Skip to main content

Table 1 Summary of the five databases used in this study

From: Learning patient-level prediction models across multiple healthcare databases: evaluation of ensembles for increasing model transportability

Name

Type

Description

Start

End

Size (million lives)

IBM Medicare Supplemental Beneficiaries (MDCR)

US Claims

Patients aged 65 or older with supplemental healthcare

2000–01–01

2019–12–31

10

IBM Medicaid (MDCD)

US Claims

Patients with government subsidized healthcare

2006–01–01

2018–12–31

28

Optum® De-Identified Clinformatics® Data Mart Database (Optum Claims)

US Claims

Patients of all ages

2000–05–01

2019–12–31

84

IBM Commercial Claims and Encounters (CCAE)

US Claims

The patients in this database are aged 65 or younger. They are employees who receive health insurance through their employer and their dependents

2000–01–01

2019–12–31

152

Optum® de-identified Electronic Health Record Dataset (Optum EHR)

US EHR

Patients of all ages

2006–01–01

2019–03–31

96