Our Huge Blind-Spot in Analyzing the Quality and Cost of Spine Care

September 29, 2013

I am continually reminded of the ineffectiveness of most clinical spine research and virtually all guidelines that collectively block any improvement in either the quality or cost of spine care.  The latest documentation is this month’s lead article in the BackLetter (“U.S. Spine Care System in a State of Continuing Decline”, Vol. 28, No. 10, October 2013).  Its focus is on the “state of crisis” in spine care within the U.S.

A new study is cited (Mafi et al: Worsening trends in the management and treatment of back pain/JAMA Intern Med/2013) that presents data that this system is actually accelerating the development of chronic pain, work disability, narcotic addiction, use of injections and surgery, and guideline-discordant care in general.  Sky-rocketing costs tied to increased utilization of unproven invasive procedures and devices have generated little if any evidence of any improvement in patient outcomes. The growing use of MRIs generates more and more mis-diagnoses and more opioid prescriptions lead to more addiction. That’s our predicament after more than 1,000 randomized clinical trials and hundreds of millions spent on research and guideline creation over the past 25 years.

What is particularly disturbing about this article is the analysis by experts as to why we are in this situation, and their ideas about solutions. They understandably, but mistakenly, put all their confidence in the recommendations of clinical guidelines.  Their explanation for this crisis is primarily clinicians’ non-compliance with guidelines and recommend that doctors provide far less treatment for both acute and chronic low back pain (LBP) and let the favorable natural history (NH) run its course to recovery for most.  That would make some sense, IF the natural history was truly as favorable as they believe.  I repeat…..if.

Many studies challenge that view of NH (i.e. Donelson et al: Is it time to rethink the typical course of low back pain? PM&R Journal. 2012;p394–401). There are very credible population-based studies that continue to be ignored by those who we would expect to be very familiar with the breadth of this literature. But those same experts seem irreversibly entrenched in the belief that a high percentage of LBP recovers on its own within 90 days.  The data supporting that belief are simply flawed and those population studies, though highly credible, remain unacknowledged.

Meanwhile, this article correctly points out that the exam room reality is that it takes far less time, with the same reimbursement, to order an MRI, write a narcotic prescription, or refer to a specialist than to explain to a patient in pain the optimism supposedly tied to NH and the rationale for not prescribing those interventions. So guideline-concordant care is just not going to happen, unless guidelines change (see below). Further, the lack of validity for any treatment gives clinicians license to prescribe whatever treatment he or she wishes. After all, if the NH is that favorable, even an ineffective, unproven treatment can appear successful to both the clinician and the patient if the patient recovers by NH.  The classic line: don’t believe your treatment was effective just because the patient got better.  Of course, the generous reimbursement for those treatments further incentivizes guideline-discordant care.

But there is an even more fundamental blind spot in these experts’ point-of-view. Throughout this entire BackLetter article, the symptom of “low back pain” is treated as “the diagnosis”. The article’s focus, just like the vast majority of RCTs and literally all guidelines, is entirely on how to treat this non-specific (NS) symptom.  That modus operandi drives the near-universal acceptance of the notion of one-size-fits-all care, or treating every patient with LBP the same, with NS-treatments like advice to keep active and reassurance of likely recovery in time.  Is it not surprising that hundreds of studies of a NS symptom would only yield NS solutions for the average patient? And is everyone truly the same?  Do none of these experts get it that a bell-shape curve and a bimodal curve focused at the two extremes of the population can have the same mean?

Such recommendations also ignore patients’ understandable sense that there is something terribly wrong and terribly painful somewhere in their back.  “Let’s find out what it is” (get an MRI), and ‘I need pain relief!!”  Again, the doctor gets paid the same for the 5 minutes to write a prescription or make a referral as for the 15 minutes required for patient education. And of course compensation is much greater if a treatment is actually provided, especially if it is invasive.  This perverse reimbursement incentive is substantial.  And all this significantly distracts from the fundamental, even essential, need to figure out how to more precisely diagnose and in turn treat these patients.

Incredibly, there is no mention in this article or, in fact, in any guideline, of the importance and value in discovering something more about the underlying pain generator.  In contrast, two different surveys of international LBP researchers spanning the past 15 years both concluded that the #1 research priority is the identification of valid LBP subgroups (Borkan et al/Spine/1998 and Pransky et al/Spine/2011).  Where is that priority in this article?  After all, isn’t that the only way we will ever move spine care toward more individualized, patient-centric, and cost-effective care?

