Scientific Evidence May Not Be Scientific, or Evidence

A story in yesterday's Wall Street Journal (WSJ) may cause concern to everyone in any workers' compensation system where treatment and impairment are to be based on "peer-reviewed, nationally recognized"scientific evidence.

The WSJ found that an astonishing amount of medical and scientific research publications have been retracted in the past few years, at an alarmingly increasing rate, calling into question both the ethics of the authors and the reliability of any subsequent research or publications based on the flawed studies.

"Just 22 retraction notices appeared in 2001, but 139 in 2006 and 339 last year," the WSJ reports. "Through seven months of this year, there have been 210, according to Thomson Reuters Web of Science, an index of 11,600 peer-reviewed journals world-wide."

For the workers' compensation community, this could be a troubling trend and means that the editors of various guidelines that are used for both treatment approval and indemnity considerations will need to more carefully vet the medical evidence that is proffered in support of any particular conclusion.

Even more, as states move to either adopt or augment their various guidelines, a question exists as to just how much reliance can be placed on any particular "standard" that may pass through utilization review.

Workers' compensation in the past 10 years has evolved from a system where the injured worker largely had carte blanche as to any and all treatment so long as it was deemed "reasonable and necessary", to more and more states implementing treatment guidelines that purport to provide best practices supported by peer-reviewed scientifically valid evidence.

I support peer-reviewed scientific evidence, particularly in treatment. Standardization is necessary whenever one deals with something as complex and varied as the human body and the body's reaction to injury or illness. While no one is the same, there are standards that are applicable and as a consequence we expect that given condition X, response Y should result if Z is applied.

It is this basic foundation upon which the trust of physicians, and those peripherally tied to the medical practice (e.g. utilization review, case management, etc.) rely upon to ensure "best practices" and outcomes measurement.

If an injured worker can challenge guidelines on the basis that any such evidence is in fact not scientifically valid, that throws medical control options out the window, and can make utilization containment systems nearly ineffective.

I'm not saying that doomsday is here for treatment guidelines or any other similar medical or scientific standard. The majority of guidelines are edited rather conservatively and require high levels of validation before a study is incorporated into the text so they are quite reliable in their application.

But the news that these errors are increasing, and at a dramatic rate, is alarming and should cause the workers' compensation professional to at least question the validity of a recommendation that is based on a study that differs from long held standards.