Increases in lumbosacral injections in the medicare population: 1994 to 2001

Janna Friedly, Leighton Chan, Richard Deyo

Research output: Contribution to journalArticlepeer-review

266 Scopus citations


STUDY DESIGN. Anecdotal reports and limited data suggest that the use of spinal injections is increasing, despite equivocal evidence about efficacy. OBJECTIVE. We sought to evaluate trends in lumbosacral injection use for low back pain, including the specialties providing the injections and the costs of care. SUMMARY OF BACKGROUND DATA. The current literature reports success rates of 18% to 90% for lumbosacral steroid injections, depending on methodology, outcome measures, patient selection, and technique. Preliminary data suggest that spinal injection rates are rising, despite ambiguity in the literature regarding their clinical effectiveness. METHODS. We used Medicare Physician Part B claims for 1994 through 2001 to examine the use of epidural steroid injections (ESI), facet joint injections, sacroiliac joint injections, and related fluoroscopy. Fee-for-service Medicare enrollees 65 years of age and older were included in this study. We used Current Procedural Technology (CPT) codes to identify the number of procedures performed each year, as well as trends in expenditures, physician specialties involved, and diagnoses assigned. RESULTS. Between 1994 and 2001, there was a 271% increase in lumbar ESIs, from 553 of 100,000 to 2055 of 100,000 patients, and a 231% increase in facet injections from 80 of 100,000 to 264 of 100,000 patients. The total inflation-adjusted reimbursed costs (professional fees only) for lumbosacral injections increased from $24 million to over $175 million. Also, costs per injection doubled, from $115 to $227 per injection. Forty percent of all ESIs were associated with diagnosis codes for sciatica, radiculopathy, or herniated disc, whereas axial low back pain diagnoses accounted for 36%, and spinal stenosis for 23%. CONCLUSION. Lumbosacral injections increased dramatically in the Medicare population from 1994 to 2001. Less than half were performed for sciatica or radiculopathy, where the greatest evidence of benefit is available. These findings suggest a lack of consensus regarding the indications for ESIs and are cause for concern given the large expenditures for these procedures.

Original languageEnglish (US)
Pages (from-to)1754-1760
Number of pages7
Issue number16
StatePublished - Jul 2007
Externally publishedYes


  • Epidural steroid injection
  • Low back pain

ASJC Scopus subject areas

  • Orthopedics and Sports Medicine
  • Clinical Neurology


Dive into the research topics of 'Increases in lumbosacral injections in the medicare population: 1994 to 2001'. Together they form a unique fingerprint.

Cite this