Paul Batalden at Dartmouth has often stated: “Every process is perfectly designed to get the results it gets.” So what is the process that has been so perfectly generating what is acknowledged in this article as this declining spine care system?

As I see it, the research cards are solidly stacked against making progress in improving care.  For example, the highly regarded Levels of Evidence (LOE) hierarchy that has become the basis for spine-related clinical research and evidence-based care ironically de-emphasizes the importance of the most important focus of LBP science: diagnostic subgroup research. The basis for that de-emphasis appears to be the belief that we simply can’t make a good diagnosis in most cases and may never be able to. If our current advanced imaging can’t do it, it’s unlikely we’ll ever be able to identify anything that will uncover the true source of pain.  Of course, theoretical explanations for LBP abound and are offered to most every patient to justify the clinician’s favorite treatment recommendation.

But the LOE construct includes an absurd approach to diagnosis research based on requiring a  “gold standard” in order to produce a credible diagnostic advance.  How can one possibly have a gold standard when 85% have no means of diagnosis?  Those who accept that research construct obviously never expect to see any headway in making a diagnosis in those 85%.

For decades, the evidence-based process in researching spine care has focused on, and been quite content with, seeking the best treatment for the average patient with this NS symptom. But we don’t have guidelines for abdominal pain, elbow pain, or knee pain, obviously because there are subgroups in each of those body regions for which there are subgroup-specific treatments validated with outcomes evidence.  But with LBP, we have entire guidelines focused on a regional symptom!

Guidelines and RCTs make a feeble attempt at subgrouping by focusing on acute vs. chronic LBP with the belief that they are two different clinical entities.  But there is now evidence that many members of those two duration-based subgroups have far more in common than they have differences (Donelson et al: Influence of directional preference on two clinical dichotomies: acute versus chronic pain and axial low back pain versus sciatica. PM&R Journal. 2012;p667-81).

The Quebec Task Force was the first comprehensive LBP literature review back in 1987.  It tried to shed light directly on this blind spot when it wrote that “the diagnosis is the fundamental source of error….. Faced with uncertainty, physicians become inventive.” But it changed nothing. According to this article, no acknowledged progress has been made.  Our inability to make a precise diagnosis not only remains the fundamental need, it’s importance is our greatest oversight that lies at the heart of our decline in spine care.

There are two other critical parts of the spine care diagnostic blind spot. First is the belief that a LBP diagnosis must be an anatomic one. The second is an entire body of literature ignored by guidelines as well as many researchers and experts that focuses on identifying and validating LBP subgroups based on clusters of clinical findings. I’ve written extensively about that here in prior blog postings and in my first book: Rapidly Reversible Low Back Pain.

Despite the policymakers’ and pundits’ concern about the decline of spine care, and despite over 1,000 RCTs, there are numerous reliability and prospective cohort studies, RCTs, and systematic reviews that strongly validate certain diagnostic subgroups.  But these studies, just like the population-based NH studies, are ignored, even though they: 1-define clinically-relevant characteristics of the large majority of acute and chronic LBP as well as those with axial pain or sciatica; and 2-identify predictably effective subgroup-specific treatments. These studies strongly suggest that acute NS-LBP can be reduced from 85% to 20% or less of the LBP population.  Those two factors alone represent a major advance in spine care that, if widely implemented, would quickly reverse the decline in spine care depicted in this article.

So why does this extremely important blind spot even exist? Why the inability or reluctance to even acknowledge these studies or see any connection between the need for a diagnosis and the perceived current spine care crisis? That discussion is for a different time and place.  But how can we possibly improve care in this huge and expensive population that struggles with this NS regional symptom without addressing and researching the topic of diagnostic subgrouping as a top priority?

What is so frustrating is that theses ignored studies provide answers!!!…..important ones.  In future blogs, I’ll write about those research strategies and discuss just how precise a diagnosis must be in order to identify predictably effective patient-specific care.  And it needn’t be an anatomic diagnosis!

I’ve run way too long, but there is much more to discuss.  I’d love to hear from you.  Please be sure to click on “Like” if this was useful to you, or “Leave a Comment” (click on “Comment” below).  Have a great week everyone.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: