Volume 63, No. 1 January 2007

LINKAGE: EBM, EMR, P4P

Physicians are not against science, or medicine based on evidence. Nor do they oppose technology, progress, efficiency, or knowledge of patients' history and the medical literature.

The label "evidence-based" is de rigueur and as ubiquitous as "low-carb," note Earl Steinberg and Bryan Luce (Health Affairs 2005;24:80-92) and perhaps as meaningful.

The RTI International University of North Carolina Evidence- based Practice Center (EPC) found that only 19 of 121 different approaches for rating the quality of individual studies met the EPC's own standards for such assessments.

Evidence-based clinical guidelines must link several bodies of evidence, with opinion often filling in the gaps.

According to an expert committee of the Institute of Medicine (IOM), only about 4% of all medical services are backed by strong evidence, and more than half by weak evidence or none. While there is great emphasis on internal validity, there is strikingly little attention to the external validity of a study or body of evidence when applied in practice. In other words, there's a lack of "medicine-based evidence" (ibid.), for all the emphasis on EBM.

Nonetheless, EBM, as clinical decision-making "tools," is to be embedded in the electronic medical record (EMR). To be distinguished from a customized digital personal health record (PHR) that a patient controls, the EMR is standardized, interoperable, and controlled by the System. A critical purpose of the EMR is apparently to collect evidence of "provider" and patient behavior, as well as, purportedly, of outcomes.

Physicians are notoriously slow to adopt change, even when "educated" by its proponents. A stronger financial incentive is needed. EBM is to determine the standards for pay for performance (P4P), and data from the EMR can be used to assess, rank, credential, and pay physicians.

"P4P is a outgrowth of the [EBM] movement," writes Michael Cannon in a Cato Institute Working Paper, to be published in 7 Yale J Health P Law & Ethics.

The Drive for Health Information Technology (HIT)

Although the Bush Administration and prominent Republicans are pushing hard for implementation of HIT, the IOM has been advocating the EMR since at least 1991. The Briefing Book on HIT by the Clinton Taskforce on Health Care Reform spells out the agenda. Savings of $78 billion by 2000 were projected, largely through "elimination of inappropriate care." Standardization and uniformity were needed to compare performance. A "seamless network" was required for sharing data nationwide. Clinicians were expected to change their behavior based on cost data.

Incoming HIT czar Robert Kolodner is former director of informatics with the Veterans Administration and served on the Clinton Taskforce. At a meeting at the Institutes of Health on IT on Sept 19, 1999, he stated: "Like it or not, when IT comes in, there will be a fundamental change in the whole process of care and who is delivering what care."

Big employers, including Wal-Mart, Intel, and British Petroleum, are planning to provide digital records for their employees, and store them in a multi-million-dollar data warehouse. The goal: reduce benefits costs (Wall St J 11/29/06).

Independence Blue Cross and 19 other Blue plans nationwide are joining to form a massive database of health records "to enable employers, consumers and medical professionals to learn better what kind of care works best" (Josh Goldstein, Inquirer 8/7/06).

Barriers to the Adoption of HIT

After the Clinton Taskforce officially disbanded, one member wrote, in June 1993: "I look forward to working with you further during this unique window in history for developing the health information infrastructure."

The window closed; momentum slowed. Now, the biggest obstacle to HIT is probably physicians in solo or small-group practice. One hurdle is that costs must be borne by physicians while most benefits will be realized by others, such as payers.

Only 13% of practices with fewer than 10 doctors had access to IT for prescribing in 2004-2005. Such practices "appear to be at risk of being left behind" (HITS 11/9/06).

According to the Medical Group Management Association, the average cost per physician to buy and implement an EMR system is $32,606. On average, implementation costs exceed vendor estimates by 25%. P4P might bring compliant physicians an extra $10,000 per year in revenue.

If hospitals are permitted to buy IT for their stable of referring physicians, the picture could change.

Physicians also care about a degradation in efficiency one reason why IT systems such as computerized order entry are not always used (iHealthBeat 11/20/06).

Then there's the risk of IT-related litigation: such as diagnostic errors related to inaccurate PHRs or flawed decision- support programs (modernhealthcare.com 9/20/06).

Although 97% of Americans say it's important for doctors to be able to access all their medical records, only 34% believe that an EMR would improve the quality of their care (HITS 12/7/06). They are very concerned about secondary uses of their medical data as to deny employment, insurance, or loans.

Physicians will be paid to accept P4P. A House bill to stave off Medicare fee cuts provides a 1.5% P4P bonus.

To win over the public, the National Committee on Vital and Health Statistics recommends using the NHIN exclusively for patient care for 5 years. Once trust has been established, the system can be expanded to secondary uses, such as "syndromic surveillance," outcomes research, and quality assurance.

