Evaluates the credibility of subgroup analyses using the ICEMAN criteria. The 10 criteria assess whether an apparent subgroup effect is likely to be a true effect modification.
Usage
assess_iceman(
subgroup_result = NULL,
is_prespecified = FALSE,
hypothesis_direction = c("none", "correct", "opposite"),
n_subgroups = 1,
biological_rationale = c("none", "weak", "moderate", "strong"),
effect_measure = c("uncertain", "consistent", "opposite"),
within_study = TRUE,
statistical_test = c("none", "informal", "formal"),
interaction_pvalue = NA_real_,
replication = c("not_applicable", "no", "yes"),
other_evidence = c("none", "weak", "moderate", "strong")
)Arguments
- subgroup_result
Subgroup analysis result (from create_subgroup_table)
- is_prespecified
Logical. Was the subgroup prespecified?
- hypothesis_direction
Character. Was direction prespecified? Options: "correct", "opposite", "none"
- n_subgroups
Integer. Total number of subgroups analyzed
- biological_rationale
Character. Strength of biological rationale: "strong", "moderate", "weak", "none"
- effect_measure
Character. Consistent with overall effect? "consistent", "opposite", "uncertain"
- within_study
Logical. Is this within-study comparison?
- statistical_test
Character. Type of interaction test used: "formal", "informal", "none"
- interaction_pvalue
Numeric. P-value for interaction (if tested)
- replication
Character. Has finding been replicated? "yes", "no", "not_applicable"
- other_evidence
Character. Supporting evidence from other sources: "strong", "moderate", "weak", "none"