G Model
ARTICLE IN PRESS
JSAMS-1144; No. of Pages 5
Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams
Original research
Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population Bronwyn K. Clark ∗ , Toby G. Pavey, Rui F. Lim, Sjaan R. Gomersall, Wendy J. Brown The University of Queensland, Centre for Research on Exercise Physical Activity and Health, School of Human Movement Studies, Australia
a r t i c l e
i n f o
Article history: Received 19 May 2014 Received in revised form 27 January 2015 Accepted 14 February 2015 Available online xxx Keywords: Questionnaire Sitting Adults Self-report
a b s t r a c t Objectives: To assess the validity of the Past-day Adults’ Sedentary Time-University (PAST-U) questionnaire, modified for a university population, compared with activPAL. Design: Participants (n = 57, age = 18–55 years, 47% female, 65% students) were recruited from the University of Queensland (students and staff). Methods: Participants answered the PAST-U questionnaire, which asked about time spent sitting or lying down for work, study, travel, television viewing, leisure-time computer use, reading, eating, socialising and other purposes, during the previous day. Times reported for these questions were summed to provide a measure of total sedentary time. Participants also wore an activPAL device for the full day prior to completing the questionnaire and recorded their wake and sleep times in an activity log. Total waking sedentary time derived from the activPAL was used as the criterion measure. Correlation (intraclass correlation coefficient, ICC) and agreement (Bland–Altman plots) between PAST-U and activPAL sedentary time were examined. Results: Participants were sedentary (activPAL-determined) for 66% of waking hours. The correlation between PAST-U and activPAL sedentary time for the whole sample was ICC = 0.64 [95% confidence interval (CI) = 0.45, 0.77]; and higher for non-students (ICC = 0.78, 95%CI 0.52, 0.91) than students (ICC = 0.59, 95%CI 0.33, 0.77). Bland-Altman plots revealed that the mean difference between the two measures was 5 min although limits of agreement were wide (95% limits of agreement: −3.9 to 4.1 h). Conclusions: The PAST-U provides an acceptable measure of sedentary time in this population, which included students and adults with high workplace sitting time. © 2015 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
1. Background Time spent in sedentary behaviour (too much sitting as distinct from too little exercise) has been detrimentally associated with several health outcomes,1,2 and with premature mortality.3–5 Sedentary behaviours are defined as activities with an energy expenditure of less than 1.5 metabolic equivalents (METs) spent in a sitting or reclining position while awake.6 Sitting time, in particular, has been the target for contemporary research as adults are spending increasing amounts of time sitting each day in work, transport and leisure activities.7–9 Methods used to measure sedentary behaviour include the use of device-based measures, such as accelerometers and inclinometers, and self-report measures, such as questionnaires and 24-h recalls.10 While device-based measures provide a measure of posture and motion with good accuracy,10 they do not provide
∗ Corresponding author. E-mail address: [email protected] (B.K. Clark).
