Junkfood Science: How low is good enough?

February 09, 2008

How low is good enough?

When we’re afraid, it’s natural to want to eliminate anything associated with a health problem. The health benefits possible when an extraordinarily high health indice is reduced doesn’t mean, however, that a lower and lower number will ensure greater and greater protection. “How low can you go?” is a concept most often based on fear, rather than good scientific evidence of benefits that outweigh the risks for us.

This week, the National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health issued a press release telling the public that one arm of a type 2 diabetes clinical trial had been stopped because of safety concerns. The study was the ACCORD trial, which was studying 10,251 adult diabetics looking to see if intensive treatments to lower blood sugars, cholesterol levels and blood pressures to even lower levels than standard care would reduce heart attacks, strokes and cardiovascular deaths. The trial began in January 2001 at 77 clinical centers across the United States and Canada. During the first four years of the trial, 257 in the intensive treatment group have died, compared with 203 within the “standard treatment group” (a difference of 3/1,000 patients). The Data and Safety Monitoring Board, an independent advisory group of multi-disciplinary experts, recommended this treatment arm be stopped.

The “standard treatment” group in this study was aiming for hemoglobin A1C levels of 7 to 7.9%. The intensive group was being treated aggressively to reduce their hemoglobin A1C levels to less than 6%, as seen in most nondiabetics. Achieving this extremely low indice required total preoccupation with monitoring and controlling their blood sugars. As Dr. John Buse, a member of the trial’s steering committee, explained to the New York Times:

“Many were taking four or five shots of insulin a day. Some were using insulin pumps. Some were monitoring their blood sugar seven or eight times a day.” They also took pills to lower their blood sugar, one or two several times a day, in addition to the pills they took for other medical conditions and to lower their blood pressure and cholesterol. The subjects also came to a medical clinic every two months and had frequent telephone conversations with clinic staff.

While this description might seem to be downplaying the importance of the ACCORD’s findings as testing an unusual and extreme intervention, it also leaves a lot of unanswered questions because new guidelines and performance measures are recommending A1C levels be kept below 7%, not below 8% as in the standard treatment group in this trial. The trial has not yet been published for anyone to examine its findings in detail and caution is warranted before overreacting.

Regrettably, concerns that surfaced in the ACCORD trial weren’t entirely surprising. They also highlighted the growing debate over health indice management. There has been a controversy within the medical community on clinical guidelines which have been recommending increasingly lower “ideal” health indices, those of healthy 20-year olds — numbers that, essentially, can only be achieved by older Americans with intense pharmacological medical management. The clinical trial evidence, however, has yet to show that such aggressive treatments offer clinical benefits that outweigh the harms for many, if not most, patients.

Increasingly aggressive clinical guidelines, issued by various professional associations and insurers, would be less worrisome if doctors were: 1). given unbiased, clear and accurate reports of the clinical trial evidence and 2). were free to apply the guidelines using their best clinical judgment and what is best for each individual patient, his/her situation and comorbidities and wishes. But increasingly, payers are assuming managed care roles [See: Welcome Home]; determining clinical practice guidelines, tests and prescriptions that providers are to follow; and incentivizing compliance by tying it to doctors’ payments.

Last April, the Centers for Medicare & Medicaid Services launched a new Pay-4-Performance program — the Physician Quality Reporting Initiative — which pays doctors who comply with 74 clinical practice measures. All toll, the quality measures encourage some 15 different prescriptions for adults over age 50. Health indice management, such as weight, cholesterol, blood pressure, and blood sugar management are especially controversial. Many doctors have objected to these P4P measures, saying they have unintended consequences, especially harmful to high risk patients, minorities, the obese, elderly and poor.

Dr. Leonard M. Pogach, M.D., MBA, at the Department of Veterans Affairs New Jersey Healthcare System, led an examination of the clinical practice guideline of the American Diabetes Association, which targets hemoglobin A1C levels less than 7%. The doctors reported that this arbitrary threshold had been achieved in clinical trials that were unrealistic in real-life clinical practices and not appropriate for certain types of patients, namely older people and those with comorbidities (other health problems). They found that more than one-third of the diabetic patients seen at 144 Veterans medical centers during 1999 and 2000 would never benefit from meeting the lower A1C values. Instead, the aggressive control of their blood sugars increased risks and would be dangerous or even fatal for certain patients.

