Up to 80 percent of all adults will eventually experience back pain, and it is a aleading reason for physician office visits, for hospitalization and suregery, and for work disability.  The annual combined cost of back pain-related medical care and disability compensation may reach $50 billion in the   U.S. Clearly, back pain is one of society's most significant non-lethal medical conditions.  And yet the prevalence of back pain is perhaps matched indegree only by the lingering mystery accompanying it.

Consider the following paradox.  The American economy is increasingly post-industrial, with less heavy labor, more automation and more robotics, and medicine has consistently improved diagnostic imaging of the spine and developed new forms of surgical and non-surgical therap.  But work disability caused by back pain has steadily risen.  Calling a physician a back-pain expert, therefore, is perhaps faint praise-medicine has at best a limited understanding of the condition.  In fact, medicine's reliance on outdated ideas may have actually contributed to the problem.   Old concepts were supported only by weak evidence such as physiological inferences and case reports, rather than by clinical findings from rigorous controlled trials.

The good news is that most back-pain patients will substantially and rapidly recover, even when their pain is severe.  This prognosis holds true regardless of treatment method or even without treatment.  Only a minority of patients with back pain will miss work because of it.  Most patients who do leave work return within six weeks, and only a small percentage never return to their jobs.  (At a given time, about 1 percent of the work force is chronically disabled because of back problems.)   Overall, then, prospects for patients with acute back pain are quite good.   The bad news is that recurrences are common; a majority of patients will experience them.  Fortunately, these recurrences tend to play out much as the original incidents did, and most patients recover agan quickly and spontaneously.

Sources of Pain

Low-back pain is a symptom that may signal various conditions affecting structures in the low back.  Part of the mystery of back pain comes from the diagnostic challenge of determining its cause in a mechanical and biochemical system of multiple parts, all of which are subject to insult.  Injuries to the muscles and ligaments may contribute, as may arthritis in the facet joints or disks.  A herniated (or "slipped") disk, in which the soft inner cushioning material protrudes through the disk's outer rim and irritates an adjacent nerve root, can be the source pain.  Or the culprit might be spinal stenosis, a narrowing of the spinal canal that can cause a pinched nerve; stenosis usually accompanies aging and wear of the disks, the facet joints and the ligaments in the spinal canal. 

Back pain also may be the result of congenital abnormalities of spine.  These odd structures are often asymptomatic but may cause trouble if severe enough.  Diseases of other parts of the anatomy, such as the kidneys, pancreas, aorta or sex organs, can be responsible as well.  Finally, back pain may be a symptom of serious underlying diseases such as cancer, bone infections or rare forms of arthritis.  Fortunately, such critical causes are extremely rare; about 98 percent of back-pain patients suffer from injury, usually temporary, to the muscles, ligaments, bones or disks.

The phsycial complexity of the lower back combines with another vexing reality to hinder diagnosis of the cause of pain: only a weak association exists between symptoms, imaging results, and anatomic or physiological changes.  Under these circumstances, most diagnostic evaluations focus on excluding extreme causes of pain--such as cancer or infection--that can be more precisely identified or on determining whether a patient has a pinched or irritated nerve.  Up to 85 percent of patients with low-back pain are tehn left without a definitive diagnosis, a nuts-and-bolts reason for their pain.  Most patients cannot recall a specific incident that brought on their suffering, and heavy lifting or injuries, though risk factors, do not account for most episodes.  Back pain often just seems to happen, and the medical community, reflecting this vagueness, has by no means reached a consensus as to the causes of garden-variety cases.

Some commonplace back pain is probably related to stress.  A study published in May by Astrid Lampe and her colleagues at the University of Innsbruck revealed connection between stressful life events and occurrences of back pain.  Previous work by Lampe found that patients without a definite physical reason for low-back pain perceived life as more stressful than a control group of back-pain patients who had definite physical damage.  John E. Sarno of the Rusk Institute of Rehabilitation Medicine at New York University Medical Center has concluded that unresolved emotionally charged states produce physical tension that in turn causes pain.  In fact, he asserts that this variety of back pain actually serves to distract patients from the potential distress of confronting their psychoilogical conflicts; Sarno has successfully treated selected patients with psychological counseling.

