Validity and Reliability of Wearable Technology Devices during Simulated Pickleball Game Play

Abstract

Pickleball is a popular sport. Also popular is wearable technology usage. Because the validity and reliability of wearable technology during pickleball is unknown, the purpose of this research was to evaluate the ability of common devices to return heart rate and estimated energy expenditure during pickleball activity. Twenty adult participants were outfitted with a portable metabolic unit and heart rate monitor (criterion measures). Experimental devices were a Garmin Instinct, Polar Vantage M2, Polar OH1, and Polar Verity Sense. Participants played simulated pickleball for 10 min. Validity measures included mean absolute percent error (MAPE) and Lin's Concordance Correlation Coefficient (CCC), whereas reliability measures included coefficient of variation (CV) and intraclass correlation coefficient (ICC). The heart rate returned lower than 10% MAPE across all devices (Instinct = 5.73-6.32%, Verity Sense = 2.92-2.97%, OH1 = 3.39-3.45%) and greater than 0.85 CCC (Instinct = 0.85-0.88, Verity Sense = 0.96-0.96, OH1 = 0.93-0.94). The CV was below 10% (Instinct = 9.30%, Verity Sense = 2.68%, OH1 = 5.01%), and ICC was above 0.7 (Instinct = 0.77, Verity Sense = 0.98, OH1 = 0.91). The energy expenditure MAPE was greater than 10% (Instinct = 27.67-28.08%, Vantage M2 = 18.87-23.38%) with CCC lower than 0.7 (Instinct = 0.47-0.49, Vantage M2 = 0.62-0.63). Reliability thresholds were met in the Vantage M2 (CV = 6%, ICC = 0.98) but not in the Instinct (CV = 15%, ICC = 0.86). The Instinct was neither valid nor reliable for estimated energy expenditure, while the Polar Vantage M2 was reliable but not valid. All devices returned valid and reliable heart rates during pickleball.

Keywords: Garmin Instinct; Polar Vantage M2; accuracy; consistency; paddle sport.

Figure 1

Example of criterion and experimental device setup utilized in the investigation.

Figure 2

Representative Bland–Altman plots denoting visual agreement of the heart rate (bpm) measured in wearable technology devices ((A) = Garmin Instinct, (B) = Polar Verity Sense, (C) = Polar OH1) and a criterion measure (Polar H10) in participants (N = 20) who completed 10 min of simulated pickleball gameplay. The colored bands represent the 95% CI for upper and lower limits of agreement and the bias.

Figure 3

Representative Bland–Altman plots denoting visual agreement of accumulated estimated energy expenditure (kcal) measured in wearable technology devices ((A) = Garmin Instinct, (B) = Polar Vantage M2) and a criterion measure (COSMED K5) in participants (N = 20) who completed 10 min of simulated pickleball gameplay. The colored bands represent the 95% CI for upper and lower limits of agreement and the bias.