We have provided the medExtractR output and gold standards for the tacrolimus and lamotrigine test sets used to develop the dose building algorithm detailed in this paper. This data comes from Vanderbilt’s EHR system. In this vignette, we show how to access this data, how to implement the algorithm, and how to compare the algorithm output to the gold standard using the tacrolimus data. More details of the functions used in the algorithm can be found in our EHR vignette for Extract-Med and Pro-Med-NLP.
Several rows of the medExtractR output for tacrolimus are shown below.
tac_mxr_fn <- system.file("examples", "tac_mxr_out.csv", package = "EHR")
tac_mxr <- read.csv(tac_mxr_fn, na = '')
tac_mxr[c(135:139,163:167,283:289,343:346),]## filename entity expr pos
## 135 X240866534_2010-01-28_4070129 DrugName Tacrolimus 839:849
## 136 X240866534_2010-01-28_4070129 DrugName Prograf 851:858
## 137 X240866534_2010-01-28_4070129 Strength 1 mg 860:864
## 138 X240866534_2010-01-28_4070129 DoseAmt 4 874:875
## 139 X240866534_2010-01-28_4070129 Frequency every twelve hours 888:906
## 163 X240866534_2010-01-28_9659069 DrugName Tacrolimus 150:160
## 164 X240866534_2010-01-28_9659069 DrugName Prograf 162:169
## 165 X240866534_2010-01-28_9659069 Strength 1 mg 171:175
## 166 X240866534_2010-01-28_9659069 DoseAmt 4 185:186
## 167 X240866534_2010-01-28_9659069 Frequency every twelve hours 199:217
## 283 X410930205_2006-06-20_3473651 DrugName Prograf 870:877
## 284 X410930205_2006-06-20_3473651 Dose 3mg 878:881
## 285 X410930205_2006-06-20_3473651 Frequency BID 882:885
## 286 X410930205_2006-06-20_3473651 DrugName prograf 943:950
## 287 X410930205_2006-06-20_3473651 Strength 1mg 951:954
## 288 X410930205_2006-06-20_3473651 DoseAmt 3 955:956
## 289 X410930205_2006-06-20_3473651 Frequency bid 961:964
## 343 X410930205_2006-06-20_2809083 DrugName prograf 560:567
## 344 X410930205_2006-06-20_2809083 Strength 1mg 568:571
## 345 X410930205_2006-06-20_2809083 DoseAmt 3 572:573
## 346 X410930205_2006-06-20_2809083 Frequency bid 578:581The first step of Part I of our algorithm is parsing the raw NLP
output. This results in a standardized form of the data that includes a
row for each drug mention and columns for all entities anchored to that
drug mention. Here, we use the parseMedExtractR function
since we are using medExtractR output as an example.
Below are the rows of the parsed output corresponding to the raw NLP output from above.
## filename drugname strength
## 1 X240866534_2010-01-28_4070129 Tacrolimus::839::849
## 2 X240866534_2010-01-28_4070129 Prograf::851::858 1 mg::860::864
## 3 X240866534_2010-01-28_9659069 Tacrolimus::150::160
## 4 X240866534_2010-01-28_9659069 Prograf::162::169 1 mg::171::175
## 5 X410930205_2006-06-20_3473651 Prograf::870::877
## 6 X410930205_2006-06-20_3473651 prograf::943::950 1mg::951::954
## 7 X410930205_2006-06-20_2809083 prograf::560::567 1mg::568::571
##
## dose route freq dosestr
## 1
## 2 4::874::875 every twelve hours::888::906
## 3
## 4 4::185::186 every twelve hours::199::217
## 5 BID::882::885 3mg::878::881
## 6 3::955::956 bid::961::964
## 7 3::572::573 bid::578::581
##
## dosechange lastdose
## 1
## 2
## 3
## 4
## 5
## 6
## 7Next, the parsed entities are paired using the buildDose
function. This results in a dataset with a column for each entity and a
row for each pairing.
The output is shown below.
## filename drugname strength dose route
## 51 X240866534_2010-01-28_4070129 Tacrolimus <NA> <NA> <NA>
## 52 X240866534_2010-01-28_4070129 Prograf 1 mg 4 <NA>
## 55 X240866534_2010-01-28_9659069 Tacrolimus <NA> <NA> <NA>
## 56 X240866534_2010-01-28_9659069 Prograf 1 mg 4 <NA>
## 104 X410930205_2006-06-20_2809083 prograf 1mg 3 <NA>
## 105 X410930205_2006-06-20_3473651 Prograf <NA> <NA> <NA>
## 106 X410930205_2006-06-20_3473651 prograf 1mg 3 <NA>
## freq dosestr dosechange lastdose drugname_start
## 51 <NA> <NA> <NA> <NA> 839
## 52 every twelve hours <NA> <NA> <NA> 851
## 55 <NA> <NA> <NA> <NA> 150
## 56 every twelve hours <NA> <NA> <NA> 162
## 104 bid <NA> <NA> <NA> 560
## 105 BID 3mg <NA> <NA> 870
## 106 bid <NA> <NA> <NA> 943We have provided the gold standard that we generated for part 1. Several rows are shown below.