Simple muscle soreness from physical activity very likely causes some back pain, as does the natural wear and tear on disks and ligaments that creates microtraumas to those structures, especially with age.  Determining the cause of a given individual's pain, however, often remains more art than science.  With spontaneous recovery the rule-once serious disease is eliminated as a factor-pinpointing an exact cause may not even be necessary in most cases.

Diagnostic Challenges

The inadequacy of definitively diagnosing the cause of back pain led my colleague Daniel C. Cherkin of Group Health Cooperative of Puget Sound and my research group at the Universtiy of Washington to conduct a national survey of physicians from different specialties. We offered standardized patient descriptions and asked our subjects how they would manage these hypothetical patients.

Reflecting the uncertainty in the state of the art, recommendations varied enormously. The results can be summmed up by the subtitle of our publication of the survey results:

  "Who you See Is What You Get."  For example, rheumatologists were twice as likely as physicians of other specialties to order laboratory tests in a search for arthritic conditions.   Neurosurgeons were twice as likely to ask for imaging tests that would uncover herniated disks.  And neurologists were three times more inclined to seek the results of electromyograms that would implicate the nerves.  If pateints are confused, they are not alone.

Until recently, doctors relied on spine x-rays, often performing one on every patient with low-back pain.  Various studies have revealed multiple problems with this approach.  First, a 10-year Swedish research effort demonstrated that at least for adults under age 50, x-rays added little of diagnostic value to office examinations, with unexpected findings in only about one of every 2,500 patients x-rayed.

Second, epidemiological research revealed that many conditions of the spine that often received blame for pain were actually unrelated to symptoms.  Large numbers of pain-free people have been x-rayed in preemployment medical exams and for military induction in some countries, and multiple studies determined that mnay spine abnormalities were as common in asymptomatic people as in those with pain.  X-rays can therefore be quite misleading.

Third, low-back x-rays unavoidably involve exposing sex organs to large doses of ionizing radiation, more than 1,000 times greater than that associated with a chest x-ray.   Last, even highly experienced radiologists interpret the same x-rays differently, leading to uncertainty and even inappropriate treatment.  The latest clinical guidelines for evaluating back pain thus recommend that x-rays be limited to specific patients, such as those who have suffered major injuries in a fall or automobile accident.

Medical experts hoped that improved diagnostic imaging instrumentation, such as computed tomographic (CT) scanning and magnetic resonance imaging (MRI), would make possible more precise diagnoses for most back-pain patients.  This promise has been illusory.  One important reason is that, as in the x-ray studies, alarming abnormalities are found in pain-free people.

A 1990 study by Scott D. Boden of the George Washington University Medical Center and his colleagues looked at 67  individuals who said they had never had any back pain or sciatica (leg pain from low-back conditions).  Herniated disks often get cited as the reason for a patient's pain, but MRI found them in one fifth of pain-free study subjects under age 60.  Half that group had a bulging blamed for pain.  Of adults older than 60, more than a third have a herniated disk visible with MRI, nearly 80 pecent have a bulging disk and nearly every one shows some age-related disk degeneration.  Spinal stenosis, rare in younger adults, occcurred in about one fifth of the over-60, pain-free group.  A similar study of 98 pain-free people, published in 1994 by Michael N. Brant-Zawadzki of HOag Memorial Hospital in Newport Beach, Calif., and his colleagues revealed that about two thirds had abnormal disks.  Detecting a herniated disk on an imaging test therefore proves only one thing conclusively: the patient has a herniated disk.

These findings suggest that many red herrings confuse imaging interpretation and that at least for some, spine abnormalities are purely coincidental and do not cause pain.   Moreover, even the best imaging tests fail to identify the simple muscle spasm or injured ligament probably responsible for pain in a substantial percentage of back patients.  All this imaging perplexity caused one orthopedic surgeon to remark, "A diagnosis based on MRI in the absence of objective clincial findings may not be the cause of a patient's pain, and an attempt at operative correction could be the first step toward disaster."  In other words, the office examination is at least as impotant as the imaging test, and surgery for patients whose back pain is associated only with abnormal imaging results can be unnecessary if not downright detrimental.  Many physicians now advocate CT scans and MRI only for those patients who are already surgical candidates for other esons.

