Activity sampling in the assessment of patients with total joint arthroplasty

Quantitative assessment of patient activity is important in evaluating the outcomes of joint prostheses, and such methods are gaining popularity. The single greatest impediment to quantitative activity assessment is patient compliance. How many days of sampling are necessary to provide reliable and accurate estimates of walking activity? The current study analyzes how well sampling for 4 consecutive days of activity compares to assessing activity for 7 or more days with the same pedometer in 131 patients with either a total hip or total knee prosthesis. The mean steps per day obtained throughout the full-length sampling (7-123 days) was strongly correlated to the one obtained from the random consecutive 4-day sample (r2 = 0.94, P < .001) with only 5 outliers. The 4-day activity assessment gave an underestimation of 4.7% (P = .5). The number of outliers increased with fewer days of sampling. Monitoring activity for 4 consecutive days yields a quantitative assessment that is within 5% of a sampling of 7 or more days.