Table 2. Primary and Secondary Outcomes Across All Treatment Groupsa.
| Outcome | Mean (95% CI) | |||||
|---|---|---|---|---|---|---|
| Usual care | Electroacupuncture | Auricular acupuncture | Difference in change from baseline, auricular acupuncture to electroacupuncture |
|||
| Change from baseline | Change from baseline | Difference from usual care in change from baseline | Change from baseline | Difference from usual care in change from baseline | ||
| Primary outcome | ||||||
| BPI pain severity score | ||||||
| Week 12 | −0.48 (−0.85 to −0.10) | −2.39 (−2.66 to −2.12) | −1.92 (−2.43 to −1.40)b | −2.03 (−2.30 to −1.76) | −1.56 (−2.07 to −1.04)b | 0.36 (∞ to 0.665)c |
| Week 24 | NA | −2.19 (−2.46 to −1.92) | NA | −1.99 (−2.27 to −1.72) | NA | 0.20 (∞ to 0.51)d |
| Secondary outcomese | ||||||
| BPI pain interference score | ||||||
| Week 12 | −0.86 (−1.28 to −0.45) | −2.76 (−3.06 to −2.46) | −1.90 (−2.40 to −1.40) | −2.52 (−2.82 to −2.22) | −1.66 (−2.16 to −1.16) | 0.24 (−0.17 to 0.64) |
| Week 24 | NA | −2.47 (−2.77 to −2.17) | NA | −2.47 (−2.78 to −2.16) | NA | 0.00 (0.41 to −0.41) |
| PROMIS-GH physical health score | ||||||
| Week 12 | 0.41 (−0.76 to 1.58) | 4.32 (3.48 to 5.15) | 3.91 (2.49 to 5.32) | 4.50 (3.64 to 5.35) | 4.09 (2.66 to 5.51) | 0.18 (−0.98 to 1.34) |
| Week 24 | NA | 3.60 (2.77 to 4.43) | NA | 4.60 (3.73 to 5.48) | NA | 1.00 (−0.18 to 2.18) |
| PROMIS-GH mental health score | ||||||
| Week 12 | −0.73 (−2.04 to 0.58) | 1.85 (0.92 to 2.78) | 2.58 (0.99 to 4.16) | 2.71 (1.76 to 3.67) | 3.44 (1.84 to 5.04) | 0.87 (−0.44 to 2.17) |
| Week 24 | NA | 1.45 (0.52 to 2.38) | NA | 2.92 (1.94 to 3.89) | NA | 1.47 (0.15 to 2.79) |
| QAQ pain medication score | ||||||
| Week 12 | 0.13 (−0.17 to 0.43) | −0.44 (−0.66 to −0.23) | −0.57 (−0.94 to −0.21) | −0.39 (−0.61 to −0.17) | −0.52 (−0.88 to −0.15) | 0.05 (−0.25 to 0.35) |
| Week 24 | NA | −0.43 (−0.65 to −0.21) | NA | −0.28 (−0.50 to −0.06) | NA | 0.15 (−0.15 to 0.46) |
Abbreviations: BPI, Brief Pain Inventory; PROMIS-GH, Patient-Reported Outcomes Measurement Information System–Global Health; NA, not applicable; QAQ, Quantitative Analgesic Questionnaire.
For each outcome, estimates are derived from a linear mixed model with baseline means constrained to be equal across study arms. The dependent variable vector included the prerandomization baseline (week 0) assessment, as well as all postrandomization assessments at weeks 4, 10, 12, 16, and 24. The independent variables were the randomization stratification variables (accrual site and baseline opioid use), treatment arm, week (categorical), and the arm-by-week interaction. A patient-level random intercept was included in the model to account for the repeated outcome measurements within patients.
Primary end point. Point estimates with 97.5% CIs (adjusted for 2 comparisons) are presented. P < .001 for both comparisons of treatment arms with usual care for differences in BPI pain severity change from baseline to week 12.
Primary end point comparison of noninferiority of auricular acupuncture to electroacupuncture at week 12. Point estimate with 1-sided 95% CI is presented. The noninferiority margin was 0.657. The 1-sided 95% CI contains the noninferiority margin; therefore, we cannot conclude that auricular acupuncture is noninferior to electroacupuncture at the prespecified P < .05 threshold. The significance for the noninferiority test was P = .055.
Secondary end point comparison of noninferiority auricular acupuncture to electroacupuncture at week 24. Point estimate with 1-sided 95% CI is presented. The noninferiority margin was 0.657. The 1-sided 95% CI did not contain the noninferiority margin, supporting the noninferiority of auricular acupuncture to electroacupuncture at week 24 at the P < .05 threshold. The significance for the noninferiority test was P = .007. This test was not prespecified as part of the primary end point comparisons and was conducted post hoc as an exploratory analysis.
Secondary outcomes were not adjusted for multiple comparisons. Results are presented as point estimates with 95% CIs.