Over the past 8 years, I have spent thousands of hours in STATA. In close collaboration with prof. Jesper Hallas, and the two data managers Morten Olesen and Martin T. Ernst, all at the University of Southern Denmark, we have learned and developed quite a few techniques that facilitate the handling of epidemiological datasets. This includes basic steps such as efficient use of loops and other single commands, a stringent folder structure and general nomenclature, and specific programs (e.g handling of time-variant exposures/confounders, an algorithm for 1:n matching, and an automatic assessment of baseline characteristics). Most importantly, however, we have worked with a concept we refer to as “full programming”, where final tables are created for direct copy/paste into e.g. Word, instead of manual transfer of each result, with its inherent mistakes.

Three years ago, I described some of our techniques in a short series of video tutorials (here). During the Summer and Fall of 2017, we aim to update these videos to include our newest programs, this time in English and with code fragments and datasets to support it. If you have any comments or suggestions for the videos, please let us know.



+45 28913340

Anton Pottegaard
Lysalleen 7F
4000 Roskilde