Academic journal article Journal of Nursing Measurement

Longitudinal Field Comparison of the RT3 and an Activity Diary with Cardiac Patients

Academic journal article Journal of Nursing Measurement

Longitudinal Field Comparison of the RT3 and an Activity Diary with Cardiac Patients

Article excerpt

This study investigated reliability and validity of two methods of measuring patients' physical activity following coronary artery bypass graft (CABG) surgery. Sixty-five patients wore an RT3 accelerometer and recorded activity in a diary at 3 weeks, 6 weeks, and 3 months postsurgery. Generalizability coefficients greater than .80 required approximately 2 to 4 days of data, fewer than reported elsewhere. Energy expenditure estimates correlated .77, .72, and .57 (p < .01), with mean RT3 estimates higher, especially when overall energy expenditure was low. Changes from 3 to 6 weeks correlated only moderately (r = .37-.46) across methods. The methods produce reliable but differing estimates of physical activity in this population. Although no method bias is evident in assessing change over time, correlations support the importance of using multiple methods.

Keywords: physical activity; accelerometer; self-report; reliability

For patients who have undergone surgery for cardiac revascularization, increasing physical activity is crucial to recovering and maintaining health, whether or not they participate in a formal program of cardiac rehabilitation. Tracking their progress in order to monitor treatment or to research the effects of an intervention requires reliable and valid methods of evaluating patients' physical activity in free-living situations. It may not be feasible to have patients return to a standardized setting for assessment after completion of rehabilitation, especially when a surgical center draws patients from a wide geographical area.

Of the several methods available to measure physical activity, accelerometers and self-report measures such as diaries are most practical for field studies. Limited research has so far been reported for the newest triaxial accelerometers, either with respect to their reliability or to how their estimates of physical activity compare with those of self-report measures. More information is also needed concerning whether previous findings generalize to clinical populations. Although some studies have looked at sedentary populations (Coleman & Epstein, 1998; Duncan, Sydeman, & Perri, 2001) or older individuals (Fehling, Smith, Warner, & Dalsky, 1999; Washburn, Janney, & Fenster, 1990), most have drawn subjects from either the general or university population, younger and in better health than is typical for a clinical population. As Coleman and Epstein (1998) noted, longitudinal studies are needed to estimate the number of days of measurement required to reliably estimate activity when subjects are becoming more active, whether due to effects of an intervention or to a natural recovery process.

Our primary aim was to estimate the number of days needed to reach adequate levels of reliability for measures of physical activity and energy expenditure obtained from the RT3 accelerometer and a daily activity diary at three stages of postsurgical recovery from coronary artery bypass surgery. Secondary aims included comparing the estimates from the two methods at each stage and evaluating whether estimates of the amount of change over time were consistent across the two methods.


Each method of measuring activity has advantages and disadvantages for use in freeliving situations. Self-report measures, such as activity logs or diaries, are inexpensive and relatively easy to administer. They can document details of type of activity and deviations from protocol. Results from activity logs have been found reasonably stable across days, with estimates of the number of days of recording needed to attain acceptable reliability (an intraclass correlation [ICC] > .80) in an adult population varying from about 8 days (Coleman & Epstein, 1998) to 2 weeks (Baranowski et al., 1999; Eason, Mâsse, Kelder, & Tortolero, 2002). However, completing activity logs or diaries places considerable burden on respondents if activity is recorded in small increments of time for such an extended period. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.