Background:​ Although the detrimental effects of physical inactivity on health are welldocumented, there is inadequate knowledge of physical activity in older adults following total knee arthroplasty (TKA). Information is lacking regarding the effect of physical activity on risk for revision following TKA, the validity of tools to measure physical activity in this population, and the levels of stationary behavior in older adults following TKA.

Objectives:​ The primary aim of this research program was to evaluate tools that measure physical activity in older adults following TKA to determine which tools are most appropriate for use in this cohort. An introductory objective was to address an identified gap in the literature on factors increasing risk for revision following TKA, including the effect of physical activity on revision risk. The primary objective was to determine the concurrent criterion validity between a reference standard research accelerometer (SWA) and two clinical tools, a) the commercially available personal activity monitor (Fitbit) and b) the self-report measure (CHAMPS). The secondary objective addressed the lack of information on stationary behavior in older adults following TKA by describing levels of stationary behavior and examining whether the stepdefined sedentary lifestyle index (SLI) may be an appropriate tool to discriminate stationary behavior.

Methods:​ A scoping review was completed to identify factors that may increase the risk of revision surgery following TKA. The primary objective was addressed in the first clinical study, a cross-sectional validation study which determined the concurrent criterion validity of a consumer-level activity monitor and a self-report questionnaire as compared to a reference standard research accelerometer when measuring physical activity in 47 older adults following TKA. The secondary objective was fulfilled in the second clinical study, a cross sectional study measuring stationary time, bouts and breaks, and examining the role of SLI on stationary behavior in 65 older adults following TKA using accelerometry.

Results:​ Increased risk of revision following TKA was found to be associated with demographic factors (younger age and African American), surgical factors (uncemented components, implant malalignment and increased surgery duration) and the health service factor of low volume hospitals. Moderate to good correlations were observed between the Fitbit and SWA for steps (ICC = 0.79), energy expenditure (ICC = 0.78) and energy expenditure <3 METS (ICC = 0.79). There was poor to moderate correlation (ICC = 0.43) between the CHAMPS questionnaire and the SWA data with the CHAMPS questionnaire reporting lower daily energy expenditures than the SWA. When examining stationary behavior in this population, participants spent 80% (13.17 hours, SD 2.30) of their waking time in stationary time and had an average of 6.06 bouts of stationary time > 30 minutes per day. The SLI had significant effects on both waking stationary time (p<0.001) and number of breaks in stationary time (p<0.001).

Conclusions:​ These findings reinforce the need for further information on levels of activity in older adults following TKA and provide guidance on how best to measure physical activity. The current literature did not suggest that physical activity increases the risk for revision following TKA;​ however, further research is needed. The Fitbit is an appropriate tool to measure physical activity in older adults following TKA using step counts, energy expenditures and time spent in lower intensity activities. Caution must be used when measuring higher intensity activities. The CHAMPS questionnaire may also be an appropriate tool to use within the limitations of selfreport measures as, although poor to moderate, the correlations were higher than those reported for other self-reported questionnaires in this population. The SLI discriminated between stationary time and breaks from stationary time suggesting clinical utility to measure stationary behavior. Older adults following TKA demonstrated high levels of stationary behavior putting them at risk for adverse health effects and suggesting that the entire spectrum of activity should be measured, not just the traditional measurements of physical activity. Achieving adequate levels of activity is essential to maximize health in older adults following TKA, who are at risk for detrimental health effects due to inactivity.