information on sedentary time in specific behaviours or contexts, and are often too expensive for large-scale studies. In contrast, selfreport questionnaires provide a low-cost and easy to use alternative for measuring sedentary behaviours, and are able to capture the type (e.g. computer use) and context (e.g. at work) of behaviour.10 These qualities make questionnaires a desirable tool for monitoring sedentary time in population health studies. However, self-report measures are susceptible to random error (i.e. inaccurate reporting) and systematic bias (e.g. social desirability).11,12 Thus, it is necessary to use high quality self-report measures to complement the information from device-based measures, in order to provide a more robust estimate of sedentary time,10 or to provide the best alternative when cost is an issue. Most commonly used sedentary behaviour questionnaires employ recall of past week or usual behaviour; however, these generally show poor validity against device-based measures. Recall of the past day may improve the accuracy of these measures.10 A recent study of the Past-day Adults’ Sedentary Time (PAST) questionnaire found it to be a valid, reliable and easily administered self-report measure.13 However, the study was conducted with a
http://dx.doi.org/10.1016/j.jsams.2015.02.001 1440-2440/© 2015 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model JSAMS-1144; No. of Pages 5
ARTICLE IN PRESS B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
2
female population diagnosed with breast cancer, a very specific group who may already be aware of their health habits. University communities include a wide range of participants, including students, research and administrative staff, who engage in activities that involve prolonged sitting.14,15 In view of the typically high sitting times in this population and specific concerns about weight gain in university students,16 the development of a self-report sedentary time measure in a university setting would provide a useful tool for future studies. The aim of this study is, therefore, to modify the PAST questionnaire for use in a university population and assess the validity of the modified version. 2. Methods The PAST questionnaire used in this study (see Appendix A) was a modification of the questionnaire used in the study by Clark and colleagues,13 which asks about sitting and lying down while awake on the previous day. The previous version of the PAST demonstrated acceptable reliability (intraclass correlation coefficient: 0.50, 95% confidence interval [CI] 0.32–0.64) and good validity (r = 0.57; 95% CI = 0.39–0.71 against activPAL) for measurement of sedentary behaviour in a population of women with a history of breast cancer.13 The modification process involved (1) discussion between the key researchers on the inclusion of new questions; and (2) interviews with university students and staff to give feedback on the questionnaire on two occasions. Students and staff who gave feedback on the questionnaire were recruited by word of mouth and were known to the research team; however, were not part of the team conducting this project. The revised version of the PAST had nine questions (Appendix A) about sedentary time for work, study, travelling, eating and drinking, watching television, using the computer, socialising and other daily activities. As it was modified for a university population it was named PAST-U. Times reported for each of the domains/behaviours were summed to create a total of self-reported time spent sitting and lying down (termed PAST-U sedentary time). Participants were recruited from the University of Queensland by convenience sampling, including word of mouth and an online university newsletter. Those who showed interest received an information sheet explaining the study and the eligibility criteria; and an invitation to join the study via email. to be eligible for the study participants had to be over the age of 18 years, healthy and ambulatory. Eligible participants provided written informed consent prior to enrolling in the study and in return for their participation received a $20 cash gratuity. Ethical clearance was obtained from the Medical Research Ethics Committee of the University of Queensland. The protocol took place over three consecutive days from August to October 2013. The majority of data collection was completed during term time with regular lecture scheduling, although 11 participants provided data during mid-semester break (no regular lectures but continuing study and assignment commitments). Day 1 involved a visit to the Human Movement Studies department. At that session, participants self-completed a questionnaire to provide demographic information including age, gender, marital status and education. Instruction on activPAL inclinometer use was then provided before the device was attached. Participants were instructed to wear the activPAL continuously until the next visit on Day 3, allowing complete 24 h wear on Day 2. Participants were also provided with an activity log to record their waking and sleeping times for Day 2 and any periods the activPAL device was removed. The second visit took place on Day 3. The activPAL and the completed activity log were collected and the participants completed the modified PAST questionnaire, which was interviewer administered. As Day 3 was scheduled for either Wednesday or Friday, the recall of the previous day was always a weekday.
