Skip to main content

Table 3 IPD NMA ANCOVA synthesis model results (model 1), EQ-5D preference score and standardised pain endpoints

From: Methods for network meta-analysis of continuous outcomes using individual patient data: a case study in acupuncture for chronic pain

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

  1. aUC usual care, SHAM sham acupuncture, ACU acupuncture, Headache group headache, migraine and TTH, Musculoskeletal group neck, shoulder and low back pain
  2. 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
  3. cDeviance information criterion (DIC) is a statistical measure of model fit and model comparison. Models with smaller DIC are preferred