CSA Big data study design | ResMed
CSA Big data study design

CSA Big data study design

This study of CSA/CompSA looked at anonymous data from PAP devices in the US1:

• Anonymous data from 133,000 telemonitored patients treated for SDB with ResMed Positive Airway Pressure (PAP) devices

• In the US

• An external international committee of experts appointed by ResMed1: Jean-Louis Pépin (France), Holger Woehrle (Germany), Atul Malhotra (USA), Peter Cistulli (Australia).

Patients agreed to be telemonitored, and data were de-identified and aggregated. As US data protection laws2 allow the use of device data for scientific purposes if de-identified, an institutional review board waiver was provided and patient informed consent was not necessary in this study.

A random sample representing 30% of the database population that began CPAP or APAP in 2015 was analysed. Of the initial patient population of 189 946, 133 006 used a PAP device ≥ 1 day with use≥ 1 hour in Week 1 and Weeks 10-13 (inclusion criteria).

The definition used for residual CSA was CAI ≥ 5/h, calculated in 1-week assessment windows. This differs from previous research because it introduces repeated measures based on real-life telemonitoring data rather than single “snapshots” of CSA.



  1. Liu et al. Trajectories of emergent central sleep apnea during continuous positive airway pressure therapy. Chest. 2017;152(4):751-60
  2. The United States Privacy Act, the Safe Harbor Act and the Health Insurance Portability and Accountability Act