tac_gs_part1 <- read.csv(system.file("examples", "tac_gs_part1.csv", package = "EHR"),
stringsAsFactors = FALSE, na = '')## filename drugname drugname_start strength dose route
## 51 X240866534_2010-01-28_4070129 Tacrolimus 839 <NA> <NA> NA
## 52 X240866534_2010-01-28_4070129 Prograf 851 1 mg 4 NA
## 53 X240866534_2010-01-28_9659069 Tacrolimus 150 <NA> <NA> NA
## 54 X240866534_2010-01-28_9659069 Prograf 162 1 mg 4 NA
## 104 X410930205_2006-06-20_3473651 Prograf 870 <NA> <NA> NA
## 105 X410930205_2006-06-20_3473651 prograf 943 1mg 3 NA
## 107 X410930205_2006-06-20_2809083 prograf 560 1mg 3 NA
## freq dosestr dosechange
## 51 <NA> <NA> <NA>
## 52 every twelve hours <NA> <NA>
## 53 <NA> <NA> <NA>
## 54 every twelve hours <NA> <NA>
## 104 BID 3mg <NA>
## 105 bid <NA> <NA>
## 107 bid <NA> <NA>The following code compares the gold standard to the Part I output and provides the recall and precision measures.
precall <- function(dat, gs) {
tp1 <- sum(dat %in% gs)
fp1 <- sum(!(dat %in% gs))
fn1 <- sum(!(gs %in% dat))
r1 <- c(tp1, tp1 + fn1)
p1 <- c(tp1, tp1 + fp1)
r <- rbind(r1,p1)
dimnames(r) <- list(c('recall','prec'), c('num','den'))
cbind(r, prop = round(r[,1] / r[,2], 2))
}
colsToCompare <- c('filename','drugname','strength','dose','route','freq',
'dosestr','dosechange','drugname_start')
tac_mxr_part1_out <- tac_mxr_part1_out[,colsToCompare]
tac_gs_part1 <- tac_gs_part1[,colsToCompare]
tacxrrow <- do.call(paste, c(tac_mxr_part1_out, sep = '|'))
gs.tacxrrow <- do.call(paste, c(tac_gs_part1, sep = '|'))
precall(tacxrrow, gs.tacxrrow)## num den prop
## recall 285 285 1
## prec 285 285 1In part II of the algorithm, the final datasets are formed containing dose intake and daily dose, and redundancies are removed at the note and date level for each patient.
This part of the algorithm requires more detailed meta data associated with each clinical note file. This is shown below using our example tacrolimus data.
bmd <- function(x) {
fns <- strsplit(x, '_')
pid <- sapply(fns, `[`, 1)
date <- as.Date(sapply(fns, `[`, 2), format = '%Y-%m-%d')
note <- sapply(fns, `[`, 3)
data.frame(filename = x, pid, date, note, stringsAsFactors = FALSE)
}
tac_metadata <- bmd(tac_mxr_part1_out[['filename']])## filename pid date note
## 51 X240866534_2010-01-28_4070129 X240866534 2010-01-28 4070129
## 55 X240866534_2010-01-28_9659069 X240866534 2010-01-28 9659069
## 104 X410930205_2006-06-20_2809083 X410930205 2006-06-20 2809083
## 105 X410930205_2006-06-20_3473651 X410930205 2006-06-20 3473651Below, a few rows of the note level and date level collapsing are shown for our example tacrolimus data.
Note level:
## filename drugname strength dose route freq dosestr
## 40 X240866534_2010-01-28_4070129 Prograf 1 mg 4 orally bid <NA>
## 42 X240866534_2010-01-28_9659069 Prograf 1 mg 4 orally bid <NA>
## 68 X410930205_2006-06-20_2809083 prograf 1mg 3 orally bid <NA>
## 69 X410930205_2006-06-20_3473651 Prograf <NA> <NA> orally bid 3mg
##
## dosechange drugname_start dosestr.num strength.num doseamt.num
## 40 <NA> 851 NA 1 4
## 42 <NA> 162 NA 1 4
## 68 <NA> 560 NA 1 3
## 69 <NA> 870 3 NA NA
##
## freq.num dose.intake intaketime dose.seq dose.daily
## 40 2 4 <NA> NA 8
## 42 2 4 <NA> NA 8
## 68 2 3 <NA> NA 6
## 69 2 3 <NA> NA 6Date level:
## filename drugname strength dose route freq dosestr
## 29 X240866534_2010-01-28_4070129 Prograf 1 mg 4 orally bid <NA>
## 42 X410930205_2006-06-20_2809083 prograf 1mg 3 orally bid <NA>
##
## dosechange drugname_start dosestr.num strength.num doseamt.num
## 29 <NA> 851 NA 1 4
## 42 <NA> 560 NA 1 3
##
## freq.num dose.intake intaketime dose.seq dose.daily
## 29 2 4 <NA> NA 8
## 42 2 3 <NA> NA 6We have provided the gold standards that we generated for part 2.
Note level:
tac_gs_part2_note <- read.csv(
system.file("examples", "tac_gs_part2_note.csv", package = "EHR"),
stringsAsFactors = FALSE, na = ''
)## filename drugname drugname_start strength dose route
## 40 X240866534_2010-01-28_4070129 Prograf 851 1 4 NA
## 41 X240866534_2010-01-28_9659069 Prograf 162 1 4 NA
## 68 X410930205_2006-06-20_3473651 Prograf 870 NA NA NA
## 70 X410930205_2006-06-20_2809083 prograf 560 1 3 NA
## freq intaketime dosestr dosechange doseintake daily
## 40 2 <NA> NA <NA> 4 8
## 41 2 <NA> NA <NA> 4 8
## 68 2 <NA> 3 <NA> 3 6
## 70 2 <NA> NA <NA> 3 6Date level:
tac_gs_part2_date <- read.csv(
system.file("examples", "tac_gs_part2_date.csv", package = "EHR"),
stringsAsFactors = FALSE, na = ''
)## filename drugname drugname_start strength dose route
## 29 X240866534_2010-01-28_4070129 Prograf 851 1 4 NA
## 42 X410930205_2006-06-20_3473651 Prograf 870 NA NA NA
## freq intaketime dosestr dosechange doseintake daily
## 29 2 <NA> NA <NA> 4 8
## 42 2 <NA> 3 <NA> 3 6The following code compares the gold standard to the Part II output
and provides the recall and precision measures for note level and date
level collapsing for dose intake and daily dose. In order to replicate
the results from this paper, we use the Part I gold standard as the
input to collapseDose.
precall <- function(dat, gs) {
tp1 <- sum(dat %in% gs)
fp1 <- sum(!(dat %in% gs))
fn1 <- sum(!(gs %in% dat))
r1 <- c(tp1, tp1 + fn1)
p1 <- c(tp1, tp1 + fp1)
r <- rbind(r1,p1)
dimnames(r) <- list(c('recall','prec'), c('num','den'))
cbind(r, prop = round(r[,1] / r[,2], 2))
}
metaData <- bmd(unique(tac_gs_part1$filename))
tacxr <- collapseDose(tac_gs_part1, metaData, 'bid')
tacxr.note <- tacxr[['note']]
tacxr.date <- tacxr[['date']]
tacxr.note$pid <- sub("_.*","",tacxr.note$filename)
tacxr.date$pid <- sub("_.*","",tacxr.date$filename)
tac_gs_part2_note$pid <- sub("_.*","",tac_gs_part2_note$filename)
tac_gs_part2_date$pid <- sub("_.*","",tac_gs_part2_date$filename)
tacxrrow.note.intake <- do.call(paste, c(tacxr.note[,c('pid','dose.intake',
'dosechange')],sep = '|'))
tacxrrow.note.daily <- do.call(paste, c(tacxr.note[,c('pid','intaketime','dose.daily',
'dosechange')], sep = '|'))
tacxrrow.date.intake <- do.call(paste, c(tacxr.date[,c('pid','dose.intake',
'dosechange')], sep = '|'))
tacxrrow.date.daily <- do.call(paste, c(tacxr.date[,c('pid','intaketime','dose.daily',
'dosechange')], sep = '|'))
gs.tacxrrow.note.intake <- do.call(paste, c(tac_gs_part2_note[,c('pid','doseintake',
'dosechange')], sep = '|'))
gs.tacxrrow.note.daily <- do.call(paste, c(tac_gs_part2_note[,c('pid','intaketime','daily',
'dosechange')], sep = '|'))
gs.tacxrrow.date.intake <- do.call(paste, c(tac_gs_part2_date[,c('pid','doseintake',
'dosechange')], sep = '|'))
gs.tacxrrow.date.daily <- do.call(paste, c(tac_gs_part2_date[,c('pid','intaketime','daily',
'dosechange')], sep = '|'))
precall(tacxrrow.note.intake, gs.tacxrrow.note.intake)
precall(tacxrrow.note.daily, gs.tacxrrow.note.daily)
precall(tacxrrow.date.intake, gs.tacxrrow.date.intake)
precall(tacxrrow.date.daily, gs.tacxrrow.date.daily)## num den prop
## recall 205 205 1
## prec 205 205 1## num den prop
## recall 205 206 1
## prec 205 205 1## num den prop
## recall 116 116 1
## prec 116 116 1## num den prop
## recall 116 117 0.99
## prec 116 116 1.00