You’d think that we’d all agree on what back pain is. Pain in the low back is almost as common as days of the week. Most everyone has had or will have back pain in the course of their lives and it is one of the most frequent reasons for seeking healthcare.1 Despite this common, everyday experience, we continue to be vexed by competing understandings of the nature of back pain.2
To be sure, our understanding of the cause of common back pain has changed over the last fifty some odd years. However, it’s questionable whether these changes mark true advances in knowledge.
At one time, we tended to refer to back pain as backache. The term implied similarity to neck ache and headache. Today, when aches of the head and neck occur without any overt injury or illness, we tend to associate them with over-exertion, working too much, tension, and unresolved conflict, among others. We allow, in other words, for the occasions when problems or people are literally “such a headache” or “a pain in the neck.” Still other times, especially when people are prone to neck ache or headache, we tend to perceive such aches and pains as a barometer for when we need to take better care of ourselves because life has become too busy or problematic in some manner.
In the past, backache was considered similarly. The gradual appearance of low back pain, without injury or illness, was readily taken as a sign that life and/or relationships were getting out of sync: backache could be due to working too much, leading to the subsequent need for more rest and relaxation in life; or it was due to the persistent inactivity of a desk job and the habit of watching night-time TV, resulting in an unmet need for more exercise; or it could be due to the persistent unresolved conflict with, say, a boss at work and the mounting pressure that it entailed over time.
Suffering from backache (or neck ache or headache), people tended to prescribe a return to healthier ways of living – resolutions to get more rest and relaxation, work less, seek the solace of the country or outdoors, or bathe in warm springs or spa.
Today, we might understand this conceptualization of back pain with the concept of stress. We use the term stress to refer to the hectic and chaotic nature of modern life. We work too much. We are our chronically over-stimulated from information, messaging, and screens. We persistently juggle between work, children and their activities, and the management of elderly parents. Despite the apparent promise of connection through social media, urban and rural loneliness are all too common. When such imbalances persist for too long, we experience them in any number of emotional and bodily ways, some of which are headache, neck ache and backache.
We all recognize the grain of truth in this way of understanding the occurrence of low back pain, especially when it comes on gradually and without warning. We used to call this type of pain backache.
A competing conceptualization of backache, however, has coexisted alongside this understanding of backache as a malady of the stress of modernity. It is backache as the result of a condition of the spine brought about by injury or degeneration. Allan and Waddell3 have a rich history of the first published medical papers beginning in the late 19th century hypothesizing backache as emanating from problems associated with the spine and subsequent reports of acute back pain due to railway injuries. By the first half of the 20th century, the intervertebral disc had sunsequently come to be understood as a source of backache. With the advent of CT scans in the 1970’s and MRI scans in the 1980’s, this conceptualization eclipsed all other ways of understanding backache, as so much of the spine had become visible. Having become observable, abnormalities of the spine were readily identified as the source of pain in the back. As a result, it is now commonplace to consider pain in the back as due to injury to the disc or degenerative changes to the spine. Indeed, this view of backache has so taken hold that we no longer readily use the term backache in preference for back pain.
There is, of course, a grain of truth in this way of understanding back pain too. Like any other part of the body, we can injure our low back. We are involved in motorcycle accidents, sports injuries, slips on the ice, and falls from ladders, and countless other ways we can injure ourselves. Sometimes, these accidents cause significant enough injuries that they can be seen on scans.
Indeed, beginning in the 1980’s, but really taking off in the 1990’s and early 2000’s, the use of CT and MRI scans became widespread for all instances of back pain. It seemed to open up a whole new world in our understanding of back pain. Specifically, we made the leap from initially holding that back pain is due to spinal abnormalities from injuries on some occasions to holding that back pain is due to spinal abnormalities on all (or most all) occasions.
