In quinterpriest/Achilles: Generates descriptive statistics for an OMOP CDM instance

Getting Started

Make sure you have your data in a supported OMOP CDM format. Achilles currently supports version 5.3 and 5.4 of the OMOP CDM. (https://github.com/OHDSI/CommonDataModel).
This package makes use of rJava. Make sure that you have Java installed. If you don't have Java already installed on your computer (on most computers it already is installed), go to java.com to get the latest version. If you are having trouble with rJava, this Stack Overflow post may assist you when you begin troubleshooting.
In R, use the following commands to install Achilles.

```r if (!require("remotes")) install.packages("remotes")

To install the master branch

remotes::install_github("OHDSI/Achilles")

To install latest release (if master branch contains a bug for you)

remotes::install_github("OHDSI/Achilles@*release")

To avoid Java 32 vs 64 issues

remotes::install_github("OHDSI/Achilles", args="--no-multiarch")

```
To run the Achilles analysis, first determine if you'd like to run the function in multi-threaded mode or in single-threaded mode. Use runCostAnalysis = FALSE to save on execution time, as cost analyses tend to run long.

In multi-threaded mode

The analyses are run in multiple SQL sessions, which can be set using the numThreads setting and setting scratchDatabaseSchema to something other than #. For example, 10 threads means 10 independent SQL sessions. Intermediate results are written to scratch tables before finally being combined into the final results tables. Scratch tables are permanent tables; you can either choose to have Achilles drop these tables (dropScratchTables = TRUE) or you can drop them at a later time (dropScratchTables = FALSE). Dropping the scratch tables can add time to the full execution. If desired, you can set your own custom prefix for all Achilles analysis scratch tables (tempAchillesPrefix).

In single-threaded mode

The analyses are run in one SQL session and all intermediate results are written to temp tables before finally being combined into the final results tables. Temp tables are dropped once the package is finished running. Single-threaded mode can be invoked by either setting numThreads = 1 or scratchDatabaseSchema = "#".

Use the following commands in R:

r library(Achilles) connectionDetails <- createConnectionDetails( dbms="redshift", server="server.com", user="secret", password='secret', port="5439")

Single-threaded mode

r achilles(connectionDetails, cdmDatabaseSchema = "cdm5_inst", resultsDatabaseSchema="results", outputFolder = "output")

Multi-threaded mode

r achilles(connectionDetails, cdmDatabaseSchema = "cdm5_inst", resultsDatabaseSchema = "results", scratchDatabaseSchema = "scratch", numThreads = 10, outputFolder = "output")

The "cdm5_inst" cdmDatabaseSchema parameter, "results" resultsDatabaseSchema parameter, and "scratch" scratchDatabaseSchema parameter are the fully qualified names of the schemas holding the CDM data, targeted for result writing, and holding the intermediate scratch tables, respectively. See the DatabaseConnector package for details on settings the connection details for your database, for example by typing

r ?createConnectionDetails

Execution of all Achilles pre-computations may take a long time, particularly in single-threaded mode and with COST analyses enabled. See file to find out how some analyses can be excluded to make the execution faster.

The SQL platforms supported by DatabaseConnector and SqlRender are the only ones supported here in Achilles as dbms. cdmVersion can be ONLY 5.x (please look at prior commit history for v4 support).