Frighten; Build Confidence

The strategy for gaining public acceptance of the NHIN resembles that of Cold War arms controllers. First, persuade the public that the threat is apocalyptic and defense impossible. Then build trust in the adversary that controls the weapons.

The IOM's alleged 100,000 deaths from medical errors, pandemic flu, loss of records in a hurricane, undetected fraud, unaffordable costs all are ways in which paper medical records can supposedly cause death and bankruptcy.

"We have to recognize that the quality of healthcare that we have in this country is poor," said John Lumpkin, senior vice president and director of the Robert Wood Johnson Foundation's healthcare group. "If we are going to reduce errors, working harder is not going to work" (HITS 10/11/06).

We must have data for cost-effectiveness analysis, for "the day may be dawning when the Medicare system will face a severe financial crisis, and the available alternatives will be far worse" (N Engl J Med 2005;353:1516-1522).

The public fears the consequences of data sharing. "NCVHS finds that the one benefit that will win over public support is better health care." Individuals will absorb some risk to privacy for identifiable personal benefit but not for societal gain. "Public participation" in the design of the NHIN will also increase public trust and "level of comfort."

Ownership

The Clinton Taskforce raised the question of ownership of health data. According to a proposed statement of principles from Heritage Foundation WebMemo 1131 by Edmund Haislmaier, submitted to the American Legislative Exchange Council, data generated in a "patient-provider" interaction belongs jointly to both parties. And "if other parties, such as a payer, are involved, then they too have a reasonable claim to ownership of that portion of the information that is generated by their involvement." This would mean that payers had the right to use and disseminate the data.

Healthcare Chernobyl?

In 2001, William Winkenwerder, who was about to become assistant secretary of defense for health affairs, told attendees at an HIT convention that "privacy and security are the Chernobyl that is waiting to happen to the healthcare industry." Through 2005, nearly 18,000 cases of medical identity theft had been reported to the Federal Trade Commission. Organized crime is becoming involved. Damages to hospitals and physician practices could easily total billions of dollars annually (HITS 10/16/06).

The street value of medical data is now $50 per record.

When fraud is discovered, the victim may be denied access to his own records, which have become commingled with others, because of the HIPAA Privacy Rule. Dangerously incorrect medical information, disseminated nationwide, may be virtually impossible to amend (HITS 10/17/06).

JCAHO may soon require hospitals to adopt new patient identification technology to prevent delivery of the wrong medical treatment to a misidentified patient. Even biometric identifiers, however, can be counterfeited or stolen. The Frauhofer Research Institute in Darmstadt, Germany, demonstrated its ability to fool fingerprint, facial scan, and retinal scan technology (DeWeese Report, November 2006).

There Is No Anonymous Information

While marketing firms, researchers, financial institutions, pharmaceutical companies, and others may claim to use de- identified information, industry experts told the National Committee on Vital and Health Statistics (NCVHS) that such information can easily be re-identified, reports Kathryn Serkes.

Secondary uses of data are far more widespread than most realize. Physicians' staff members may be selling "anonymized" information from EMRs for up to $100,000 per practice, without the knowledge of patients or physicians.

The CDC is already collecting information from emergency rooms for biosurveillance.

Security Lacking

According to an August 2006 report of the Government Accountability Office, Medicare computers have numerous vulnerabilities in user identification and authorization, cryptography, system boundary protection, and monitoring of security-related events. These could allow access to personally identifiable information and lead to interruptions in claims processing (www.gao.gov/new.item s/d06750.pdf).

Providers and payers have made little progress in complying with HIPAA security requirements. By October, 56% of providers (up from 50% in January) and 80% of payers claimed to be compliant (HITS 10/17/06).

Introducing New Errors

Introducing new technology into complex systems can create new pathways to adverse drug events. Bar-code medication administration systems reduced some types of errors but degraded nurse-physician coordination, caused nurses to drop activities during heavy workloads, increased prioritization of monitored activities during goal conflicts, and decreased flexibility (J Am Med Inform Assoc 2002;9:540-553).

The EBM Agenda

According to Bernardine Healy, M.D., "the autonomy and authority of the doctor, and the subsequent variability in care, are the problems that EBM wants to cure." EBM "breaks the lock- hold" of the medical profession over how medicine is practiced and compensated. Through a "microfascist" process of "sift and sliver," EBM discards physician judgment and more than 90% of the medical literature (US News 9/11/06).

AAPS Calendar

Feb 9, 2007. Thrive, Not Just Survive V, Dallas, TX

Feb 10, 2007. Winter Board meeting, Dallas, TX

Oct 10-13, 2007. 64th annual meeting, Cherry Hill, NJ.

