I am looking for a specific type of a dataset in order to develop a statistical and/or machine learning model for health data. This dataset should contain histories of many patients with the diagnoses/procedures that they underwent and the corresponding time labels. I would like to hope that there is such a dataset (or similar) used for machine learning algorithms, which is freely available and not constrained by patient privacy issues, etc.

All suggestions will be greatly appreciated.


Check out ipums.org. They have two sources that may be beneficial to you:

Health Surveys: Historical and contemporary U.S. health survey data from the National Health Interview Survey (NHIS) (1963-present) and the Medical Expenditure Panel Survey (MEPS) (1996-present).

Global Health: Health survey data for Africa and Asia, including harmonized data collections for Demographic Health Surveys (DHS) and Performance Monitoring for Action (PMA)

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.