Dr. Rodney A. Hayward, M.D., also called into question the efficacy of performance measures based on getting all patients’ A1C levels below 7%. Arbitrary number management — which encourage the prescribing of multiple drugs “to reach the magic number” — can have the unintended consequences of removing incentives for doctors to help patients with extremely high blood sugars who could most truly benefit by reducing levels, even if they’re never able to achieve 7%. Meanwhile, he said, doctors are rewarded for focusing on the more attainable modest reductions (say, from 8% to 7%) that will have minor impact on preventing complications while putting those patients at risk for side effects. “Treating numbers, for the sake of guidelines, may conflict with treating patients,” he wrote.

Last fall, we learned “evidence-based medicine” often isn’t, and how various biases in the interpretations and dissemination of clinical trial results can and do distort the conclusions that are reported. The major UKPSDS (United Kingdom Prospective Diabetes Study) trial was the first randomized controlled clinical trial of type 2 diabetes medications and medical care that had been conducted in 25 years. It found that tight glucose control among more than 5,000 diabetics followed for 20 years had reduced surrogate endpoints, but had no effect on diabetes-related or all-cause mortality. Independent reviewers found that the evidence contradicted the positive spins of the findings that doctors were being given, yet were being used to support more aggressive management guidelines. While these guidelines are one-size-fits all, a number of reviewers have noted that the adverse effects put some patients at additional risk.

Readers of JFS, Dr. Roy M. Poses, M.D., and Dr. Robert M. Centor, M.D., know that doctors had been questioning the aggressive management of blood sugars to attain A1C levels below 7%, even before the ACCORD trial results.

Whenever policies, mandates and performance measures, such as aggressive health indice guidelines, diverge from the soundest science or lack a body of supportive evidence, the conflicts of interests behind them become increasingly troubling. As Dr. Poses wrote:

A lot of people make a lot of money from intensive diabetes treatments. Drug-makers sell more drugs and biologics, device-makers sell more diabetes tests, and doctors are paid for more office visits. Pharmaceutical, biotechnology, and device companies have marketed diabetes tests and treatments with great enthusiasm. Also, these companies have fostered conflicts of interest that may have influenced how physicians, academics, and organizational leaders think and act. These companies have paid physicians and academics as consultants, speakers, advisors, etc. These companies have also lavishly contributed to professional societies and not-for-profit organizations that are positioned to influence diabetes care... [His full post describes the funding conflicts.]

Dr. Centor echoed the concerns, adding that the studies being used to support these guidelines “have assumed that we treat patients through proxy laboratory measurements. In diabetes, we treat the sugar... the LDL-cholesterol level. In all cases, the common wisdom is wrong.” As he urges:

These studies should give caution to the “bean counters” and the measurement geeks. Caring for patients is much more complex than treating numbers. We have a responsibility to treat patients, and treating patients requires that we consider both the benefits and risk of any intervention.

In our minds, health indices — surrogate measures associated with various health problems — have become actual diseases themselves and the focus of our concerns and fears. Yet, it seems every week the news has been reporting another problem surfacing when symptoms of diseases (health indices) are treated, or treatments are evaluated by changes in surrogate endpoints, that don’t always translate into better clinical outcomes.

The recent examples of clinical trials that used surrogate endpoints with unanticipated clinical outcomes — the ENHANCE trial, the RIO trials, etc. — will hopefully serve as a reminder to healthcare professionals of the importance of randomized, placebo-controlled, double-blind clinical trials measuring actual clinical outcomes; and the need for all of us to examine that evidence and be cautious of assuming guidelines developed by professional trade associations and insurers have accurately interpreted the evidence unincumbered by financial interests.

For consumers, the take home message is not to fear for your health based on health risk indices or sensationalized news stories. It’s important to become informed and active partners with your doctor in your healthcare. The emphasis being trusting “your doctor” who knows you, is looking out for you and has your best interests — not third-party insurance companies or government bureaucrats who don’t.

Bookmark and Share