From caf7141274a216cc9862ec8726098a0d7cf6c0dd Mon Sep 17 00:00:00 2001 From: Vincent Shields Date: Mon, 20 Nov 2023 19:17:46 -0800 Subject: [PATCH] Update README.md --- README.md | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 56 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index ff26112..24e2b36 100644 --- a/README.md +++ b/README.md @@ -19,16 +19,68 @@ install requirements pip install -r requirements.txt ``` -## Usage example +or install the sofware here + +## Use Case ```sh python new_height_vals.py ``` -You will be prompted to enter the old listing first, then the new listing (the listing with the changes you are looking for). Make sure you pull them from the exact same source in EDC. -You will then be prompted to enter the columns you wish to match. Make sure to add spaces between. +Sites often enter data in EDC systems later than the respective study documents indicate that they should. This is often refered to as data latency, and it happens at every Clinical Trial. +However, some cases are more egregious than others. Lets create a mock Phase III clinical trial as an example: + +# CLINICAL STUDY PROTOCOL +Protocol Title: A Phase 3, Randomized, Stratified, Observer-Blind, +Placebo-Controlled Study to Evaluate the Efficacy, and Safety of Insomnistop in +Adults Aged 18 Years and Older +Protocol Number: inso7385 +Sponsor Name: TheSleepSpace + +| Primary endpoint: Improvemnt of sleep quality| Statistical Analysis Methods| +|--------------|-----------| +| | A Statisticaly significant decrease in the Sleep Quality Assessment from baseline to Last Patient Visit | + [](![jcsm 13 3 479 t03](https://github.com/vlshields/NewLineListingVals/assets/25306963/f3d6937f-62c3-4049-8778-399c3db61b8b) + + +Theres no mention of specific prohibited medications in the mock protocol, but there are known medications that could effect the studies primary endpoints. Sometimes patients taking such medications +should have not met inc/exc in the first place. Other times, the study team my have to render their results unusable. + +Lets take Concamitant medications for example. Perhaps a study coordinator neglected to enter a medication in the patients folder, maliciciously or not. Sure DM could have programatic flags for this in +EDC,but in a trial with over 1000 patients screened, I think a way to pull automated reports should be available for CMs and DMs. + +First lets make some fake data: + +```sh +# you can modify the fake data however you wish in this file +python fake_listing_data.py +``` +We pull this first report in november. Take a look at patient 715 +Screenshot 2023-11-20 031700 + +We pull another listing in december, take a look at this patient again: + +Screenshot 2023-11-20 031108 + +We can see that there were medications entered later for this patient. We would have to go to EDC to look at exactly who, why, and when this data was entered. In a large clinical trial, I tool like this could help us spot +these misshaps more quickly. + +## How to use + +1. Open the program in any directory besides C:\User\. I suggests making a new folder on your desktop or documents and keeping it there. You can run it the pythonic way or simple click on the .exe file. +2. Click and drag your **first** line listing after prompted. Then drag the listing you want to see changes on. +3. THe program should generate an excel file called NewVals{DateofExecustion}.xlsx. Under the indicator columnm, uncheck both to see the old (some become deleted or updated) and new values. + + image + +Some data get removed as well: + +image + + + ## Development setup -Make sure to have excel installed. You will find a file called NewVals{todaysdate}.xlsx in your working directory. Go to the indicator column and deselect "Both" +Make sure to have excel installed or libreoffice installed. I have not tested it in google sheets but if someone does please let me know how it works! ## Release History