Complicating the situation still further is the fact that most patient with acute low-back pain simply get better and quickly.  A study comparing treatment outcomes found no differences in functional recovery times among patients who saw chiropractors, family doctors or orthopedic surgeons.  Cost, on the other thand, varied substantially, with family doctors costing least and surgeons most.  The Hippocratic admonition "First, do not harm" may be the most importants counsel with regard to this condition--the favorable natural history of acute low-back pain is hard to beat.

Extended bed rest was once regarded as the standard therapy.  This approach was based on the rationale that some patients experience at least transient relief when lying down, as well as on the physioligcal observation that pressures in the intervertebral disks are lowest when patients are prone.  But a guilty-looking disk may be innocent, and most patients improve naturally.  Nevertheless, recommendations of one to two weeks of strict bed rest were the norm until about 10 years ago.  Bed rests's fall from favour has been almost as dramatic as the reversal in status suffered by that former favorite of primary care, blood-letting.  Extended bed rest is now considered anathema, and resuming normal activities as much as possible may be the best option for patients with acuted back pain.

Watchful Waiting as Treatment

When bed rest was till the standard, my group tested it by comparing seven days of bed rest with just two days.  The results were striking.  After three weeks and three months, there were no differences in pain relief, in days of limited activity, in daily functioning or in satisfaction with care.  The only difference was that, obviously, patients with longer bed rest missed more work.  Severity of a patients's pain, duration of pain, and abnormalities found in the office examination offered no predictive value for how long the patient would be off th ejob.  In fact, data analysis showed that the only factor that predicted the duration of the patient's absence from work was our recomendtion for how long to stay in bed.

Other studies have confirmed and extended these findings.  Four days of bed rest turns out to be no more effective than two days-or even no bed rest at all.  The fear that activity would exacerbate the situation and delay recovery proved to be unfounded. Studies have shown that people who remained active despite acute pain experienced less future chronic pain (defined as pain lasting three months or more) and used fewer health care services than patients who rested and waited for the pain to diminish. (The fact that bed rest is ineffective does not meant that everyone can return to their normal jobs immediately, however.

Some pleple with physically demanding jobs may be unable to go back to their normal work as quickly as people with more sedentary occupations.  Neverhteless, it is often useful to have patients with abck pain return to some form of light work until they have recovered more fully.)

Recent research has also challenged the effectiveness of other types of passive treatment.  For example, several studies concluded that traction for the low back simply does not work.  More controversially, there is growing evidence that transcutaneous electrical nerve stimulation (TENS), which delivers mild electric current to the painful area, has little if any long-term benefit.  Similarly, injections of the facet joints with cortisone like drugs appear to be no more effective than injections with saline solution.

In contrast, there is growing evidence for exercise as an important part of the perevention and treatment of back problems for those suffering from either chronic or acute back pain.  No single exercise best, and effective programs combine aerobics for general fitness with specific training to improve the strength and endurance ofthe back muscles.

An exhaustive review of clinical studies of exercise and back pain found that structured exercise programs prevented recurrences and reduced workd absences in patients with acute pain who regularly took part soon after an episode of back pain had subsided.   The preventive power of exercise was stronger than the effect of education (for example, how to lift) or of abdominal belts that limit spine motion.  Patients experiencing chronic pain also benefited from exercise.  In contrast to acute back-pain sufferers, who did better during a pain episode by resuming normal activities than through exercise, chronic back-pain episode by resun=ming normal activities than through exercise, chronic back-pain patients substantially improved by exercising even with their pain.

The inability of conventional medical practice to "cure" a large percentage of back-pain patients has no doubt led the condition to be a major eason patients seek various forms of alternative treatment, including chiropractic care and acupuncture.   Chiropractic is the most common choice, and evidence accumulates that spinal manipulation may indeed be an effective short-term pain remedy for patients with recent back  problems.  Whether chiropractic or other alternative treatments can impart long-term pain relief remain back pain most likely leads to a belief in whatever treatment is employed and probably accounts for the large number of therapeutic options with passionate advocates.