The activPAL device (Version 3, Pal Technologies Ltd, Glasgow, UK) is a thigh-worn inclinometer accelerometer, which continuously records posture and movement (time spent sitting/lying, standing or stepping). The device was sealed with a nitrile finger cot and a layer of opsite and attached to the skin with a transparent film (TegadermTM Roll, 3MTM ) in order to provide a waterproof barrier. The attachment was made to the right thigh (midline on the anterior aspect), which was prepared with an alcohol swab and a patch of hypafix to minimise skin irritation. The activPALs were initialized (default settings used) and data were downloaded using activPalTM Professional Software, v6.1.2 Research Edition (Pal Technologies Ltd, 2010). Estimates for time spent sitting/lying were derived from the event file, which includes time intervals per day in seconds. Wake and sleep times recorded in the activity log were used to extract activPAL data for the times participants were awake on Day 2 and continuing to bed time if past midnight on Day 2. Participants’ activPAL data were considered valid if they reported wearing the device for all waking hours with less than 30 min removal. The time spent sitting or lying down recorded by the device while awake was termed PAL sedentary time. While not strictly a gold standard measure of sitting time, the activPAL has shown good agreement with direct observation of sitting time (99% agreement17 ; and mean bias of 7.7 min per day18 and 0.19%19 ) and so was used as the criterion measure for assessing the validity of the PAST questionnaire. Data were analysed using SPSS version 21 (IBM Corporation) with statistical significance set at p < 0.05. Findings are presented for the overall sample; and for students and non-students separately to provide information for researchers undertaking research in solely student or non-student working populations. Descriptive statistics (N, mean, standard deviation [SD], median and interquartile range [IQR]) were used to describe the characteristics of the sample. Intraclass correlation coefficients (ICC) with 95% confidence intervals (CI) were used to determine the correlation between PAST-U sedentary time and PAL sedentary time; with the addition the of Pearson’s correlation coefficients (r) so as to compare to existing validity studies for sedentary behaviour questionnaires. Bland-Altman plots20 were used to examine differences between PAST-U–PAL sedentary time and the average of the two measures, with mean difference (MD) and 95% limits of agreement (LoA; ±1.96 SD) reported. Statistical difference between sedentary times from the PAST-U and the activPAL were examined using Student’s t-tests. 3. Results Modifications made for the university version of the PAST questionnaire were the addition of questions on sitting time for study, for socialising and for meals and the incorporation of sitting for hobbies into the ‘all other sitting’ question. These modifications were based on consultation with the research team (addition of a question on sitting for study); and two sets of interviews with students and staff. At the first set of interviews, seven university students provided critical feedback. The main recommendations were to have ‘time spent sitting while socialising’ as a standalone question and to have ‘time spent on hobbies’ combined into the ‘all other activities’ question. On the second occasion, feedback from six university students and two members of staff was received. An additional suggestion was to have ‘time spent sitting for meal times’ as a standalone question, and to have examples for each question so as to be clear on the definition of activities in the various questions (e.g. paid work would include activities such as babysitting, administrative/clerical work; use of computer for leisure would include activities such as Facebook, YouTube, Skype or online shopping). Fifty-eight participants consented to take part in the study (38 students and 20 staff). All 58 participants wore the activPAL device
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model JSAMS-1144; No. of Pages 5
ARTICLE IN PRESS B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
3
Table 1 Attributes of the study sample.
Age, median (interquartile range) (years) Female, n (%) Body mass index, mean ± SD (kg m−2 ) Education: completed post-high school education, n (%) Awake time, mean ± SD (h) Sedentary time, activPAL mean ± SD (h)
Overall (n = 57)
Student (n = 37)
26 (23, 31) 27 (47%) 23.82 (3.05) 52 (91%) 16.09 (1.90) 10.65 (2.42)
24 (22, 27) 17 (46%) 23.42 (3.11) 34 (92%) 16.09 (2.03) 10.76 (2.46)
Non-student (n = 20) 33 (30, 36) 10 (50%) 24.56 (2.87) 18 (90%) 16.11 (1.68) 10.43 (2.03)
Time awake and sedentary is in hours on the day recalled. Standard deviation (SD).
