Treatment Risks Must Be Put Into Context
In his article, "NSAID: 'Nuff Said on the Issue of Toxicity," Anthony Rosner, PhD, LLD [Hon.], LLC, begins by asking: "What am I missing?"
Of course, it is then incumbent upon the person making that recommendation to provide sufficient information to the patient in order to obtain a truly informed consent for that treatment.
By citing only risk factors in his article, Dr. Rosner is missing the opportunity to provide sufficient information supporting either or neither intervention. It would be illogical and unethical to use information about only the risk of one treatment to support another treatment; each treatment must be evaluated on its own merits and risks, and then put into perspective with other treatments similarly evaluated.
His article does not make it clear why he is comparing NSAIDs to neck manipulation. Is he proposing a comparison of neck manipulation and NSAIDs for the treatment of neck pain, treatment of low back pain, treatment to maintain health, or for some other indication? He erroneously cites data about efficacy of lumbar manipulation for low back pain without citing data for efficacy of cervical manipulation.
It is entirely appropriate to recommend a risky treatment if there are sufficient benefits to justify the risk when comparing to the other available treatments, and comparing to non-treatment. Appropriate recommendation depends also upon the severity of the problem one is treating. Even an effective low-risk treatment may be too risky if the problem is insignificant.
On the other hand, sometimes a severe problem, left untreated, would predictably lead to death, so a high-risk recommendation is appropriate if sufficiently likely to be life-saving. Some patients may use NSAID medications without sufficient reason to justify the risks, and some patients may undergo neck manipulation without sufficient reason to justify the risks.
A 2001 article in The Hamilton Spectator stated that neck manipulations are often performed on patients without neck pain, including patients without any complaints: "If you go to a chiropractor, even with lower back pain, there's a 66 percent chance the chiropractor will want to manipulate your neck."
A survey in the January 2000 JMPT reported 90 percent of all chiropractors believe all patients should be treated with spinal manipulation, even patients with no symptoms, because the chiropractor believes he/she is maintaining the patient's health.
In addition, a risk vs. benefit analysis may suggest there are risk factors in an individual that would guide the selection of a particular treatment in that individual, but might be a different choice for another individual at different risk.
There are certainly some patients who would be vulnerable to severe injury from NSAID treatment, and who should not receive it. There are certainly some patients who would be vulnerable to severe injury from cervical manipulation, and who should not receive it.
Another thing Dr. Rosner is missing is the need to completely and comprehensively review studies of safety and efficacy, instead of cherry-picking studies that support his premise. For example, he cited NEJM for evidence of NSAID risk, but did not cite its estimate of as many as one in 20,000 spinal manipulations causing stroke, a number much greater than the RAND article he quoted.
In "Changing Views of Chiropractic ... and a National Reappraisal of Nontraditional Health Care," RAND researchers could say only that spinal manipulation benefits some people with acute low-back pain. They didn't find enough data from well-designed studies to say anything about chiropractic's value for chronic low-back pain or low-back pain that involves an irritated sciatic nerve; about the complication rate of chiropractic treatment; about the number of manipulations needed to get the maximum response; or about the cost-effectiveness of manipulation compared with other types of conservative care, such as some forms of physical therapy or even home self-care. (Emphasis added)
A systematic review by the Cochrane Collaborative did not find evidence that cervical manipulation was more effective than cervical mobilization for neck pain; and also noted there was evidence for publication bias that limited the reliability of concluding there was benefit.
Dr. Rosner is right to be concerned about the potential for risk of all treatments. However, I believe these risks must be put into context.
Steven L. Zeitzew, MD
West Los Angeles
Editor's note: Dr. Rosner provides a response to Dr. Zeitzew's letter to the editor in our next (May 1) issue.
Dynamic Chiropractic encourages letters to the editor to discuss issues relevant to the profession and/or to respond to a previously published article. Submission is acknowledgment that your letter may be published in a future issue of the publication. Email here to submit your letter; please include your full name, relevant degree(s) obtained, as well as the city and state in which you practice.