lb update for percent differentials for certain lab tests

[bms63]

We will use this in our ad_adlb script.

https://github.com/pharmaverse/admiral/blob/main/R/derive_param_wbc_abs.R

[bms63]

All patients right only have absolute values - lets just update some of those existing values to what we need rather than create new observations.

[bms63]

See admiral.test/inst/data_scripts/update_ds.R for some guidance

[bms63]

# A tibble: 11 × 6
   USUBJID PARAMCD  AVAL PARAM                       VISIT         DTYPE      
   <chr>   <chr>   <dbl> <chr>                       <chr>         <chr>      
 1 P01     WBC     33    Leukocyte Count (10^9/L)    CYCLE 1 DAY 1 NA         
 2 P01     WBC     38    Leukocyte Count (10^9/L)    CYCLE 2 DAY 1 NA         
 3 P01     LYMLE    0.9  Lymphocytes (fraction of 1) CYCLE 1 DAY 1 NA         
 4 P01     LYMLE    0.7  Lymphocytes (fraction of 1) CYCLE 2 DAY 1 NA         
 5 P01     ALB     36    Albumin (g/dL)              CYCLE 2 DAY 1 NA         
 6 P02     WBC     33    Leukocyte Count (10^9/L)    CYCLE 1 DAY 1 NA         
 7 P02     LYMPH   29    Lymphocytes Abs (10^9/L)    CYCLE 1 DAY 1 NA         
 8 P02     LYMLE    0.87 Lymphocytes (fraction of 1) CYCLE 1 DAY 1 NA         
 9 P03     LYMLE    0.89 Lymphocytes (fraction of 1) CYCLE 1 DAY 1 NA         
10 P01     LYMPH   29.7  Lymphocytes Abs (10^9/L)    CYCLE 1 DAY 1 CALCULATION
11 P01     LYMPH   26.6  Lymphocytes Abs (10^9/L)    CYCLE 2 DAY 1 CALCULATION

Hey @yangc71 and @millerg23 This is the output from the derive_param_wbc_abs example. Is USUBJID P02 also supposed to have a Calculation done for them?

Also, I'm a little confused on my notes. Can you tell me again what we were updating here? In my notes, I just have update existing values for WBC, but that is it! Looking at the data and this example just makes me a litte more confused :( sorry! Please more guidance...tyty!

[millerg23]

Find parameters holding absolute differential lab test results, and replace them with Percent differential lab test results. Would suggest just to do this with a few subjects, and possibly just a subset of visits within these subjects. The function to convert Percent to Absolute should then produce the Absolute values from the percent values (using the percent and the WBC value). Note: the function to convert percent to absolute, will only do so if the Absolute value does not exist at that time point. Updates rquired are:

1. Subset on some subjects (and visits?), and the lab differential lab tests (in CDISC test data these are: LBTESTCD = 'BASO', 'EOS', 'LYM'. 'MONO', 'PLAT').
2. Over-write the following variables: LBTEST, LBTESTCD, LBSTRESU, LBSTRESN, LBSTRESC (?), and update ranges?? Possibly also original values for completeness (ie. LBORRES, etc).

[bms63]

Thanks @millerg23. I still like to try and make this update with your guidance!

[millerg23]

To keep things simple for the newly created Percent Differential lab tests, just set the variables holding the original results to the SI (standard unit) results. ie once LBSTRESN/C created, then set LBORRES = LBSTRESC. Likewise LBORRESU with LBSTRESU, LBORNRLO/LBORNRHI with LBSTNRLO/LBSTNRHI. @bms63 if you let me know what lab tests you decide to choose, I can supply reasonable normal ranges to you.

[bms63]

LYM - Reference Ranges

lbstresu = "fraction of 1"; rrlo = 0.22; rrhi = 0.44;