for all waking hours on the day recalled and completed the PASTU questionnaire. For one participant the activPAL did not record all waking hours on the day of wear so their data were excluded from the analysis. The characteristics of the 57 participants who were included in this analysis are presented in Table 1. Most participants had completed post-high school education. There were close to even numbers of men and women. Participants were awake for 16.1 h (SD: 1.9) on average on the day of monitoring, and of this time an average of 10.7 h (SD: 2.4) was recorded as sedentary time on the activPAL (66% of awake time). The student group was younger than the non-student group but there were no other differences between the groups in terms of demographic characteristics or activPAL measures of sedentary time (Table 1). Table 2 shows the total sedentary time reported in the PAST-U questionnaires overall and for each group (students and nonstudents). Participants reported on average over 10 h of sitting or lying while awake (10.72 ± 2.42 h). Mean PAST-U sedentary time was similar in the student group (10.74 ± 2.65 h) and the nonstudent group (10.70 ± 1.98 h). Table 2 also shows the reported sitting times for the individual items in the PAST-U questionnaire. Highest median time reported for the whole sample was sitting for study, followed by television viewing, using the computer and eating. The lowest reported times were for reading as only 10 participants reported any reading for leisure time. The most frequently reported item was sitting for eating (56 participants) followed by leisure time computer use (46 participants). For the student population, highest median sitting time for an individual item was sitting for study, followed by using the computer and then eating; and for non-students highest median sitting time was sitting for work followed by television viewing. There was a moderate correlation between PAST-U sedentary time and PAL sedentary time (ICC = 0.64, 95%CI 0.45, 0.77; r = 0.63, 95%CI = 0.44–0.76). A similar correlation was found for the student population (ICC = 0.59, 95%CI 0.33, 0.77; r = 0.58, 95%CI = 0.32–0.76), though the strength of this correlation was lower than for the non-student population (ICC = 0.78, 95%CI 0.52, 0.91; r = 0.78, 95%CI = 0.51–0.91). The Bland-Altman plot showing agreement between PAST-U and PAL sedentary time for the total sample is shown in Fig. 1. Differences between sedentary time on the PAST-U and activPAL were normally distributed. The mean difference between the two measures was small (0.08 h, SD 2.04 h), indicating that the PAST-U questionnaire produced estimates of sedentary time that were about 5 min more than the sedentary time recorded by the activPAL. However, 95% limits of agreement were wide (−3.92 to 4.07 h), indicating that differences between measures for some individuals were large. Most participants (84%) were within two hours difference between PAST-U and PAL sedentary time. When the agreement was assessed separately for students (−0.02 h, SD 2.35 h) and non-students (0.26 h, SD 1.34 h) mean differences were similar; however, limits of agreement were wide, although greater for students (−4.63 to 5.59 h) than non-students (−2.37 to 2.89 h) (plots not shown). T-tests showed that differences between the PAST-U sedentary time and PAL sedentary time
for the whole sample or for the students and non-students were non-significant. 4. Discussion The aim of this study was to modify the PAST questionnaire, recalling sedentary time on the previous day, to suit a university population and validate it against a device-based measure. The findings suggest that the PAST-U has acceptable levels of validity when compared to sedentary time from the activPAL, in terms of correlation (ICC = 0.64) and agreement (mean difference: 5 min) at the group level; however, estimates were poor at an individual level, as suggested by wide limits of agreement. This questionnaire may, therefore, be most appropriate for use in large-scale studies rather than for studies requiring estimates of an individual’s sedentary time. The strength of the correlation found between the PAST-U and activPAL derived sedentary time was similar to that found in the original validity study of the PAST questionnaire (r = 0.57).13 In comparison with another study which used an activPAL as criterion measure, the correlation for the PAST-U was higher than for the International Physical Activity Questionnaire (IPAQ) on weekdays (r = 0.41) and higher than the Domain Specific Questionnaire for weekdays (r = 0.30).21 Although there is no recommended standard for the acceptable strength of correlations between sedentary behaviour questionnaires and criterion measures, physical activity researchers have set a standard of r > 0.5 (against the accelerometer as criterion) for questionnaires measuring moderate and vigorous physical activity.22 Our findings show correlations above this level for the whole sample and for both the student and non-student subsamples. Very few other studies have used activPAL devices as the criterion measure for self-report measures of sedentary time. One other study that used this criterion measure examined the validity of an interviewer-administered, web-based time-use 24-h Physical Activity Recall. When sedentary time from the 24-h recall was validated against the activPAL, correlations were higher (men, = 0.81; women, = 0.81), than found for the university version of the PAST.23 In terms of agreement, the mean difference (men, MD = 0.72 h, women, MD = 0.75 h) was greater than for the PASTU, but the limits of agreement were smaller (95% LoA men = −2.61 to 4.05; women, −2.21 to 3.71).23 The stronger correlations and smaller limits of agreement for the 24-h recall suggest that it may provide better estimates of sitting time at the individual level than the PAST-U. However, the longer time spent completing such detailed recalls puts a higher burden on participants compared to questionnaires, which also provide information on the time spent in specific sedentary behaviours and can be completed in a shorter time (the original PAST questionnaire was completed in a median time of 7 min)13 . Given the lower participant burden and greater accuracy at the group level (evidenced by a small mean difference on the Bland Altman Plots) the PAST-U may be more suited to large scale surveillance studies. The PAST-U provides information on sedentary time in specific contexts for both students and staff. For students, the most
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model
ARTICLE IN PRESS
JSAMS-1144; No. of Pages 5
B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
4
Table 2 Sedentary time (h) reported on the PAST-U items.
