PURPOSE: Wearable technology has popularized the sport and fitness industries in recent years, making it easy for users to monitor various health-related and performance measures, including lactate threshold (LT). As devices grow in sophistication and popularity, independent validation of these devices are needed in order to determine their usefulness in driving training decisions and tracking physiological changes. Therefore, the purpose of this study is to validate the LT estimate in the Garmin fēnix 6 fitness tracker watch.
METHODS: Six apparently healthy runners (5 male, 1 female, 173.2 ± 4.2 cm, 69.3 ± 4.2 kg, 23.2 ± 2.9 kg/m^2, 38.22 ± 4.76% muscle mass, 19.4 ± 9.25% fat mass, 48.8 ± 34.3 km ran/week, all reported as mean ± standard deviation) participated in two data collection days between 2 and 7 days of each other. The first day consisted of a treadmill based graded exercise test to determine lactate threshold via blood lactate levels. The treadmill speed was increased every 3 minutes until LT was achieved. The second day consisted of an outdoor track-based run that was guided by the Garmin device. Participants wore the fēnix 6 watch on their wrist and HR monitor around their chest. The test consisted of increasing stages of intensity until the device was able to produce a LT estimate. Data analysis was performed to assess validity by determining the mean absolute percentage error (MAPE), Lin’s Concordance Correlation Coefficient (CCC), and two-tailed paired t-tests. Comparisons were made between the fēnix 6 and laboratory-based measurements, including speed at LT, speed at the onset of blood lactate accumulation (OBLA) and HR at LT. The device may be considered valid if the MAPE is <10% and CCC >0.7.
RESULTS: The Garmin fēnix 6 significantly underestimated the speed at which LT occurred compared to lab-based measurements for both LT and OBLA (p < 0.05), but had no difference in HR at LT. The MAPE was 8.38%, 10.41%, and 6.20% for the fēnix 6 compared to LT, OBLA and HR, respectively. The CCC was 0.85, 0.79, and -0.03 for LT, OBLA, and HR, respectively.
CONCLUSION: Despite the small sample size, the Garmin fēnix 6 appears to be valid in the estimate of LT when compared to lab-based measurements, although it appears to have poor agreement when comparing OBLA or HR values at LT.