The primacy of what we might call the injury model for understanding back pain is evident even in those instances in which there is no overt injury associated with the onset of back pain. In response to the gradual appearance of back pain, we tend to look to what we might think of as micro-injuries: “I must have slept wrong,” “I must have tweaked my back,” or “I twisted the wrong way.” Conceptualizing it as an injury, we subsequently seek medical evaluation and oftentimes want a scan to see what’s going on with the spine. Indeed, both the general population and healthcare professionals now seem to simply assume the association of common back pain with spinal abnormalities. Subsequently, it’s now commonplace to want to look first to the vertebrae, discs and ligaments when seeking an explanation of common back pain.
The term for what’s wrong in many of these instances has tended to be spinal- or disc degeneration. It’s a way to refer to the medical equivalent of what we might think of as micro-injuries. Degenerative changes to the spine are the result of slow, wear and tear.
Moreover, such degenerative changes can and have been, of course, the object of medical and surgical intervention. Physical therapists, physiatrists, interventional pain physicians, orthopedic and spine surgeons – all are experts in the evaluation and delivery of interventions aimed at degenerative conditions of the spine.
Having reached its apex in the early 2000’s, this way of conceptualizing back pain as the result of overt or subtle injury, defined by abnormality of the spinal structures, keeps back pain squarely within the realm of healthcare, rather than lifestyle. People have come to look, not to what they can do to alleviate backache, but to healthcare providers with an expertise in the evaluation and treatment of abnormalities of the vertebrae, disc and ligaments. In this way, people with back pain must become patients, it seems, because they have little control over the abnormalities of their vertebrae, discs, and ligaments once they have them. The implication is that people with back pain went from having a problem that they resolved themselves to patients with a medical condition, akin to an injury, that healthcare providers treat.
During this period of shifting emphasis towards spinal abnormalities as the explanation of back pain, the period of the 1990’s to the 2010’s saw the use of scans, interventional procedures, and surgeries grow at exponential rates.4
Looking back, we might question the logic of what happened. Just because some instances of common back pain are due to injury doesn’t mean that all instances of common back pain are the result of injury. In logic, this type of error is called affirming the consequent and it’s considered a fallacy. It is, however, exactly what happened in our understanding of pain in the back.
We were captivated by a model for understanding a particular set of a phenomenon, through which we came to see all instances of the given phenomenon.5 We sometimes think that science and knowledge comes from empirical observation, which then lead us to developing models for understanding those observations. Of course, it can happen this way, but the reverse can also be true. Sometimes, our models determine what empirical observations we make. Science and knowledge can be a reciprocal process in this regard. It’s the latter that occurred from the 1990’s to at least the early 2010’s, but even to some extent to this day. Our injury model of back pain has determined how we perceive the empirical data of back pain.
Our captivation to this model was aided, of course, by the development of high-powered abilities to perceive spinal structures – the CT and MRI machines in the 1970’s and 1980’s, respectively. With them, we could find abnormalities of the spine even when the onset of back pain occurred without any overt precipitating injury – we could still find micro-injuries, if you will, in the form of degenerative changes of the spine. Thus, the injury model of back pain could still explain back pain even when there was no demonstrable outward injury. In other words, we could see the apparent spinal correlates to those instances when we explain to ourselves that we “must have” have injured our backs with a tweak or a wrong twist, even though we hadn't previously observed such a tweak or twist when the pain had come on subtly and progressively.
Despite the power of explanatory models determining what we see, empirical data can still break through in the form of counter-factuals to the model and lead to a change in the conceptual model itself. It takes, however, time, sometimes on the order of years, and it can be aided by changes in society.6
One early instance of counter-factual data, aided by events in society, was that despite the exponential growth of scans and treatments for spinal degeneration, disability claims for back pain grew at a corresponding exponential rate.7
How could this be? If common back pain is due to injury and we have the capabilities to both accurately assess these injuries with the use of scans and treat them with, say, the use of physical therapy and spinal interventions and surgeries, people with back pain should be get better and go back to work at high rates. It’s not what happened, though. With the practice of scanning and treating spinal abnormalities, people have been becoming more disabled. It is hard to escape the possibility that, in at least some instances of common back pain, we have been assessing and treating the wrong things – things that in fact were not the cause of these instances of back pain.