Study Work Transport TV viewing Computer/electronic device use Reading Eating Socializing Others Composite sedentary time PAST, mean ± SD
Overall (n = 57)
Student (n = 37)
Non-student (n = 20)
2.50 (0, 5.50) 0 (0, 6.00) 0.67 (0.13, 1.08) 1.00 (0, 2.00) 1.00 (0.33, 1.75) 0 (0, 0) 1.00 (0.50, 1.50) 0.17 (0, 1.0) 0 (0, 0.50) 10.72 (2.42)
4.00 (2.00, 6.50) 0 (0, 0.29) 0.50 (0.04, 1.00) 1.00 (0, 2.00) 1.00 (0.50, 2.00) 0 (0, 0) 1.00 (0.58, 1.50) 0.33 (0, 1.00) 0 (0, 0.50) 10.74 (2.65)
0 (0, 0) 6.00 (5.13, 7.00) 0.88 (0.50, 1.23) 1.75 (0.06, 2.00) 0.50 (0, 1.00) 0 (0, 0.19) 0.75 (0.38, 1.00) 0.08 (0, 0.50) 0 (0, 0.69) 10.70 (1.98)
Data for PAST-U sedentary times are presented as median (interquartile range) except where indicated. Standard deviation (SD).
prevalent behaviour was studying, followed by using the computer and electronic devices. The finding that studying was the primary purpose of sedentary time in students is unsurprising and has been found previously in another study on sedentary behaviour patterns of university students.14 Long hours of sitting for work by non-student participants (median 5.53 h) was also expected and similar to findings for working participants in the previous study of the PAST13 and other studies of occupational sitting.24–26 Although absolute agreement between the measures was similar in the student and non-student population, correlation was better for the non-students suggesting this measure may be useful in working populations with high sedentary time. Participants in this study were sedentary for 66% of their waking hours which is a similar finding to other studies that have measured sedentary time using the activPAL in office workers (67% waking hours sedentary21 and 67% of working hours sedentary26 ) but more than the proportion of sedentary time found in the previous study on the PAST in breast cancer survivors (58%).13 The university population, therefore, appears to represent a population that experiences high sedentary time, similar to office workers. The use of the activPAL device as a criterion measure for the PAST-U questionnaire is a major strength of the study. The activPAL has shown high agreement with direct observation of sitting17–19 and is, therefore, a good criterion for evaluating correlation and agreement in self-report measures of sedentary time. The inclusion of multiple items assessing sedentary time in various contexts is also a strength. Composite measures of sitting time have been shown to provide better correlations with device-based measures
of sedentary time than single item sitting time questions.10,18,27 However, one of the issues with composite measures compared to single item sitting time questions is increased participant burden. A suggestion to further refine the PAST-U questionnaire is to remove or combine certain questions with very low level responses (e.g. reading), thus shortening the completion time. One limitation of this study is that the estimates of sedentary time spent in individual behaviours or domains may not perfectly reflect the proportion of time spent sedentary for each purpose, due to the possibility of multi-tasking and overlapping of time. For example, the component item ‘television viewing’ was asked earlier in the questionnaire, so the time reported could have included sedentary time from other components experienced simultaneously such as time spent eating, because participants were told not to report time for subsequent components if they had already reported it. This could affect the true value of the sedentary times for some components but does not affect estimates of total sedentary time. Another limitation is the single administration of the PAST-U. Behaviours and schedules vary daily and usual time spent sedentary may not be accurately assessed by a single pastday recall. Assessing multiple administrations of the PAST-U over several weekdays and weekend days is an area for improvement in future studies. The final limitation of this study is the sample. As expected in a University sample, the participants had all completed high school education. For broader use of the PAST-U questionnaire validity should also be assessed in other populations including those with lower levels of education. Participants were also possibly more
Fig. 1. Bland–Altman plot for overall PAST-U sedentary time and PAL sedentary time. The solid line represents the mean difference (MD) between the two measures and the dashed lines are the 95% LoA. Students are represented by solid grey diamonds (
), non-students are clear diamonds (
).
