end, larger hospital EMR space (see: Why Does Epic Keep Hammering Cerner? Mr. HIStalk's Opinion; Is Cerner Modifying Its EMR Business Model?). John Moore who blogs over at Chilmark Research recently posted a long piece suggesting that Cerner is crafting a new strategy in order to compete more effectively with Epic. He suggests, first, that It emphasizes support for "communities of care" (i.e, city, region, state, employer). Cerner also intends to provide a "PHR with an ecosystem of third party apps." This will serve as a replacement for the now departed Google Health product (see: Google Health Calls It Quits; Lessons Learned about PHRs or Not). Here is a link to John's note (see: Stepping in Where Google Health Left Off) and below is an excerpt from it:

Cerner has been in the HIT business now for 31+ years having grown to one of the leading EHR vendors in the market. You’ll usually find their systems ...in large healthcare organizations. This sector of the EHR market is seeing fierce competition as Epic seems to pick up one win after another at the expense of others....While continuing to go head-to-head with Epic, it appears that Cerner has also chosen to take a different tack, adopting a philosophy of: if you can’t beat them straight up, change the rules of the game....[The] company is truly looking to remake itself into one that adopts an open approach to not only sharing information ...but provides a foundational “network of services” to enable “communities of care.” ...Cerner is also looking to land additional multi-stakeholder, HIE contracts with their partner Certify Data Systems....But what may be even more interesting then what they have done in the HIE market, is what Cerner intends to do in the broader consumer market....Cerner has been eating its own “dog food” for the past year using Cerner Health to promote health & wellness among their employees ....Cerner Health will target a number of other health & wellness areas, with programs that include built-in incentives....For some time now, employers and payers have been looking to better manage their populations to lower medical loss ratios (MLRs). Providers will be looking to do the same as they take on a greater share of the risk via new contracts ...and future Accountable Care Organizations (ACOs). Cerner Health intends to serve both employers and provider needs in this regard with “Health Graphs,” a conceptual analytics framework that combines multiple data streams to provide an accurate view of population health at the community level....Cerner Health will go head-to-head with HealthVault by offering a PHP with a published software development kit (SDK) for third party independent software vendors (ISVs) by year-end. This will enable an ecosystem of applications to potentially sit on top of the Cerner Health stack....In addition to releasing the SDK at year-end, Cerner will also open the doors to any and all consumers/patients to store their personal health information (PHI) on the Cerner Health PHP. 

There is no question in my mind that this "new" strategic direction by Cerner reflects the long-held core values of the company. Neal Patterson, the CEO and co-founder of the company, has been pursuing this broad vision - a community approach to healthcare delivery and IT -- for more than 25 years. Needless to say, and as John emphasizes, the company has been extremely active in care and wellness activities for its own employees. From the perspective of public health professionals, much of the medical school academic community, and politicians, this insightful approach will certainly be applauded. However, I think that there is one small fly in the ointment. Can Cerner compete more efficiently with Epic with this shift in strategy? In short, can the product be sold to hospital and health system CEOs? In my opinion, the answer to this question is probably no. Here are the reasons why I hold this view.

Most health system and hospital CEOs tends to stay focused on bottom-line issues. Their goal is to maximize the profits of their largely non-profit organizations. This goal is achieved by ramping-up their inpatient admissions and holding onto these patients for no more than three to five days in the hospital. During this time, the number of imaging procedures and lab tests is maximized and these generate the most profits. Their interest in outpatient visits has always been muted because they are less profitable. They are not enthusiastic about exchanging clinical information with other hospitals because they consider it proprietary to their own institutions. Their interest in preventive medicine is minimal to non-existent. Their interest in personal health records (PHRs) is roughly equivalent to that of most healthcare consumers, which is low. They will jump through any ACO hoop demanded of them to the extent that they want to minimally satisfy the feds and employers, hoping that this trend will soon burn out.

It has been demonstrated by market behavior that the CEOs of most large hospitals will spend tens, if not hundreds of millions of dollars, for an Epic EMR. The do so with the knowledge that it will support their preferred business model. Their expectation is that it will work OK and not torment the majority of physicians and nurses. As I opined in a previous note (see: Some Additional Insights into the Epic Corporate Culture): [The CEOs] only care whether the Epic software they have purchased for millions of dollars can be "slammed-in" and runs "adequately" in their hospitals. Will they spend a comparable amount of money for a community network of services, support for an ACO and PHR, and an ecosystem of applications to sit on top of the Cerner software suite (see: The Feasibility of Using the Epic EMR as a "Platform" to Extend Its Functionality)? In my opinion, not so much. In other words, I think that Epic is now providing to health system and hospital executives roughly what they want and need to pursue their favored business model.