Robustness and value of data from the Oura Ring for personal monitoring in extreme environments with implications for clinical use: an observational study in an unassisted ski traverse of Antarctica

Objective To evaluate the completeness and quality of Oura Ring data in a remote environment. Setting The INSPIRE22 medical research expedition, a 950 km, 47-day unassisted trek from the Ronne Ice Shelf/land junction to the South Pole. Participants Nine participants (six male) from either the UK military or civilians with experience of extreme environments and/or expedition medicine. Primary and secondary outcome measures The primary outcome measures were missing data from the Ora Ring for movement measurement during six distinct activity periods within a day and heart rate (HR) during sleep. The secondary outcome measures were the agreement between movement measurement by the Oura Ring and a wrist-worn GENEActiv device for both an average day and the activity periods. Results Overall, 3.2% of the minute-by-minute movement values were missing with a day-to-day variation of 0% to 83%. For the six activity periods, the median lowest incidence of missing values across participants was 4.44% (range 0–13.3%) during night-time and the highest 16.7% (range 6.67–60.0%) during sleep preparation. Agreement between the GENEActiv and Oura Ring measurements was activity-period dependent (r 2 =0.05–0.44). A Bland-Altman plot demonstrated a mean difference between the two devices for an average day of 0.41 (95% CI −0.45 to 1.27). For HR, the number of nights when data were available across participants ranged from 27 to 47. Within those nights, the median percentage of missing values across participants ranged from 5.99% to 18.2%. Conclusion Research and healthcare relying on wearable technology should prioritise devices providing unprocessed data for robustness.