This shows you the differences between two versions of the page.
Both sides previous revision Previous revision | |||
documentation:software:methods_library [2017/04/21 08:29] schuemie |
documentation:software:methods_library [2018/12/14 13:12] (current) schuemie |
||
---|---|---|---|
Line 1: | Line 1: | ||
====== Methods library ====== | ====== Methods library ====== | ||
- | The methods library consists of a set of R packages that can be used to execute observational studies against data in the Common Data Model (CDM) format. The following packages are available: | + | The methods library consists of a set of R packages that can be used to execute observational studies against data in the Common Data Model (CDM) format. |
- | * [[https://github.com/OHDSI/SelfControlledCohort|SelfControlledCohort]], previously also known as Observational Screening. | + | For more information, see the [[https://ohdsi.github.io/MethodsLibrary/|Methods Library website]]. |
- | * [[https://github.com/OHDSI/SelfControlledCaseSeries|SelfControlledCaseSeries]], an implementation of the Multiple SCCS method allowing many covariates (e.g. all drug exposures) to be included in the model. | + | |
- | * [[https://github.com/OHDSI/IcTemporalPatternDiscovery|IcTemporalPatternDiscovery]] | + | |
- | * [[https://github.com/OHDSI/CohortMethod|CohortMethod]], for performing new-user cohort studies using large-scale propensity scores (e.g. including all drugs, conditions, procedures, comorbidity indices, etc.). | + | |
- | * [[https://github.com/OHDSI/CaseControl|CaseControl]], for performing case-control studies with options to match on age, gender, visit data, provider, and length of observation, as well as adjusting for many covariates | + | |
- | * [[https://github.com/OHDSI/CaseCrossover|CaseCrossover]], for performing case-crossover studies with options to adjust for time-trends in exposure (case-time-control), and specifying multiple control windows. | + | |
- | + | ||
- | All OHDSI methods are designed to be able to run customized one-off studies for a particular exposure-outcome pair, but also across a large set of pairs, and using many different predefined analysis choices. | + | |
- | + | ||
- | **Supporting packages** | + | |
- | + | ||
- | From the OMOP experiment we learned that it is important to measure and understand the operating characteristics of methods in the settings that they're used, and to use these measured operating characteristics to calibrate the estimates produced by the methods. To facilitate these tasks, the following packages have been developed: | + | |
- | + | ||
- | * [[https://github.com/OHDSI/MethodEvaluation|MethodEvaluation]], for evaluating the performance of a method against established reference sets and simulated data. | + | |
- | * [[https://github.com/OHDSI/EmpiricalCalibration|EmpiricalCalibration]], for estimating the standard error distribution of a method using negative controls, and computing calibrated p-values. | + | |
- | + | ||
- | **Using the OHDSI R packages** | + | |
- | + | ||
- | All packages have package manuals describing the functions available in the package, and most packages have vignettes that describe how to use the package in a more user-friendly way. You can access the manuals and vignettes from the front pages of each Github repository. | + | |
- | + | ||
- | [[documentation:r setup|How to set up your R environment]] to run the methods library. | + |