An approach to improving efficiency of actigraphy data in research: use of statistical software to facilitate batch import.

Abstract

Despite the recent discontinuation of Philips Actiwatch devices, volumes of actigraphy data remain unanalyzed, and these and similar devices continue to be used for clinical and research applications. In this report, we present a Stata "do" file that automates the process of importing and appending raw data files downloaded from Philips Actiware software. For example, in less than 60 seconds, our do file imported, appended, and cleaned 189,596 epochs of data from raw 48-hour actigraphy data files of 35 critically ill patients, yielding a single unified dataset with usable variables ready for analysis. Portable and scalable, this approach can facilitate error-free generation of single- or multi-day actigraphy datasets from an unlimited number of patients. Although designed for Stata, this simple executable file can be adapted for other data platforms and applied to new and existing actigraphy data files and modified as needed for other actigraphy-based efforts.