An S7 class for representing meta-analysis results including pooled estimates, heterogeneity statistics, and study weights.
Usage
MetaResult(
estimate = integer(0),
ci = integer(0),
ci_level = 0.95,
p_value = NA_real_,
method = "",
n = NA_integer_,
metadata = list(),
model = "random",
effect_measure = "hr",
heterogeneity = list(Q = NA_real_, Q_df = NA_integer_, Q_pvalue = NA_real_, I2 =
NA_real_, H2 = NA_real_, tau2 = NA_real_, tau = NA_real_),
weights = NULL,
prediction_interval = NULL,
study_results = list()
)Arguments
- estimate
Numeric pooled effect estimate
- ci
Numeric vector c(lower, upper)
- ci_level
Numeric confidence level
- p_value
Numeric p-value
- method
Character string for statistical method
- n
Integer number of studies
- metadata
List of additional metadata
- model
Character string: "fixed" or "random"
- effect_measure
Character string: "hr", "or", "rr", "rd", "md", "smd"
- heterogeneity
List with Q, Q_df, Q_pvalue, I2, H2, tau2, tau statistics
- weights
Numeric vector of study weights
- prediction_interval
Numeric vector c(lower, upper) for prediction interval
- study_results
List of individual study ComparisonResult objects
Examples
result <- MetaResult(
estimate = 0.80,
ci = c(0.70, 0.91),
ci_level = 0.95,
p_value = 0.001,
n = 5L,
model = "random",
effect_measure = "hr",
heterogeneity = list(
Q = 15.2, Q_df = 4L, Q_pvalue = 0.004,
I2 = 73.7, H2 = 3.8, tau2 = 0.025, tau = 0.158
),
method = "REML with Knapp-Hartung adjustment"
)
result@estimate
#> [1] 0.8
result@heterogeneity$I2
#> [1] 73.7