NPI Update

The unanswered questions on the NPI, posted in the AAPS FAQ on our website, remain unanswered. Please send us copies of communications you receive from hospitals or pharmacies regarding the NPI!

To access the official government FAQs, go to http://questions.cms.hhs.gov, and enter "NPI" in the search field.

Items of special interest include:

ID 2623: "NPIs may also be used to identify...providers on prescriptions, in internal files to link proprietary provider identification numbers and other information, in coordination of benefits between health plans, in patient medical record systems, in program integrity files, and in other ways."

ID 6147: "The NPI will eventually be the standard identifier for e-prescribing under Part D."

ID 8063: "All CMS education on the NPI is also free." A company that offers a $110 manual is not CMS.

ID 7298: If you get an NPI, you need to keep a copy of the notification from the Enumerator as you will need it "when you enroll or make a change to your Medicare enrollment information."

Tip of the Month: The decision in Stowers v. Clinton Cent. School Corp. (Ind. App.) upheld the admissibility of a release form signed by an athlete. The student football player and his mother acknowledged the inherent potential for injuries. This was relevant to the school's defense in the wrongful death action brought by the parents. Medical release of liability forms should be admissible also.

Government Keeps Data Secret

Despite the 1999 Shelby Amendment requiring that all data from federally funded trials be made available to the public through the Freedom of Information Act, no data have actually been released, despite numerous requests. All are considered to fall under exemptions. The secrecy is costing lives.

With great fanfare, the NIH announced guidelines for using high-dose steroids in patients with spinal cord injuries, 6 weeks before the study was published in the New England Journal of Medicine. Physicians were skeptical because they were seeing more deaths in patients treated in this way a concern that was borne out in a subsequent study. But the raw data underlying the new "standard of care" are still under wraps.

FOIA requests have also been denied for data related to the finding of smaller brains in children diagnosed with attention deficit/hyperactivity disorder. Critics questioned whether this might be an effect of drug treatment. Likewise, data withheld by researchers might show that Prozac, the only antidepressant approved by the FDA for use in children, is unsafe (Jeanne Lenz, New Republic 10/30/06).

Non-Consented Research

The FDA's Emergency Consent Rule provides for waiver of consent only if "the human subjects are in a life-threatening situation, available treatments are unproven or unsatisfactory... [and] randomized placebo-controlled investigations [are] necessary to determine the safety and effectiveness of particular interventions" (21 CFR 50.24).

The FDA approved an exception in the case of an experiment with PolyHeme, an artificial blood product despite unanimous agreement that human blood is proven safe and effective. The concept of "community consent" was used, although communities were not informed that in a previous trial conducted on surgery patients, 10 of 81 persons who received PolyHeme suffered heart attacks and died, compared with zero heart attacks in 71 persons receiving blood.

A test focusing on situations in which blood was out of reach for hours wasn't conducted because assembling patients for such a trial would be "too difficult and time consuming."

Linda Gorman asks: "Are they increasing risk for these patients unnecessarily just because the FDA is demanding an RCT and won't accept observational evidence?... It is always so easy to cite the good of the many from the losses of a few. Taking private property is OK as long as tax revenues increase. Now it appears to be OK to risk lives to get data faster."

"Special Duty"

When physicians prescribe pain management, they incur the risk that the patient will misuse the medication, leading to criminal prosecution of the physician.

There is a similar legal theory that is spreading in the area of civil litigation. Dr. Peter McConarty prescribed an oral hypoglycemic medication for a patient. About 45 minutes after discharge from the hospital, the patient lost consciousness while driving (allegedly because of low blood sugar) and hit a motorcyclist. The injured motorcyclist sued the doctor under the theory that the doctor had a "special relationship" with him and a "special duty" not to expose him to unreasonable risk of harm. The lawsuit alleges that the doctor should have warned the patient not to drive. Massachusetts Judge Peter W. Agnes, Jr., recently refused to dismiss the suit.

Three states that currently recognize this "special duty" are California, Michigan, and Delaware. Five states that do not recognize it are Missouri, Iowa, Texas, Kansas, and Florida.

Worker Fired for Smoking Off the Job

Scott Rodrigues has sued The Scotts Co., the lawn-care giant, for terminating him because it found nicotine in his urine. The company has a new policy of not hiring workers who smoke, in order to reduce its benefits costs.

The Supreme Judicial Court has upheld a Massachusetts law that makes smokers ineligible to be hired as police officers or firefighters (Boston Globe 11/30/06).

Florida Pain Patient Loses Appeal

Richard Paey, a pain patient who has a morphine pump prescribed for him in jail, was sentenced to 25 years because of his doctor's testimony that he was forging prescriptions. He received the mandatory minimum sentence for "trafficking" although he had not sold any drugs.

