See also “Pro-Demographic” in “2. EHR Vignette for Structured Data” of EHR package.

Introduction

This tutorial describes how to use the Pro-Demographic module to process demographic data (see Choi et al.\(^{1}\) for details).

To begin we load the EHR package and create some directories to work from.

# load EHR package
library(EHR)

td <- tempdir()
checkDir <- file.path(td, 'checks') # directory for interactive checking
dir.create(checkDir)
dataDir <- file.path(td, 'data') # directory for processed data
dir.create(dataDir)
rawDataDir <- system.file("examples", "str_ex2", package="EHR")
drugname <- 'fent'
LLOQ <- 0.05

Demographic Data

We will use example demographic data to demonstrate the Pro-Demographic module. The raw data is shown below.

demo.in <- read.csv(system.file("examples", "str_ex2","Demographics_DATA.csv", package="EHR"))[,c(1,2,3,4,5)]
head(demo.in,10)

   subject_id subject_uid gender weight height
1      1106.0    34364670      0   5.14  59.18
2      1444.0    36792472      1   5.67  62.90
3      1465.0    36292449      0  23.67 118.02
4      1520.0    34161967      0  14.07  97.04
5      1524.0    37857374      1  23.40 102.80
6      1550.0    37826262      1   6.21  62.03
7      1566.0    35885929      0 101.32 176.26
8      1596.0    38340814      1   6.79  62.99
9      1607.1    38551767      0  10.81  74.95
10     1607.0    38551767      0   2.76  45.94

Each row in the data represents an individual surgical date. In this case, each individual is assigned a unique ID and has a recorded gender, weight, and height for each surgery. The goal of Pro-Demographic is to make the demographic data suitable for merging with medication data for population pharmacokinetic analysis while apply exclusion criteria in a reproducible manner which preserves and organizes the original data.

Preparing the Demographic Data

td <- tempdir()
dir.create(file.path(td, 'data2'))
dataDir <- file.path(td, 'data2')
saveRDS(demo.in, file=file.path(dataDir,"demo.rds"))

Running `run_Demo`

run_Demo will configure the demographic data, apply exclusion criteria and saving a record of exclusion rules/excluded individuals. The argument toexclude takes an expression evaluating to TRUE/FALSE indicating which individuals will be excluded from the final dataset.

low_wgt <- function(x, val=1) {  x < 6 }
demo.out <- run_Demo(demo.path = file.path(dataDir, "demo.rds"),
                     demo.columns = list(id = 'subject_id'),
                     toexclude = expression(low_wgt(weight)))

The number of subjects in the demographic data, who meet the exclusion criteria: 9

Processed Data

The demographic data is now ready for further processing.

head(demo.out)

$demo
   subject_id subject_uid gender weight height
1      1106.0    34364670      0   5.14  59.18
2      1444.0    36792472      1   5.67  62.90
3      1465.0    36292449      0  23.67 118.02
4      1520.0    34161967      0  14.07  97.04
5      1524.0    37857374      1  23.40 102.80
6      1550.0    37826262      1   6.21  62.03
7      1566.0    35885929      0 101.32 176.26
8      1596.0    38340814      1   6.79  62.99
9      1607.1    38551767      0  10.81  74.95
10     1607.0    38551767      0   2.76  45.94
11     1724.0    39087607      0   5.21  58.16
12     1770.1    39418554      1   5.44  60.44
13     1770.0    39418554      1   5.38  60.39
14     2157.0    42023523      1   4.20  54.95
15     2162.0    42044808      0   5.00  55.68
16     2164.0    41221120      0   5.80  61.90
17      466.1    28579217      0  21.99 116.90
18      466.0    28579217      0   9.23  77.08

$exclude
[1] 1106.0 1444.0 1607.0 1724.0 1770.1 1770.0 2157.0 2162.0 2164.0

References

Choi L, Beck C, McNeer E, Weeks HL, Williams ML, James NT, Niu X, Abou-Khalil BW, Birdwell KA, Roden DM, Stein CM. Development of a System for Post-marketing Population Pharmacokinetic and Pharmacodynamic Studies using Real-World Data from Electronic Health Records. Clinical Pharmacology & Therapeutics. 2020 Apr;107(4):934-43. doi: 10.1002/cpt.1787.

Pro-Demographic