David Cundiff, MD: Bias Issues Inherent in Bravewell’s Focus on Evidence-Based Medicine in Reform
John Weeks
Summary:In this response to the Bravewell Collaboration’s listing of evidence-based medicine (EBM) and reimbursement reform as key factors in any healthcare transformation, David Cundiff, MD points out that “randomized controlled clinical trials and other tools of EBM are mostly funded by special interests and interpreted by researchers paid by those special interests.” Thus “the evaluation of EBM trials is and will remain controversial.” Cundiff, author Money Driven Medicine, argues that one-third of the HEDIS measures which shape insurer decisions on what get covered, are “highly questionable.” Sponsors of these measures, Cundiff notes, include a who’s who of major pharmaceutical firms. Can one even imagine a level playing field for an evidence-based inclusion of integrative, natural therapeutic approaches?
In this column, Cundiff, author of Money Driven Medicine, focuses on two of the eight key factors in health reform which were listed by the Bravewell Collaborative following the recent Bravewell-funded Institute of Medicine Summit on Integrative Medicine and the Health of the Public. (For the eight Bravewell factors, see Forum on the IOM: Bravewell’s Post-Summit Statement on “Key Factors” in Any Health Reform Plan.) The two are use of EBM and the needed for reimbursement reform. Cundiff focusing on significant issues related to the value of the HEDIS outcomes which shape behavior of 90% of the nation’s healthplans. He suggests that up to a third of their key indicators – the evidence upon which coverage of humans are shaped – “are highly questionable,” providing evidence to back his assertions. This is stimulating, if despair-inducing, reading.
_____________________________ Challenges in Bravewell’s Focus on Evidence-Based Medicine and Reimbursement as Reform Factors
David Cundiff, MD
Of the Bravewell Collaborative’s eight
“key factors that should be included in health reform,” six would
probably appeal to the vast majority of stakeholders. However, the two factors
that relate to allocation of resources present major challenges to finding a
practical way forward with health care reform:
(1).
Evidence-based medicine (EBM) is the only acceptable standard.
Researchers and practitioners alike concurred that health care should be
supported by evidence and urged further research and testing to expand the
evidence base for integrative models of care.
(2). The
reimbursement system must be changed to a system that rewards outcomes rather
than procedures with changes that incentivize physicians to focus on the health
outcomes of their patients.
Let me
address a major problem with basing medical resource allocation on EBM.
Sponsor of the HEDIS measures
Randomized
controlled clinical trials and other tools of EBM are mostly funded by special interests
and interpreted by researchers paid by those special interests. For that and
other reasons, the evaluation of EBM trials is and will remain CONTROVERSIAL.
Therefore, health care reform should not be predicated on being able to find
the right, all-knowing, unbiased experts to draw up government-endorsed
guidelines determining what interventions should and should not be covered by
insurance.
For example, the National Committee for Quality Assurance (NCQA), a private, non-profit organization
markets “Health Plan
Employer Data and Information Set” (HEDIS), a tool used by
more than 90% of America’s health plans to measure performance on important
dimensions of care and service. Employers, consultants and consumers use HEDIS
data to help them select health plans. HEDIS data also are the centerpiece of
most health plan “report cards” that appear in national magazines and
local newspapers. Corporate financial sponsors of HEDIS include the following
drug companies: Boehringer Ingelheim Pharmaceuticals, Inc., Merck & Co.,
Inc., McNeil Pediatrics, Division of Ortho-McNeil-Janssen Pharmaceuticals,
Inc., Pfizer Inc., Daiichi Sankyo, Inc., Sanofi Pasteur, Abbott, Lilly, Bristol-Meyers
Squibb, and Novo Nordisk.
Of
the clinical indicators by which physician organizations and health plans are
rated by HEDIS, many are based on good evidence in the peer reviewed medical
literature or based on common sense such that randomized trials would be
unethical. However, the following guidelines, about one-third of the HEDIS
quality measures, are highly questionable:
Immediate beta-blocker pills for heart
attack patients: In 1999, a HEDIS
report extolled the value of this quality measure: “One shining example is
use of beta blockers after a heart attack. Since 1996, plans’ combined
performance on this measure has jumped from 62.5 percent to almost 80 percent.
Considering the amount of attention given this measure in the last two years,
it could be argued that HEDIS has done more to elevate awareness of
beta-blocker use as a proven and relatively inexpensive way for physicians and
health plans to save lives than did all of the relevant clinical guidelines
that preceded it.” However, in 2007, HEDIS quietly
withdrew this quality measure saying widespread achievement of near maximum
use of beta-blockers in post-MI patients and little variance among health plans
resulted in it having little value as a quality measure to differentiate health
plans. Then,
in December 2008, the American Heart Association and American College of
Cardiology withdrew this guideline with a euphemism indicating that the
guideline was never valid: Guideline deleted because of “Increased complexity
of decision making and controversy about the magnitude of net benefit.” A
New York Times article indicated that the Centers for Medicare and Medicaid
Services (CMS) sent hospital administrators around the country received “a
short, unceremonious e-mail” withdrawing this quality indicator a month later.