Another set of empirical data that arose in the 1990’s to challenge the injury model of back pain was the discovery that spinal abnormalities are common even among people without back pain. The use of scans, when turned on people without back pain, find the very same things that are found in people with back pain, and which are used to explain back pain.8, 9 As it turns out, disc herniations and degenerative changes are common in people, with or without back pain. Jarvik, et al.,10 and Borenstein, et al.,11 subsequently followed their subjects without back pain over a period of three and seven years, respectively, to see if the presence of such degenerative changes predict later onset of back pain. The progression of degenerative changes over time had no statistical correlation with who later developed back pain.
To place the importance of these findings in perspective, we might use an analogy of the brown-haired bank robber. Suppose that a bank was being robbed one morning and the police had received a tip that the bank robber had brown hair. They surrounded the bank and out came a young man with brown hair. They promptly arrested him and placed him in the back of a squad car, confident that they had found the robber. Upon entering the bank to tell the customers and staff that they were now safe, the police came to realize that many people in the bank had brown hair. Had they found the culprit? They’d have to admit that their confidence could no longer be so certain.
Similarly, having become able to find spinal abnormalities with the occurrence of back pain, we had become confident that back was due to these spinal abnormalities. However, once we came to recognize that we find such abnormalities commonly in people without back pain, we can't and shouldn't so certain that we have found the culprit of back pain.
At the time of the publication of these findings in the 1990’s and early 2000’s, these findings were indeed puzzling, and made a splash among healthcare providers. Nonetheless, they weren’t sufficient to alter the model through which we had come to understand and perceive back pain. We continued, as we still do today, to understand back pain as the result of injury or degeneration of the spine.
I recall occasions in the 2000’s when patients reported to me, “I never knew how much pain I was in until my surgeon read me my MRI results.” Still, to this day, I have patients who report to me that their interventional pain physician or surgeon, when reading their MRI results, exclaim in wonder as to how they can even walk with a spine like they have. Rather than privileging the facts, in this case that the patient doesn’t report much pain or can readily walk, and thus interpret the scans as not representative of their pain and abilities, they privileged the findings of the scans and then told the patients that they are actually worse than they are. Such is the power of the injury model to captivate and determine our understanding of the phenomenon in front of us.
If, however, we are able to step out of the injury model for understanding back pain, we might not wonder such things. We would understand better why people don’t have a lot of pain or difficulty walking, even when finding the presence of spinal abnormalities on scans. CT and MRI scans commonly reveal such findings in people with or without back pain.
Moreover, we might come to understand that these findings have little or no relationship to pain levels or functional abilities. We might, in other words, realize that we are pointing to the wrong things when attempting to explain many instances of common back pain.
This conclusion bears out in large-scale studies that fail to find either statistical or clinically meaningful relationships between back pain and spinal degenerative changes.12, 13 It’s hard to make a case that we have found the cause of back pain when the purported cause doesn’t even correlate, or correlates poorly, with back pain. The most charitable thing we can say is that the lion’s share of what accounts for back pain isn’t captured by scans. Back pain is likely due to many different things that when put together adds up to pain in a person’s back and that what’s found on a scan is just one small factor that accounts for any given person’s back pain.
While we can, of course, injure our back, as we can any other location of our body, should we then infer that all back pain is due to injury? When we so commonly find spinal abnormalities upon scanning the backs of people who report an injury, it is tempting to think that spinal abnormalities are the objective correlate to the reported injury. Similarly, it is tempting to think that such spinal abnormalities are the cause of back pain when it occurs gradually, without a precipitating injury, as we can find spinal abnormalities in these cases too. Thus, it can come to seem that all back pain is the result of spinal abnormalities, whether as the result of injury or more subtle degeneration of the spine.
