CHICAGO – Clinical trials are serious business. In addition to being big business (e.g. untold billions of dollars every year), they also involve serious ethical issues and choices.

Most of us are aware of the fundamental ethical principle of informed consent underlying all clinical trials. What is less obvious is that conducting a clinical trial carries with it the responsibility to obtain and share its results (whether positive or negative and even if the clinical trial is a “failure”). It is a failure – indeed a tragedy – when the lessons learned from such studies are not fully appreciated.

Clinical trials involving surgery are especially problematic.

While placebo-controlled trials for pharmaceuticals are accepted from a statistical, practical and ethical standpoint, the same cannot be said for controlled trials for surgery involving sham procedures (e.g. surgeries in which the patient is “opened” but no actual therapeutic intervention takes place). Instead, such trials are often designed to compare surgery with some other, non-surgical treatment.

In this column, I will discuss one of these trials – the recent spine patient outcomes research trial (SPORT) as published in the Journal of the American Medical Association (JAMA) in Nov. 2006 – and share with you from my perspective one of the most important lessons learned.

Two JAMA articles headlined “Surgical vs. Non-Operative Treatment For Lumbar Disk Herniation: The SPORT Observational Cohort” and “SPORT: A Randomized Trial” summarized this important clinical trial. They were also accompanied by two editorials.

The trial results were significant enough to raise the attention of the national media including an article in the New York Times headlined “Study Questions Need to Operate on Disk Injuries”. Indeed, it seems there as many cures for back pain as there are people with back pain.

For anyone with excruciating back pain, it would be an incredible relief to finally have a definitive answer on what works and what doesn’t. There’s one problem, though.

This clinical trial – the largest and most comprehensive of its kind – could not provide a definitive answer (a “yes” or a “no”). Instead, the basic conclusions were that discectomy relieved the back pain more quickly, most people eventually improved in equal fashion and conservative waiting did not seem to cause harm.

Tenuous as it may seem, even this conclusion was significantly muddied by the rather unique trial design. In order to accommodate the challenges of conducting a surgical trial along with the fact that medical opinion was strongly in favor of surgery, the trial was designed so patients could refuse to be randomized to one of the two treatment arms.

Patients could request – both before and during the course of the trial – to cross over to the other treatment. Patients who were initially on the conservative treatment arm could decide that they “couldn’t stand it any more” and elect for surgery. Other patients who were scheduled for surgery could decide they didn’t want to accept the risk and instead went for physical therapy.

Because of the extremely high number of people crossing over (45 percent to 60 percent of patients) and to a lesser extent a certain degree of missing data, “such analyses are significantly confounded and should not be used to compare different strategies,” JAMA contributing editor David Flum wrote in his accompanying editorial.

Instead of answering the question of which treatment was better, the trial seemed to suggest another important message. James Weinstein, a professor of orthopedic surgery at Dartmouth and the lead author of the study, said (regardless of the treatment) that “nobody got worse [and] we never knew that until we did the study”. The New York Times article also quoted Flum:

“Everyone was hoping the study would show which was better. Everyone was surprised by the tremendous number of crossovers in both directions,” [Flum] added. That muddied the data.

However, it would seem that this huge crossover rate also provided another, extremely important lesson: patients not only value a good outcome but they also value choice. The starkly dichotomous nature of the treatments (e.g. surgery versus waiting) and the high number of crossovers clearly send a message that patient choice is absolutely critical and integral to the therapeutic process.

For treatments whose outcomes are roughly equivalent, the availability of options may be much more important for patients than the usual indices of outcomes and morbidity. It is unclear why some patients chose surgery. Did they want to “get it over with” or did they have lower, short-term pain thresholds than those who elected to “wait it out”?

What about those who chose not to go under the knife? Were their pain thresholds higher yet their risk tolerance less? While the study yielded indefinite results, this much is for sure: patients want and certainly act upon choice.

It was disappointing to read further in the JAMA editorials that designing clinical trials with sham-controlled surgeries (e.g. operations in which the patient is cut but the actual restorative procedure is not done) “may be the only effective and ethical next step”. While I can understand the sentiments behind that, the holy grail of scientific rigor does raise some questions.

While evidence-based medicine is certainly a critical component of rigorous and scientific medicine, rigidly applying the concept that one and only one treatment must certainly prevail does a disservice to the diversity of experience, expectations and circumstances that enter into any medical decision.

It is believed that randomized, double-blind clinical trials are the cure against biased medical decisions. However, this blinds one to perhaps the most important bias: the notion that there should be a binary (“yes” or “no”) outcome to any treatment comparison.

While the SPORT trial was in relative terms a limited success with less-than-definitive results, it would be an even greater failure not to recognize the larger lesson learned: whether right or wrong, patients do value choice.

Previous Columns in 2007:
Gov. Blagojevich Announces IllinoisCovered to Insure 1.4 Million in Illinois (3/5/2007)
Medical Design Excellence Awards Offer Decisive Glimpse Into Future of Health Care (2/20/2007)
What’s More Important in Medicine: Diagnostics, Therapeutics or Prognosis? (2/5/2007)
Lance Armstrong and the Future of Cancer Care (1/22/2007)
Subtle But Powerful, Publication Bias Goes Beyond Financial Incentives (1/9/2007)
Click for 2006 column archive.

E-Mail This Article to a Friend or Colleague
View This Article in Printer-Friendly Format