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model JSAMS-1144; No. of Pages 5
ARTICLE IN PRESS B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
aware of their sitting time as they had been informed the study would measure sitting time. 5. Conclusion The PAST-U questionnaire shows acceptable validity for estimating sedentary time in a student and working population and can provide useful contextual information about sedentary time in different domains. This supports earlier research that past-day recall of sedentary time provides a viable alternative to commonly used recall of past week or usual behaviour. The PAST-U shows promise for use in population surveillance studies and to complement device-based measures of sedentary time. Practical implications • The PAST-U questionnaire shows validity properties that are similar to the original version. • The PAST-U shows acceptable validity for estimating sedentary time in student and working populations at a group level. • The accuracy of the PAST-U is not suitable for measurement of individual’s sedentary time. Acknowledgement The study was funded by a University of Queensland (UQ) Startup Grant (UQ no. 2012002122). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jsams.2015.02.001. References 1. Proper K, Singh A, van Mechelen W et al. Sedentary behaviors and health outcomes among adults: a systematic review of prospective studies. Am J Prev Med 2011; 40:174–182. 2. Williams DM, Raynor HA, Ciccolo JT. A review of TV viewing and its association with health outcomes in adults. Am J Lifestyle Med 2008; 2(3):250–259. 3. Chau JY, Grunseit AC, Chey T et al. Daily sitting time and all-cause mortality: a meta-analysis. PLoS ONE 2013; 8(11):e80000. 4. Dunstan DW, Barr ELM, Healy GN et al. Television viewing time and mortality: the Australian diabetes, obesity and lifestyle study (AusDiab). Circulation 2010; 121:384–391. 5. Pavey TG, Peeters GG, Brown WJ. Sitting-time and 9-year all-cause mortality in older women. Br J Sports Med 2015; 49:95–99.
5
6. Sedentary Behaviour Research Network. Standardized use of the terms “sedentary” and “sedentary, behaviours”. Appl Physiol Nutr Metab 2012; 37: 540–542. 7. Owen N, Bauman A, Brown W. Too much sitting: a novel and important predictor of chronic disease risk? Br J Sports Med 2009; 43(2):81–83. 8. Chau J, Merom D, Grunseit A et al. Temporal trends in non-occupational sedentary behaviours from Australian time use surveys 1992, 1997 and 2006. Int J Behav Nutr Phys Act 2012; 9(1):76. 9. Church T, Thomas D, Tudor-Locke C et al. Trends over 5 Decades in U.S. occupation-related physical activity and their associations with obesity. PLoS ONE 2011; 6:e19657. 10. Healy GN, Clark BK, Winkler AE et al. Measurement of adults’ sedentary time in population-based studies. Am J Prev Med 2011; 41(2):216–227. 11. Affuso O, Stevens J, Catellier D et al. Validity of self-reported leisure-time sedentary behavior in adolescents. J Negat Results Biomed 2011; 10(1):2. 12. Bassett Jr DR, Fitzhugh EC. Establishing validity and reliability of physical activity assessment instruments. 3, in Epidemiological methods in physical activity studies, Lee IM, editor, New York, NY, Oxford University Press, 2009. 13. Clark BK, Winkler E, Healy GN et al. Adults’ past-day recall of sedentary time: reliability, validity, and responsiveness. Med Sci Sports Exerc 2013; 45(6):1198–1207. 14. Rouse PC, Biddle SJH. An ecological momentary assessment of the physical activity and sedentary behaviour patterns of university students. Health Educ J 2010; 69(1):116–125. 