This explanation of back pain can seem so compelling until, at least, we come to find similar levels of spinal abnormalities in people without back pain. We now have about three decades of repetitive studies coming to the same finding. Most of these studies find no statistically significant relationship between back pain and spinal abnormalities. Even in the minority of studies that do show statistical significance, the relationship is poor and of questionable clinical significance. At best, spinal abnormalities play only a minor role in the cause of back pain.
This fact might shed some light on why spine surgery, despite its widespread use, has failed to produce any conclusive empirical studies showing its effectiveness. The largest and longest study to date, the SPORT trial for lumbar disc herniation, has shown no considerable difference between those who have surgery for disc herniation and those who do not.14, 15, 16
This study was originally designed as a longitudinal, clinical trial, meaning that subjects, who all had disc herniations in their low back, were randomly assigned to either receive surgery or conservative care without surgery (the clinical trial aspect of the study) and then followed to obtain data on their status over the course of 1, 2, 4, and 8 years (the longitudinal aspect). When the data was compared for the two groups, strictly defined by those who remained in their randomized groupings over the course of the study (which is how a clinical trials are run, for instance, for testing the effectiveness of a medication), outcomes were no different between those who received surgery and those who did not receive surgery, at any of the follow-up periods from 1 to 8 years.
If this clinical trial tested the effectiveness of a medication, or of a psychological treatment, no one would question the conclusion that the treatment failed to work. It was no more effective than usual care, which is to say that adding surgery to the overall treatment provided no value. Surgery for disc herniation is no more effective than not having surgery.
However, a funny thing happens when we are captured by an explanatory model of a particular phenomenon and are then met with counterfactual data. Counterfactual data are hard to believe. It’s no truer when it comes to the specific situation of a widespread treatment for spinal abnormalities like surgery. We all collectively might want to ask: how can this be that surgery for disc herniation doesn’t work? Surely, it can’t be true.
As it turns out, the study had some problems, which ironically leads to a reframing of the data towards an alignment with the explanatory model that back pain is due to spinal abnormalities. Specifically, the study suffered from a lot of crossover of randomized subjects. Crossover occurs when subjects of a clinical trial fail to stay in their randomized groupings and so don’t end up getting the treatment to which they were randomized to get. So, in the instance of the SPORT trial, some subjects randomized to get surgery subsequently decided against getting surgery after all; and some of the subjects randomized to conservative care without surgery eventually decided to get surgery anyway. The investigators did the best they could given this circumstance and compiled the data in accordance to the re-assorted groupings and compared those who received surgery against those who did not receive surgery.
Because the subjects in these newly sorted groupings were not fully randomized anymore, this lack of randomization introduces additional variables that might influence the effectiveness of the procedure, outside of the actual procedure itself. There might be something unique to the individuals who decide for or against a certain treatment that in turn influences the effectiveness of the treatment, something that might have been more evenly distributed and therefore washed out, if they had stuck to the initial randomization of who got which treatment. We know, for instance, that motivation and belief in the effectiveness or lack of effectiveness of a treatment can influence the effectiveness of a treatment. It's feasible that such motivation and belief may have played a role in the degree of crossover from one randomized treatment to another, and, as such, they may have also played a role in the degree of effectiveness that those treatments exhibited in the study.
Nonetheless, the data derived from the re-sorting of the groupings might shed some light on the effectiveness of the procedure, even if it is not as rigorous of a study as a clinical trial. The name for this type of study that the investigator’s secondarily pursued, given the circumstances, is called an observational study. The results of an observational study are not considered as conclusive as the results of a clinical trial.
What they found was that, at the 4-year and 8-year follow-up periods, both the group that received surgery and the group that did not receive the surgery improved, but that there was a statistically significant difference in favor of the surgery. In other words, those who had received the surgery improved a little bit more than those who id not receive surgery. This small improvement amounted to about a 13% greater reduction in pain for those who received surgery, which is roughly the equivalent of a one-point greater reduction on the commonly used 0 to 10 scale. There was no difference in the rate of returning to work between those who received surgery and those who didn’t.
