MHserviceDemo {glmertree} | R Documentation |
Artificial mental-health service outcomes dataset
Description
Artificial dataset of treatment outcomes (N = 3739) of 13 mental-health services to illustrate fitting of (G)LMM trees with constant fits in terminal nodes.
Usage
data("MHserviceDemo")
Format
A data frame containing 3739 observations on 8 variables:
- age
numeric. Variable representing age in years (range: 4.8 - 23.6, M = 11.46).
- impact
numeric. Continuous variable representing severity of and impairment due to mental-health problems at baseline. Higher values indicate higher severity and impairment.
- gender
factor. Indicator for gender.
- emotional
factor. Indicator for presence of emotional disorder at baseline.
- autism
factor. Indicator for presence of autistic disorder at baseline.
- conduct
factor. Indicator for mental-health service provider.
- cluster_id
factor. Binarized treatment outcome variable (0 = recovered, 1 = not recovered.
- outcome
numeric. Variable representing treatment outcome as measured by a total mental-health difficulties score assessed about 6 months after baseline, corrected for the baseline assessment. Higher values indicate poorer outcome.
Details
Dataset was modelled after Edbrooke-Childs et al. (2017), who analyzed a sample of $N = 3,739$ young people who received treatment at one of 13 mental-health service providers in the UK. Note that the data were artificially generated and do not reflect actual empirical findings.
References
Fokkema M, Edbrooke-Childs J & Wolpert M (2021). “Generalized linear mixed-model (GLMM) trees: A flexible decision-tree method for multilevel and longitudinal data.” Psychotherapy Research, 31(3), 329-341. doi:10.1080/10503307.2020.1785037
See Also
Examples
data("MHserviceDemo", package = "glmertree")
summary(MHserviceDemo)
lt <- lmertree(outcome ~ 1 | cluster_id | age + gender + emotional +
autism + impact + conduct, data = MHserviceDemo)
plot(lt)
gt <- glmertree(factor(outcome > 0) ~ 1 | cluster_id | age + gender +
emotional + autism + impact + conduct,
data = MHserviceDemo, family = "binomial")
plot(gt)