| caseControl {EmpiricalCalibration} | R Documentation |
Odds ratios from a case-control design
Description
Odds ratios from a case-control design
Usage
data(caseControl)
Format
A data frame with 47 rows and 4 variables:
- drugName
Name of the drug
- groundTruth
Whether the drug is a positive (1) or negative (0) control
- logRr
The log of the incidence rate ratio
- seLogRr
The standard error of the log of the incidence rate ratio
Details
A dataset containing the odds ratios (and standard errors) produced using a case-control design. The outcome is upper GI bleeding, the drug of interest (groundTruth = 1) is sertraline. Also included are 46 negative control drugs, for which we believe there to be no causal relation with upper GI bleeding. We used a database of medical records from general practices in the USA, the General Electric (GE) Centricity database, which contains data on 11.2 million subjects. We restricted on study period (start of 1990 through November 2003), age requirements (18 years or older), available time prior to event (180 days), number of controls per case (6), and risk definition window (30 days following the prescription). Controls were matched on age and sex. Cases of upper GI bleeding were identified on the basis of the occurrence of ICD-9 diagnosis codes in the problem list. These codes pertain to esophageal, gastric, duodenal, peptic, and gastrojejunal ulceration, perforation, and hemorrhage, as well as gastritis and non-specific gastrointestinal hemorrhage. For more information on this set see Schuemie et al (2014).
References
Schuemie MJ, Ryan PB, Dumouchel W, Suchard MA, Madigan D. Interpreting observational studies: why empirical calibration is needed to correct p-values. Statistics in Medicine 33(2):209-18,2014