The majority in the District Court of Appeal of Florida, Second District, held that it lacked the authority to mitigate Paey's harsh sentence. Dissenting Judge James Seals wrote: "It is illogical, absurd, cruel, and unusual for the government to put Mr. Paey in prison for twenty-five years for foolishly and desperately pursuing his self-help solution to his medical management problems, and then go to prison only to find out that the prison medical staff is prescribing the same... medication he had sought on the outside but could not legitimately obtain."

Correspondence

Performance Standards for Hospital CEOs? According to the Buffalo News, the median compensation in the U.S. for hospital CEOs is $806,000. And while some doctors are cheering a 1% fee increase as a "great victory," hospital CEO salaries went up 7.2%! Even at Erie County Medical Center, which is $17.5 million in the red, the CEO makes nearly $400,000 per year.

About 75% of hospital CEO compensation packages have "incentives" accounting for 25% of the total. These incentives are probably tied in some manner to cost containment. The reason we are seeing so many sham peer reviews nationwide may be that the CEO's compensation depends on removing expensive doctors from the medical staff.
Lawrence R. Huntoon, M.D., Ph.D., Lake View, NY

Flawed Standards. P4P has been the goal of the left for more than a decade. Groups like the Robert Wood Johnson Foundation funded the Leapfrog Group, staffed the federal agencies, and pushed for federal departments. If an executive agency controls the standards, and payment depends on adherence to them, then the agency controls medical care, with no need for federal legislation.

The standards are based on fundamentally flawed premises. Meta-analyses of randomized controlled trials (RCTs) are considered the most reliable form of information, but there is a substantial body of literature showing that meta-analyses are statistical nonsense. The group average shows little about the effects of treatment in an individual: P4P standards for tight blood sugar control have increased problems with fainting.

The experts pushing P4P have never seen an extension of government control they didn't like. No one knows whether P4P will lead to better health. We have no measure of the value of the medical dollar unless consumers are spending their own money. But getting consumers and doctors out of the business of making medical decisions is the whole point of P4P.
Linda Gorman, Independence Institute, Golden, CO

Where the Money Is. According to Modern Healthcare, JCAHO's 2005 form 990 reveals that it took in more than $106 million while spending out almost $101 million. It paid President Dennis O'Leary total compensation and benefits of $1,972,598, up from $875,231 in 2004. Russell Massaro, executive vice president of accreditation and certification operations, was paid $635,857. JCAHO's publishing division is giving the AMA a run for its money: it took in $40 million, up from $33.9 million in 2004, and paid its CEO $515,285.

It seems there's more money in regulating medicine than in practicing medicine.
Kathryn A. Serkes, Washington, D.C.

Objective Standards. The subjective/objective line is grey. Finding objectivity in P4P is more difficult than locating electrons within an atom.
Steven Bassett, Oak Park, IL

Comparative Fees. In 1986, a woman had a hip replacement. Medicare paid the orthopedic surgeon $4,500. In 2006 she came back for the other hip. The Medicare fee is now $1,200.

The same woman said her golden retriever was also having a hip replacement. The fee was $10,000: $3,000 up front for the prosthesis, $7,000 for the surgeon.
Alieta Eck, M.D., Somerset, NJ

Non-Participation. P4P and other government interventions would collapse if physicians would refuse to participate. Bureaucrats cannot function without physician input. But doctors are afraid of a free market. Some would experience a huge bust if churning ceased. Borderline eye examinations, glaucoma exams, lid surgeries, treatments for macular degeneration many of these would probably stop.
Robert P. Gervais, M.D., Mesa, AZ

Opted Out. [From a letter to Sen. Murkowski] ...I opted out of Medicare when it first became possible to do so in 1998 and have been out ever since. My primary motivation was that Medicare seemed to want to get in the middle of the patient- doctor relationship and interfere with patient care. One example was the permitted schedule of tests and the cross-linking coding to justify them. Exceptions required a great hassle. I concluded Medicare really didn't want physicians to find things because then government would have to pay for them. There was an interminable wait to discuss any issue with Medicare personnel. My staff was unhappy with the myriad regulations, which no one seemed to understand....
Alex Baskous, M.D., M.P.H., Anchorage, AK

Standards. I don't want "standardized" medicine. I want customized medicine. I want each and every one of us to get the best treatment for our unique needs and preferences. The best treatment is determined by a host of factors: genetics, history, environmental exposures, comorbidities, moral values, lifestyle choices, and available resources.
Greg Scandlen, Consumers for Health Care Choices

Evidence. While meta-analysis is conceptually attractive, no evidence supports overriding the treating physician's decisions with an RCT-based guidelines issued by a third party who has never seen the patient. HHS found only nine RCTs on EBM-based compensation: little data support this approach.
Richard Dolinar, M.D., Phoenix, AZ