At the other end of the strategic spectrum is surgery.  Most specialists agree that disk surgery is appropriate only when thre is a combination of definite disk hernia on an imaging test, a corresponding pain syndrome, signs of nerve root irritation and failure to respond to six weeks of nonsurgical treatment.  For patients with these findings, surgery can offer faster pain relief.  Unfortunately, patients who do not meet all these standards also often go under the knife, and there is extensive literature on failed low-back surgery.  Indeed, if the pain is not actually from disk herniation, surgical repair of a disk cannot be expected to end it.

Surgical Interventions

The scapegoating of the herniated disk deserves further reflection.  Herniated diskds are most common in adults betwee ages 30 and 50, and most patients whose pain is actually caused by a disk herniation have leg pain with numbness and tingling as the primary symptom; their back pain is often less severe.  A positive MRI should only support a physical examination that investigates a constellation of effects--such as nerve root irritation, reflex abnormalities and limited sensation, muscle strength and leg mobility--to implicate the disk definitvely as the factor in pain.

Recent studies shwo that even for patients with a herniated disk, spontaneous recovery is the rule.  Studies using repeated MRI revealed that the herniated part of the disk often shrinks natureally over time, and about 90 percent of patients will experience gradual improvement over a period of six weeks. Thus, only aboout 10 percent of patients with a symptomatic disk herniation would appear to require surgery.  And because most back pain is not caused by herniated disks, the actual proportion of back-pain patients who are surgical candidates is only about 2 fpercent.

Herniated disks nonetheless remain the most common reason for backsurgery.  A long-term follow-up study of 280 patients, performed by Henrik Weber of Ullevaal Hospital in Oslo and published in 1983, raises serious questions about the enthusiasm for surgical intervention.  Although patients who had surgery had faster pain relief than did patients treated conservatively, the differences evaporated over time.  At the four- and 10-year follow-ups, the two groups of patients were virtually indistinguishable.   Thus, reasonable people might have preferences for different medical interventions, and there is growing recognition that these preferences should be an important consideration in treatment decisions.

Spinal stenosis is the most common reason for back surgery in those over age 65.   National hospital survey data show stenosis correction to be the most rapidly increasing form of back surgery.  Surgery for herniated disks increased 39 percent between 1979 and 1990; stenosis surgeries increased 343 percent.  Reasosns for this rapid rise are unclear but may simply reflect the ability of the new CT and MRI scans to reveal stenosis.  Unfortunately, the indications for surgery in this condition are even less clear-cut than they are for herniated disks.  As a result, there are enormous variations, even within the U.S., in rates of surgery for spinal stenosis.   For example, by analyzing Medicare claims, my group found approximately 30 stenosis surgeries in Rhode Island for every 100,000 people older than 65 but 132 in Utah.

Surgery for this condition is more complex than simple disk surgery.  Spinal stenosis tends to occur at multiple levels within the spine rather than at a single level, as is usually true for herniated disks.  Furthermore, these patients are older and therefore more susceptible to complications of surgery.  In addition, we know less about the long-term effectiveness of surgical and nonsurgical approaches for treating spinal stenosis than we do about mangement of herniated disks.  Because syptoms of spinal stenosis often remain stable for years at a time, decisions are rarely urgent, and the preferences of the patient should again play an important role.

Classifying as trivial a condition that annually drives millions of Americans to their knees and drains $50 billion from the economy would be a mistake.  A collective shrug at the condition, however, may be the most appropriate, albeit unsatisfying, societal attitude.  Nearly everyone will have back pain, and we should perhaps simply accept it as part of normal life.  Once serious conditions get ruled out, a sufferer is usually best served by simply attempting to cope as well as possible with a condition that will almost certainly improve in days or a few weeks.  The wide vriability in surgical recommendations should make all back-pain experts circumspect, and the patient's wishes should carry considerable wight in treatment choice.

Tfhe mysterious nature and economic ost of back pain are driving a growing interest in research, and the coming years may reveal the fundamental aspects of this problem in more detail.  In the meantime, for most back-pain patients the sterotypical phsycian advisory to "take two aspirin and call me in the morning" comes to mind.  A richer and better course of action might be to take pain relievers as needed, stay in good overall physical condition, keep active thorugh an acute attack if at all possible and monitor the condition for changes over a few days or a week.  Bakc pain's power to inflict misery is great, but that power is usually transient.  In most cases, time andperseverance will carry a patient through to recovery.