The most charitable interpretation of these longitudinal findings of both the randomized clinical trial aspects of the study and the observational aspects of the study are that surgery for a herniated disc might show a small reduction in pain four years and eight years down the road, when compared to not getting surgery. Again, both the surgery group and the non-surgery group showed improvements, but the surgery group improved just a little more than those who did not receive surgery.
Nonetheless, this small improvement is of questionable clinical significance – is the difference, say, of having a pain level of a 7 rather than an 8, four to eight years down the road, enough of a difference to really make a difference in the actual experiences of one’s life?
Thus, the SPORT Trial, which is the best study to date of surgery for disc herniation is far from conclusive. It points to the conclusion that surgery for disc herniation might produce a small improvement in pain level over time.
This small effect for pain reduction but not work status improvement falls in line with what we have been discussing – spinal abnormalities might play a causal role in back pain, but if they do, it is a small role, and so targeting them surgically produces only small improvement. Again, all of this is to suggest that the lion’s share of what causes common back pain lies outside of what scans can identify and what therapies targeting such abnormalities can effectively treat.
Despite decades of data pointing to a cause of common back pain as more complex than simply the presence of spinal abnormalities, a large portion of society continue to have a default understanding that back pain is due to an injury for which spinal degenerative changes are its correlate.17, 18 This default understanding involves, of course, a subsequent default implication that first-line treatments of back pain should target these degenerative changes.
We continue this default understanding because we are captured by an explanatory model that back pain is due to injury, and spinal degenerative changes are the correlate to injury. We continue to do so at the peril of persistent pain, disability and cost.
What, then, of our alternative understanding of back pain – that of back pain as backache, similar to a headache?
With each time we develop a headache, we don’t automatically think that we must have injured our head, and subsequently seek medical evaluation and a scan to determine the specific nature of the injury that we figure we must have had. We also don’t automatically think that we must seek out therapies that target the presumed injury. Rather, we tend to naturally think of it as due to tension and stress, or working too much, or not getting enough sleep, or just the hectic pace of our modern lifestyle. We subsequently think that we need to take better care of ourselves and engage in some healthy changes to the life we live.
Maybe, it’s time to return to a similar understanding of back pain. We could start by going back to referring to back pain as backache.
For more information, see also:
1. St. Sauver, J. L, Warner, D. O., Yawn, B. P., Jacobson, D. J., McGree, M. E., Pankratz, J. J., Melton, L. J., Roger, V. L., Ebbert, J. O., & Rocca, W. A. (2013). Why do patients visit their doctors? Assessing the most prevalent conditions in a defined US population. Mayo Clinic Proceedings, 88(1), 56-67. doi: 10.1016/j.mayocp.2012.08.020
2. Hartvigsen, J., Hancock, M. J., Kongsted, A., Louw, Q., Ferreira, M. L… Lancet Low Back Pain Series Working Group. (2018). What low back pain is and why we need to pay attention. Lancet, 391(10137), 2356-2367. doi: 10.1016/S0140-6736(18)30480-X
3. Allan, D. B., & Waddell, G. (1989). An historical perspective on low back pain and disability. Acta Orthopaedica Scandinavica, 60(suppl. 234), 1-23. doi: 10.3109/7453678909153916
4. Brooks, M. I., Deyo, R. A., Mirza, S. K., Turner, J. A., Comstock, B. A., Hollingworth, W., & Sullivan S. D. (2008). Expenditures and health status among adults with back and neck problems. Journal of the American Medical Association, 299, 656-664.
5. Wittgenstein, L. (1953). Philosophical Investigations. New York: Macmillan.
6. Kuhn, T. S. (1996). The Structure of Scientific Revolutions (3rd Edition). Chicago: University of Chicago Press.
7. Deyo, R. A., Mirza, S. K., Turner, J. A., & Martin, B. I. (2009). Overtreating back pain: Time to back off? Journal of the American Board of Family Medicine, 22(1), 62-68. doi: 10.3122/jabfm.2009.01.080102
8. Boden, S. D., Davis, D. O., Dina, T. S., Patronas, N. J., & Wiesel, S. W. (1990). Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects: A prospective investigation. Journal of Bone and Joint Surgery: American Volume, 72(3), 403-408.
