Skip to contents

Assess Risk of Bias using RoB 2

Source: R/bias_rob2.R
assess_rob2.Rd

Assesses the risk of bias for a single study using the Cochrane RoB 2 tool. The overall judgment is automatically calculated based on domain judgments using the following rules:

  • "Low": All domains rated "Low"

  • "High": Any domain rated "High" OR 2+ domains rated "Some concerns"

  • "Some concerns": All other combinations

Usage

assess_rob2(
  study_id,
  d1_randomization,
  d2_deviations,
  d3_missing_data,
  d4_measurement,
  d5_selection,
  d1_support = "",
  d2_support = "",
  d3_support = "",
  d4_support = "",
  d5_support = "",
  outcome = "",
  overall = NULL,
  overall_justification = NULL,
  metadata = list()
)

Arguments

study_id

Character. Study identifier.

d1_randomization

Character. Domain 1 judgment: "Low", "Some concerns", or "High". For randomization process.

d2_deviations

Character. Domain 2 judgment: "Low", "Some concerns", or "High". For deviations from intended interventions.

d3_missing_data

Character. Domain 3 judgment: "Low", "Some concerns", or "High". For missing outcome data.

d4_measurement

Character. Domain 4 judgment: "Low", "Some concerns", or "High". For measurement of the outcome.

d5_selection

Character. Domain 5 judgment: "Low", "Some concerns", or "High". For selection of the reported result.

d1_support

Character. Supporting text for Domain 1 (optional).

d2_support

Character. Supporting text for Domain 2 (optional).

d3_support

Character. Supporting text for Domain 3 (optional).

d4_support

Character. Supporting text for Domain 4 (optional).

d5_support

Character. Supporting text for Domain 5 (optional).

outcome

Character. Outcome being assessed (optional).

overall

Character. Overall judgment. If NULL, auto-calculated.

overall_justification

Character. Justification for overall judgment. If NULL and overall is auto-calculated, a default justification is generated.

metadata

List. Additional metadata (optional).

Value

A RoB2Result object with the assessment results.

Examples

if (FALSE) { # \dontrun{
# Basic assessment with auto-calculated overall
result <- assess_rob2(
  study_id = "STUDY001",
  d1_randomization = "Low",
  d2_deviations = "Low",
  d3_missing_data = "Low",
  d4_measurement = "Some concerns",
  d5_selection = "Low",
  d4_support = "Outcome assessor not blinded",
  outcome = "Overall Survival"
)
result@overall
result@judgment_counts

# Manual overall judgment
result2 <- assess_rob2(
  study_id = "STUDY002",
  d1_randomization = "High",
  d2_deviations = "Low",
  d3_missing_data = "Low",
  d4_measurement = "Low",
  d5_selection = "Low",
  overall = "High",
  overall_justification = "Inadequate randomization procedure"
)
} # }