Calling example("example_smk_ume")
will run an unrelated mean
effects (inconsistency) NMA model with the smoking cessation data, using the
code in the Examples section below. The resulting stan_nma
object
smk_fit_RE_UME
will then be available in the global environment.
Examples
# Set up network of smoking cessation data
head(smoking)
#> studyn trtn trtc r n
#> 1 1 1 No intervention 9 140
#> 2 1 3 Individual counselling 23 140
#> 3 1 4 Group counselling 10 138
#> 4 2 2 Self-help 11 78
#> 5 2 3 Individual counselling 12 85
#> 6 2 4 Group counselling 29 170
smk_net <- set_agd_arm(smoking,
study = studyn,
trt = trtc,
r = r,
n = n,
trt_ref = "No intervention")
# Print details
smk_net
#> A network with 24 AgD studies (arm-based).
#>
#> ------------------------------------------------------- AgD studies (arm-based) ----
#> Study Treatment arms
#> 1 3: No intervention | Group counselling | Individual counselling
#> 2 3: Group counselling | Individual counselling | Self-help
#> 3 2: No intervention | Individual counselling
#> 4 2: No intervention | Individual counselling
#> 5 2: No intervention | Individual counselling
#> 6 2: No intervention | Individual counselling
#> 7 2: No intervention | Individual counselling
#> 8 2: No intervention | Individual counselling
#> 9 2: No intervention | Individual counselling
#> 10 2: No intervention | Self-help
#> ... plus 14 more studies
#>
#> Outcome type: count
#> ------------------------------------------------------------------------------------
#> Total number of treatments: 4
#> Total number of studies: 24
#> Reference treatment is: No intervention
#> Network is connected
# \donttest{
# Fitting an unrelated mean effects (inconsistency) model
smk_fit_RE_UME <- nma(smk_net, refresh = if (interactive()) 200 else 0,
consistency = "ume",
trt_effects = "random",
prior_intercept = normal(scale = 100),
prior_trt = normal(scale = 100),
prior_het = normal(scale = 5))
smk_fit_RE_UME
#> A random effects NMA with a binomial likelihood (logit link).
#> An inconsistency model ('ume') was fitted.
#> Inference for Stan model: binomial_1par.
#> 4 chains, each with iter=2000; warmup=1000; thin=1;
#> post-warmup draws per chain=1000, total post-warmup draws=4000.
#>
#> mean se_mean sd 2.5%
#> d[Group counselling vs. No intervention] 1.13 0.02 0.78 -0.35
#> d[Individual counselling vs. No intervention] 0.90 0.01 0.27 0.40
#> d[Self-help vs. No intervention] 0.33 0.01 0.58 -0.85
#> d[Individual counselling vs. Group counselling] -0.30 0.01 0.63 -1.53
#> d[Self-help vs. Group counselling] -0.61 0.01 0.68 -1.97
#> d[Self-help vs. Individual counselling] 0.17 0.02 1.08 -1.90
#> lp__ -5765.42 0.19 6.27 -5778.23
#> tau 0.93 0.01 0.22 0.58
#> 25% 50% 75%
#> d[Group counselling vs. No intervention] 0.62 1.09 1.64
#> d[Individual counselling vs. No intervention] 0.72 0.90 1.07
#> d[Self-help vs. No intervention] -0.05 0.33 0.71
#> d[Individual counselling vs. Group counselling] -0.70 -0.30 0.09
#> d[Self-help vs. Group counselling] -1.04 -0.59 -0.18
#> d[Self-help vs. Individual counselling] -0.52 0.15 0.87
#> lp__ -5769.53 -5765.44 -5761.08
#> tau 0.77 0.90 1.05
#> 97.5% n_eff Rhat
#> d[Group counselling vs. No intervention] 2.73 2256 1.00
#> d[Individual counselling vs. No intervention] 1.47 1130 1.00
#> d[Self-help vs. No intervention] 1.46 1946 1.00
#> d[Individual counselling vs. Group counselling] 0.96 2221 1.00
#> d[Self-help vs. Group counselling] 0.73 2494 1.00
#> d[Self-help vs. Individual counselling] 2.35 3461 1.00
#> lp__ -5753.65 1104 1.00
#> tau 1.46 1096 1.01
#>
#> Samples were drawn using NUTS(diag_e) at Wed Mar 6 13:06:25 2024.
#> For each parameter, n_eff is a crude measure of effective sample size,
#> and Rhat is the potential scale reduction factor on split chains (at
#> convergence, Rhat=1).
# }
# \dontshow{
if (requireNamespace("pkgdown", quietly = TRUE) && pkgdown::in_pkgdown()) {
assign("smk_net", smk_net, .GlobalEnv)
assign("smk_fit_RE_UME", smk_fit_RE_UME, .GlobalEnv)
}
# }