By Michael Rubin, MD
Professor of Clinical Neurology,
Weill Cornell Medical College

Dr. Rubin reports no financial relationships relevant to this field of study.

Synopsis: In an analysis of a large dataset from several large integrated health care systems of patients older than 65 years of age with new-onset low back pain, early spine imaging did not alter management or outcomes, but added considerable cost to their care.

Source: Jarvik JG, et al. Association of early imaging for back pain with clinical outcomes in older adults. JAMA 2015;313:1143-1153.

Back pain is the third most common cause for doctor visits, affects 85% of adults at some point in their lives, costs the United States more than $100 billion annually, and forces almost 50% of sufferers to give up sex as a consequence. Yet, in most instances, episodes are self-limited, resolve without therapy, and, in 85%, remain without a specific cause ever being established. Can anything be done to at least reduce the cost of low back pain? Perhaps limiting early magnetic resonance imaging (MRI) of the lumbar spine in new-onset, low back pain in the elderly can be done. And if so, would this be harmful to the patient?

To compare outcomes among older patients with new-onset low back pain who underwent early MRI of the lumbar spine vs those who did not, a prospective observational cohort was recruited from three integrated health care systems, comprising Harvard Vanguard, Henry Ford Health System, and Kaiser Permanente Northern California, consisting of 5239 patients, 65 years of age or older, who, between March 2011 and March 2013, presented to their primary care physician with new-onset low back pain, defined as having had no visit for back pain in the prior 6 months. Early imaging was defined as undergoing lumbar spine radiographs, computed tomography (CT), or MRI within 6 weeks of this initial visit. Physical limitation due to back pain was the primary outcome measure, as assessed by the Roland-Morris Disability Questionnaire, whereas secondary outcome measures included noting the number of falls with injuries that occurred over the prior 3 weeks, rating average back and leg pain intensity over the prior week on a scale of 0-10, and completing the Brief Pain Inventory interference scale, the Patient Health Questionnaire, and the EuroQol health status measure. Statistical analysis comprised the McNemar tests for categorical variables, paired t-tests for continuous variables, and linear mixed-effects models to obtain adjusted differences between groups.

Among the 5239 participants, 386 were excluded from analysis due to premature withdrawal, unavailability of data, cancer visit, or lumbar spine surgery within the prior year; bone scan within 6 weeks; or death within a year. Of four patients who died of cancer within the year, two had early imaging and two did not. Among the 4853 remaining who underwent propensity score matching, 1174 had early radiographs, 349 had CT or MRI, 1353 served as matched controls, and 1977 were not in a matched set. None of the groups differed significantly on the primary outcome measure, the Roland-Morris disability questionnaire. Of patient-reported outcomes, only leg pain intensity was significantly lower at months 3, 6, and 12 among those who underwent early imaging compared to those who did not. However, this difference, only 0.5 points on the 0-10 point pain numerical rating scale, was clinically unimportant, given that 2-3 points represents a clinically important difference. In contrast, costs were 40% and 50% higher among the radiograph and CT/MRI groups, as measured by mean total relative value units (RVUs), translating into approximately $1380 and $1430, respectively, per patient. Cancer rates were comparable across all groups in the year following the study.

Clinical outcomes do not appear to be altered in elderly patients who undergo early imaging for new-onset low back pain, and, although not a randomized, double-blind, controlled study, circumspection with respect to early imaging may be advised in treating such patients.


Even brief exposure to a number of modifiable factors, both physical and psychosocial, can increase the risk of a new episode of low back pain.1 Using a case-crossover design, 1639 patients were screened, of which 999 patients were included, with new onset low back pain, between October 2011 and November 2012, from 300 primary care clinics in Sydney, Australia. Moderate or vigorous physical activity, or being distracted during an activity or task, increased the risk of new onset LBP by an odds ratio of 2.7 and 25.0, respectively, with risk being highest between 7 a.m. and noon. Alcohol consumption and sexual activity were not associated with new onset low back pain. New approaches for the prevention of LBP should address these findings.


  1. Steffens D, et al. What triggers an episode of acute low back pain? A case-crossover study. Arthritis Care Res 2015;67;403-410.