Answers are grounded in this paper and cite their sources.
Study Design: Retrospective, observational study. Objective: To compare adverse outcomes among Medicare fee-for-service (FFS) beneficiaries with LBP receiving acupuncture versus primary care with prescription drug therapy (PDT). Summary of Background Data: In 2020, the Centers for Medicare and Medicaid Services initiated coverage of acupuncture for chronic low back pain (LBP). Materials and Methods: We analyzed 2020 to 2023 Medicare claims data for noninstitutionalized Medicare FFS beneficiaries enrolled under Parts A, B, and D for ≥365 days prior to and 180 days after the initial visit. We compared rates of outpatient adverse events between two mutually exclusive cohorts of patients with episodes of LBP, assembled according to initial treatment received. The acupuncture (ACU) exposure group was defined by an initial visit for LBP with an acupuncture procedure code (97810, 97811, 97813, 97814, 20560, 20561). The prescription drug therapy (PDT) group was defined by an initial visit for LBP with a primary care physician and an analgesic prescription filled within seven days, without acupuncture. Specifically, six-month outcomes included outpatient adverse drug events (ADEs) and pneumothorax. Inverse propensity weighting balanced beneficiary characteristics. Adjusted logistic and Poisson regression models with robust SEs compared outcomes between groups. Results: The ACU group contained 13,113 beneficiaries and the PDT group 366,817. Pneumothorax risk (RR=0.259, 95% CI: 0.118–0.556, P <0.001) and cumulative adverse event risk (RR=0.346, 95% CI: 0.191–0.628, P <0.001) were significantly lower in the ACU group. Adjusted odds of any adverse event trended lower in the ACU group (OR=0.558, 95% CI: 0.282–1.106, P =0.094), as did ADE risk (RR=0.540, 95% CI: 0.222–1.312, P =0.174). Conclusions: Among Medicare beneficiaries with LBP, the risk of adverse outcomes was lower with acupuncture than with primary care and PDT. However, differences may reflect referral patterns and patient selection rather than treatment effects.
No takes yet. Share an insight, caveat, or question.
Grout et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: