This paper describes an R-R Interval (RRI) outlier exclusion method that can exclude RRI outliers including misdetected R waves caused by artifacts, which are frequently observed when using wearable electrocardiogram (ECG) devices. The method distinguishes targeted misdetected R waves based on statistical ECG values that can reflect the occurrence of artifacts, and annotates all detected R waves with those results for evaluating RRI measurement status. Experimental results confirmed that the proposed method is effective to improve the accuracy of both time and frequency domain measures of heart rate variability than the conventional one.