Join KRI Analysis_Input by Denominator Type

Groups bound Analysis_Input data by denominator type and pivots numerators into separate columns for shared bootstrap processing.

This function validates that all KRIs sharing the same DenominatorType have identical AccrualThreshold values. This is critical because different AccrualThreshold values cause different Denominator calculations due to date adjustment logic in Input_CountSiteByMonth, which would lead to incorrect results when pivoting the data.

Usage

JoinKRIByDenominator(dfInput, dfMetrics, bSkipValidation = FALSE)

Arguments

dfInput

data.frame. Bound Analysis_Input from BindResults. Must contain columns: MetricID, GroupID, GroupLevel, Numerator, Denominator, StudyID, MonthYYYYMM, and DenominatorType.

dfMetrics

data.frame. Metrics metadata from MakeMetric. Must contain columns: MetricID and AccrualThreshold.

bSkipValidation

logical. If TRUE, skips validation checks that require collect() operations (AccrualThreshold consistency, GroupLevel consistency, and Denominator value matching). Use this to improve performance with large database tables when you are confident your workflow configuration is correct. Default: FALSE.

Value

Named list with one tibble per denominator type. Each tibble has common columns plus Numerator columns renamed by MetricID.

Examples

dfInput <- data.frame(
  MetricID = c("kri0001", "kri0001", "kri0003", "kri0003"),
  GroupID = c("Site1", "Site2", "Site1", "Site2"),
  GroupLevel = "Site",
  Numerator = c(5, 3, 2, 1),
  Denominator = c(100, 80, 100, 80),
  StudyID = "AA-1",
  MonthYYYYMM = 202301,
  DenominatorType = "Visit"
)

dfMetrics <- data.frame(
  MetricID = c("kri0001", "kri0003"),
  AccrualThreshold = c(180, 180)
)

lJoined <- JoinKRIByDenominator(dfInput, dfMetrics)

names(lJoined)
#> [1] "Visit"

# Example of error condition (different AccrualThreshold for same DenominatorType)
if (FALSE) { # \dontrun{
dfMetrics_mismatched <- data.frame(
  MetricID = c("kri0001", "kri0003"),
  AccrualThreshold = c(180, 25) # Different values!
)

# This will throw an error:
lJoined <- JoinKRIByDenominator(dfInput, dfMetrics_mismatched)
} # }

# Example 3: Skip validation for performance with large database tables
if (FALSE) { # \dontrun{
# When working with lazy tables and confident about data quality:
lJoined <- JoinKRIByDenominator(
  dfInput = tbl_lazy_input,
  dfMetrics = dfMetrics,
  bSkipValidation = TRUE
)
} # }