Skip to main content
. 2016 Oct 6;16:131. doi: 10.1186/s12874-016-0224-1

Table 3.

IPD NMA ANCOVA synthesis model results (model 1), EQ-5D preference score and standardised pain endpoints

IPD NMA ANCOVA results: EQ-5D preference score and SMD endpointsa Model 1, ANCOVA, change in outcome score, adjusted for baseline median MCMC posterior sample (95 % CrI)
Change EQ-5D Change standardised pain
Relative treatment effects Osteoarthritis of the knee SHAM vs UC 0.057 (0.013, 0.095) 0.271 (-0.007, 0.537)
ACU vs UC 0.079 (0.042, 0.114) 0.703 (0.399, 0.984)
ACU vs SHAM 0.022 (-0.014, 0.060) 0.438 (0.121, 0.715)
Headache SHAM vs UC 0.052 (0.010, 0.095) 0.332 (0.022, 0.669)
ACU vs UC 0.056 (0.021, 0.092) 0.588 (0.311, 0.869)
ACU vs SHAM 0.004 (-0.035, 0.042) 0.256 (-0.073, 0.560)
Musculoskeletal SHAM vs UC 0.059 (0.017, 0.101) 0.063 (-0.241, 0.378)
ACU vs UC 0.082 (0.047, 0.116) 0.588 (0.334, 0.863)
ACU vs SHAM 0.023 (-0.008, 0.053) 0.527 (0.323, 0.735)
Between-study variance 0.001 (0, 0.003) 0.090 (0.049, 0.170)
Total residual deviance b 15,850 (15,480; 16, 230) 17,060 (16,660; 17,450)
Deviance information criterion c -6,420.4 37,394.2

a UC usual care, SHAM sham acupuncture, ACU acupuncture, Headache group headache, migraine and TTH, Musculoskeletal group neck, shoulder and low back pain

bFor the EQ-5D endpoint models used approx. 14800 observations; for the SMD endpoint models used approx. 15900. Models should be preferred when total residual deviance mean posterior is close to the actual number of data points

cDeviance information criterion (DIC) is a statistical measure of model fit and model comparison. Models with smaller DIC are preferred