Skip to contents

Transform Cumulative Counts to Study Level

Transform_CumCount.Rd

Transforms site-level counts into study-level metrics by aggregating across groups (sites/countries) by calendar month and calculating cumulative sums. Supports both in-memory data frames and dbplyr lazy tables. Handles multiple Numerator columns (e.g., Numerator_kri0001, Numerator_kri0003).

Note: StudyMonth is calculated by this function based on MonthYYYYMM order. Date normalization should be done at the input level (Input_CountSiteByMonth).

Usage

Transform_CumCount(dfInput, vBy, tblMonthSequence = NULL)

Arguments

dfInput

data.frame or tbl. Output from Input_CountSiteByMonth with columns: GroupID, Numerator (or Numerator_*), Denominator, StudyID, MonthYYYYMM. Note: Input should contain monthly (not cumulative) counts. This function calculates cumulative sums at the study level and adds StudyMonth.

vBy

character. Vector of column names for grouping (e.g., "StudyID" or c("StudyID", "BootstrapRep")).

tblMonthSequence

tbl_lazy, data.frame, or NULL. For lazy table inputs: Optional pre-generated month sequence with only a MonthYYYYMM column (output of GenerateMonthSeq()). Must contain consecutive months with NO gaps. If NULL, attempts to create temp table (requires write privileges). If data.frame provided, will be written to temp table.

Value

A data.frame with study-level aggregated cumulative counts and ratios. Output columns: vBy columns, MonthYYYYMM, StudyMonth, Numerator columns, Denominator, Metric columns, GroupCount.

Examples

# Generate input data
dfSubjects <- clindata::rawplus_dm
dfNumerator <- clindata::rawplus_ae
dfDenominator <- clindata::rawplus_visdt

dfInput <- Input_CountSiteByMonth(
  dfSubjects = dfSubjects,
  dfNumerator = dfNumerator,
  dfDenominator = dfDenominator,
  strNumeratorDateCol = "aest_dt",
  strDenominatorDateCol = "visit_dt"
)

# Transform to study level and calculate StudyMonth
dfTransformed <- Transform_CumCount(
  dfInput = dfInput,
  vBy = "StudyID"
)

print(dfTransformed)
#> # A tibble: 194 × 7
#>    StudyID        MonthYYYYMM StudyMonth Numerator Denominator Metric GroupCount
#>    <chr>                <dbl>      <int>     <int>       <int>  <dbl>      <int>
#>  1 AA-AA-000-0000      200309          1         1           2  0.5            1
#>  2 AA-AA-000-0000      200310          2         1           5  0.2            3
#>  3 AA-AA-000-0000      200311          3         1          10  0.1            5
#>  4 AA-AA-000-0000      200312          4         3          16  0.188          7
#>  5 AA-AA-000-0000      200401          5         4          33  0.121         12
#>  6 AA-AA-000-0000      200402          6         5          50  0.1           14
#>  7 AA-AA-000-0000      200403          7        11          78  0.141         19
#>  8 AA-AA-000-0000      200404          8        23         114  0.202         23
#>  9 AA-AA-000-0000      200405          9        34         161  0.211         29
#> 10 AA-AA-000-0000      200406         10        39         213  0.183         35
#> # ℹ 184 more rows