However, a search of the CMS
website for “beta-blocker and heart attack” is negative.
This uncontrolled vaccination experiment may result
in an epidemic of adults getting severe
chicken pox in 10 – 20 years.
Vaccination
against chicken pox (varicella) for children: Immunity from varicella
vaccination lasts no more than 5 years compared with lifetime immunity from
having chicken pox (usually a mild uncomplicated early childhood illness). The
older the age of acquiring chicken pox the worse the complications and the
higher the risk of death. This uncontrolled vaccination experiment may result
in an epidemic of adults getting severe chicken pox in 10 – 20 years.
Treatment of depression with
antidepressant medication: The efficacy of selective serotonin uptake and
release drugs (SSRIs, the most widely prescribed antidepressants) is
not more than placebo. They
clearly increase suicidal attempts and may cause some suicides. Non drug
treatments like aerobic exercise, stress management techniques (yoga,
meditation, etc.), communing with nature, and gardening are not mentioned in
the HEDIS guidelines.
Breast cancer mammography screening:
Controversy abounds in interpreting mammogram screening trials. The most
rigorous trial of all, a Canadian study, found no reduction of breast cancer
deaths in premenopausal
or postmenopausal
cardiovascular events by > 50% in 5
years compared with < 30% with pills.
Blood cholesterol screening and drug
treatment to lower low density lipoprotein (LDL) cholesterol: A trial of
treatment of hypercholesterolemia with Mevacor (a statin drug) in subjects
eating the American Heart Association diet compared with a “portfolio diet”
(i.e., fruits, vegetables, almonds, soy protein foods, oats, barley, etc) alone
showed no
significant difference in lowering of LDL cholesterol. The “Dr. Dean Ornish
Program to Reverse Heart Disease” including a similar diet, reduces adverse
cardiovascular events by > 50% in 5 years compared with < 30% with pills.
Treating stage 1 high blood pressure
(systolic BP 140 – 159 or diastolic BP 90 – 99) with drugs: Published trials
have not shown that drug treatment of mild hypertension saves lives or reduces
adverse events. The widely reported ALLHAT trial
comparing 4 blood pressure lowering drugs for stage 1 hypertension showed that
subjects taking water pills (thiazide diuretics) had fewer adverse clinical
outcomes than those on the newer patented meds. However, none of the 15
placebo-controlled randomized trials of thiazide diuretics involved exclusively
stage 1 patients. Five of those 15 trials showed no benefit with thiazides, and The Oslo trial,
combining stage 1 and 2 hypertension subjects, reported a significant INCREASE
in deaths with thiazides at 10 years (14 versus 3).
For type 2
patients with no symptoms of diabetes, a more aggressive diet and
exercise
prescription would be a reasonable alternative to drugs.
Treating people with type 2
diabetes who have no symptoms (frequent urination, excessive thirst, blurry
vision etc.) with blood sugar lowering drugs: Of at least 15 complications of
diabetes studied in the
largest type 2 diabetes trial, only one was significantly benefited with
drugs. Patients on sugar lowering drugs required fewer laser treatments for
diabetic eye disease. However, drug treatment increased weight and
episodes of hospitalization due to diabetic coma (hypoglycemia). For type 2
patients with no symptoms of diabetes, a more aggressive diet and exercise
prescription would be a reasonable alternative to drugs.
Flu shots for healthy adults 50 – 64
years old: Over
90% of the 36,000 deaths per year from complications of influenza are in the
elderly (> age 65). One-third of people 50 – 64 years of age have chronic
diseases that place them at increased risk for the complications of flu. There
is no evidence that vaccinating healthy people 50 – 64 reduces the morbidity or
mortality from the complications of flu.
Pay for performance (P4P) schemes based
on HEDIS are flawed in many process quality measures and are
As currently practiced based on HEDIS
standards, P4P is bad medicine. Clinical trials and EBM definitely have a place
in health care, but it should be to inform the clinical decisions of individual
physicians about individual patients. Given the inevitable controversy involved
in many clinical decisions, they should not be the basis of government and
academic medicine experts, many with financial conflicts of interest, to
determine a one-size-fits-all system of resource allocation.
David Cundiff, MD
Comment: When members of the integrative practice community think about shifting national policy, we typically focus on making changes via members of Congress or the White House. Yet there are huge additional playgrounds for lobbyists, for “government affairs” and “public affairs” professionals which shape the policy soup. This include entities such as NCQA and its crowning EBM glory, HEDIS. Cundiff underlines the importance of being present in these playgrounds. Corporate and industry lobbyists certainly know their way around these money bars.
“Until we really start talking about WHY the current
system behaves as it does, why it rewards who it rewards and excludes who it
excludes, until we speak plainly about who’s benefiting from the status quo
(and someone is clearly benefiting), we’ll never really get anywhere.”
Well-said. And fully-affirmed by Cundiff’s column. EBM, as practiced, is not what it is cracked up to b
