OBJECTIVES: The purpose of this study was to compare self-reported pain and "improvement" of patients with symptomatic, magnetic resonance imaging-confirmed, lumbar disk herniations treated with either high-velocity, low-amplitude spinal manipulative therapy (SMT) or nerve root injections (NRI). METHODS: This prospective cohort comparative effectiveness study included 102 age- and sex-matched patients treated with either NRI or SMT. Numerical rating scale (NRS) pain data were collected before treatment. One month after treatment, current NRS pain levels and overall improvement assessed using the Patient Global Impression of Change scale were recorded. The proportion of patients, "improved" or "worse," was calculated for each treatment. Comparison of pretreatment and 1-month NRS scores used the paired t test. Numerical rating scale and NRS change scores for the 2 groups were compared using the unpaired t test. The groups were also compared for "improvement" using the χ(2) test. Odds ratios with 95% confidence intervals were calculated. Average direct procedure costs for each treatment were calculated. RESULTS: No significant differences for self-reported pain or improvement were found between the 2 groups. "Improvement" was reported in 76.5% of SMT patients and in 62.7% of the NRI group. Both groups reported significantly reduced NRS scores at 1 month (P = .0001). Average cost for treatment with SMT was Swiss Francs 533.77 (US \$558.75) and Swiss Francs 697 (US \$729.61) for NRI. CONCLUSIONS: Most SMT and NRI patients with radicular low back pain and magnetic resonance imaging-confirmed disk herniation matching symptomatic presentation reported significant and clinically relevant reduction in self-reported pain level and increased global perception of improvement. There were no significant differences in outcomes between NRI and SMT. When considering direct procedure costs, the average cost of SMT was slightly less expensive.