9. Jensen, M. C., Brant-Zawadzki, M. C., Obuchowski, N., Modic, M. T., Malkasian, D., & Ross, J. S. (1994). Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine, 331, 69-72. doi:10.1056/NEJm199407143310201
10. Jarvik, J. G., Hollingworth, W., Heagerty, P. J., Haynor, D. R., Boyko, E. J., & Deyo, R. A. (2005) Three-year incidence of low back pain in an initially asymptomatic cohort. Spine, 30, 1541-1548.
11. Borenstein, D. G., O’Mara, J. W., Boden, S. D., Lauerman, W. C., Jacobson, A., Platenberg, C., Schellinger, D., & Wiesel, S. W. (2001). The value of magnetic-resonance imaging of the lumbar spine to predict low-back pain in asymptomatic subjects: A seven-year follow-up study. Journal of Bone and Joint Surgery: American Volume, 83(9), 1306-1311. doi: 10/2106/00004623-200109000-00002
12. Brinjikji, W., Luetmer, P. H., Comstock, B., Bresnehan, B. W., Chen, L. E., Deyo, R. A., Halabi, S., Turner, J. A., Alvins, A. L., James. K., Wald, J. T., Kallmes, D. F., & Jarvik, J. G. (2016). Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. American Journal of Neuroradiology, 36(4), 811-816.
13. Corniola, M. V., Stienen, M. N., Joswig, H., Smoll, N. R., Schaller, K., Hildebrandt, G., & Gautschi, O. P. (2016). Correlation of pain, functional impairment, and health-related quality of life with radiological grading scales of lumber degenerative disc disease. Acta Neurochirurgica, 158(3), 499-505.
14. Weinstein, J. N., Tosteson, T. D., Lurie, J. D., Tosteson, A. N., Hanscom, B., Sinner, J. S., Abdu, W. A., Hilibrand, A. S., Boden, S. D., & Deyo, R. A. (2006). Surgical vs. nonoperative treatment for lumbar disk herniation: The Spine Patient Outcomes Research Trial (SPORT): A randomized trial. JAMA, 296(20), 2441-2450. doi: 10.1001/jama.296.20.2441
15. Weinstein, J. N., Lurie, J. D., Tosteson, T. D., Tosteson, A. N., Blood, E., Abdu, W. A., Herkowitz, H., Hilibrand, A. S., Albert, T., & Fischgrung, J. (2008). Surgical versus non-operative treatment for lumbar disk herniation: Four-year results for the Spine Patient Outcomes Research Trial (SPORT). Spine, 33(25), 2789-2800. doi: 10.1097/BRS.0b013e318ed8f4
16. Lurie, J. D., Tosteson, T. D., Tosteson, A. N., Zhao, W., Morgan, T. S., Abdu, W. A., Herkowitz, H. & Weinstein, J. N. (2014). Surgical versus nonoperative treatment for lumbar disk herniation: Eight-year results for the spine patient outcomes research trial. Spine, 39(1), 3-16. doi: 10.1097/BRS.0000000000000088
17. Weber, C., Behbahani, M., Baardsen, R., Lehmberg, J., Meyer, B., & Shiban, E. (2107). Patients’ beliefs about diagnosis and treatment of cervical spondylosis with radiculopathy. Acta Neurochirurgica, 159(12), 2379-2384. doi: 10.1007/s00701-017-3356-0
18. Franz, E. W., Bentley, J. N., Yee, P. S., Chang, K. W., Kendall-Thomas, J., Park, P., & Yang, L. J. Patient misconceptions concerning lumbar spondylosis diagnosis and treatment. Journal of Neurosurgery, Spine, 22(5), 496-502. doi: 10.3171/2014.10.SPINE14537
Date of initital publication: 12-6-2021
Date of last modification: 1-3-2021]]>