Format Bayesian Meta-Analysis Result for IQWiG Submission — format_bayesian_result

Formats a bayesian_meta_result object according to IQWiG Methods v8.0 requirements. Includes effect estimate, credible interval, heterogeneity, and probability statements with proper formatting (semicolons in CIs, specific decimal places, etc.)

Usage

format_bayesian_result_iqwig(
  bayesian_result,
  digits_estimate = 3L,
  digits_ci = 3L,
  digits_tau = 3L,
  locale = get_locale(),
  include_prob = TRUE,
  include_interpretation = TRUE,
  ci_brackets = c("[", "]")
)

Arguments

bayesian_result: A bayesian_meta_result object from bayesian_meta_analysis()
digits_estimate: Integer. Decimal places for estimate (default: 3)
digits_ci: Integer. Decimal places for CI bounds (default: 3)
digits_tau: Integer. Decimal places for heterogeneity (default: 3)
locale: Character. Locale for decimal separator: "en" or "de" (default: current pharmhand locale)
include_prob: Logical. Include probability statements (default: TRUE)
include_interpretation: Logical. Include text interpretation (default: TRUE)
ci_brackets: Character vector of length 2 for CI brackets (default: c("[", "]") per IQWiG)

Value

A list with formatted strings for:

estimate

Formatted point estimate

ci_95

Formatted 95% credible interval as "lower; upper" with brackets

tau

Formatted heterogeneity estimate

tau_ci

Formatted heterogeneity CI

probability_statement

Probability statement string

interpretation

Text interpretation

full_text

Complete formatted summary text

raw

Raw values used to construct the formatted output, with fields:

point_estimate
ci_lower
ci_upper
tau
tau_ci_lower
tau_ci_upper
effect_measure
n_studies
prob_beneficial

Examples

if (FALSE) { # \dontrun{
yi <- log(c(0.75, 0.82))
sei <- c(0.12, 0.15)
result <- bayesian_meta_analysis(
  yi = yi, sei = sei, effect_measure = "hr"
)
formatted <- format_bayesian_result_iqwig(result)
print(formatted$full_text)
} # }