Check for duplicate TR records

This check looks for duplicate TR records and returns a data frame. Only applies to assessments by Investigator, selected based on uppercased TREVAL = "INVESTIGATOR" or missing or TREVAL variable does not exist.

check_tr_dup(TR)

Arguments

TR

dataframe with variables USUBJID, TRCAT, TRLINKID/TRLNKID, TRTESTCD, TRSTRESC, TRDTC, TRSPID (if it exists)

Author

Joel Laxamana

Examples

# example with an error
TR <- data.frame(
 USUBJID  = c(1,1,2,2),
 TRCAT    = c(1,1,2,2),
 TRTESTCD = c(1,1,2,2),
 TRLINKID = c(1,1,2,2),
 TRDTC = c(rep("2016-01-01",2), rep("2016-06-01",2)),
 TRSTRESC = c(1,1,2,2),
 TRSPID = "FORMNAME-R:19/L:19XXXX",
 TREVAL = "INVESTIGATOR",
 stringsAsFactors = FALSE
)

check_tr_dup(TR)
#> [1] FALSE
#> attr(,"msg")
#> [1] "There are 2 duplicated TR records. "
#> attr(,"data")
#> # A tibble: 2 × 7
#> # Groups:   USUBJID, TRCAT, TRTESTCD, TRLINKID, TRSPID, TRDTC, TRSTRESC [2]
#>   USUBJID TRCAT TRTESTCD TRLINKID TRSPID                 TRDTC      TRSTRESC
#>     <dbl> <dbl>    <dbl>    <dbl> <chr>                  <chr>         <dbl>
#> 1       1     1        1        1 FORMNAME-R:19/L:19XXXX 2016-01-01        1
#> 2       2     2        2        2 FORMNAME-R:19/L:19XXXX 2016-06-01        2

TR1 <- TR
TR1$TRSPID <- NULL

check_tr_dup(TR1)
#> [1] FALSE
#> attr(,"msg")
#> [1] "There are 2 duplicated TR records. "
#> attr(,"data")
#> # A tibble: 2 × 6
#> # Groups:   USUBJID, TRCAT, TRTESTCD, TRLINKID, TRDTC, TRSTRESC [2]
#>   USUBJID TRCAT TRTESTCD TRLINKID TRDTC      TRSTRESC
#>     <dbl> <dbl>    <dbl>    <dbl> <chr>         <dbl>
#> 1       1     1        1        1 2016-01-01        1
#> 2       2     2        2        2 2016-06-01        2

TR2 <- TR
TR2$TREVAL <- NULL

check_tr_dup(TR2)
#> [1] FALSE
#> attr(,"msg")
#> [1] "There are 2 duplicated TR records. "
#> attr(,"data")
#> # A tibble: 2 × 7
#> # Groups:   USUBJID, TRCAT, TRTESTCD, TRLINKID, TRSPID, TRDTC, TRSTRESC [2]
#>   USUBJID TRCAT TRTESTCD TRLINKID TRSPID                 TRDTC      TRSTRESC
#>     <dbl> <dbl>    <dbl>    <dbl> <chr>                  <chr>         <dbl>
#> 1       1     1        1        1 FORMNAME-R:19/L:19XXXX 2016-01-01        1
#> 2       2     2        2        2 FORMNAME-R:19/L:19XXXX 2016-06-01        2

# example with no records flagged because issues only among IRF records
TR3 <- TR
TR3$TREVAL <- "INDEPENDENT ASSESSOR"
check_tr_dup(TR3)
#> [1] TRUE

# example with required variable missing
TR4 <- TR
TR4$TRLINKID <- NULL
check_tr_dup(TR4)
#> [1] FALSE
#> attr(,"msg")
#> [1] "TR is missing both the TRLINKID and TRLNKID variables. "