15. Alkhajah TA, Reeves MM, Eakin EG et al. Sit-stand workstations: a pilot intervention to reduce office sitting time. Am J Prev Med 2012; 43(3): 298–303. 16. Vella-Zarb RA, Elgar FJ. The ‘freshman 5’: a meta-analysis of weight gain in the freshman year of college. J Am Coll Health 2009; 58(2):161–166. 17. Ryde GC, Gilson ND, Suppini A et al. Validation of a novel, objective measure of occupational sitting. J Occup Health 2012; 54(5):383–386. 18. Kozey-Keadle S, Libertine A, Lyden K et al. Validation of wearable monitors for assessing sedentary behavior. Med Sci Sports Exerc 2011; 43(8):1561–1567. 19. Grant P, Ryan C, Tigbe W et al. The validation of a novel activity monitor in the measurement of posture and motion during everyday activities. Br J Sports Med 2006; 40(12):992–997. 20. Bland JM, Altman GA. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986:307–311. 21. Kozey-Keadle S, Libertine A, Staudenmayer J et al. The feasibility of reducing and measuring sedentary time among overweight, non-exercising office workers. J Obes 2012:1–10. 22. van Poppel MN, Chinapaw MJ, Mokkink LB et al. Physical activity questionnaires for adults: a systematic review of measurement properties. Sports Med 2010; 40(7):565–600. 23. Matthews CE, Keadle SK, Sampson J et al. Validation of a previous-day recall measure of active and sedentary behaviors. Med Sci Sports Exerc 2013; 45(8):1629–1638. 24. Healy GN, Eakin EG, Lamontagne AD et al. Reducing sitting time in office workers: short-term efficacy of a multicomponent intervention. Prev Med 2013; 57(1):43–48. 25. Thorp AA, Healy GN, Winkler E et al. Prolonged sedentary time and physical activity in workplace and non-work contexts: a cross-sectional study of office, customer service and call centre employees. Int J Behav Nutr Phys Act 2012; 9:128. 26. Ryde GC, Brown HE, Peeters GM et al. Desk-based occupational sitting patterns: weight-related health outcomes. Am J Prev Med 2013; 45(4):448–452. 27. Clemes SA, David BM, Zhao Y et al. Validity of two self-report measures of sitting time. J Phys Act Health 2012; 9:533–539.
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
ARTICLE IN PRESS
JSAMS-1144; No. of Pages 5
Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Science and Medicine in Sport journal homepage: www.elsevier.com/locate/jsams
Original research
Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population Bronwyn K. Clark ∗ , Toby G. Pavey, Rui F. Lim, Sjaan R. Gomersall, Wendy J. Brown The University of Queensland, Centre for Research on Exercise Physical Activity and Health, School of Human Movement Studies, Australia
a r t i c l e
i n f o
Article history: Received 19 May 2014 Received in revised form 27 January 2015 Accepted 14 February 2015 Available online xxx Keywords: Questionnaire Sitting Adults Self-report
a b s t r a c t Objectives: To assess the validity of the Past-day Adults’ Sedentary Time-University (PAST-U) questionnaire, modified for a university population, compared with activPAL. Design: Participants (n = 57, age = 18–55 years, 47% female, 65% students) were recruited from the University of Queensland (students and staff). Methods: Participants answered the PAST-U questionnaire, which asked about time spent sitting or lying down for work, study, travel, television viewing, leisure-time computer use, reading, eating, socialising and other purposes, during the previous day. Times reported for these questions were summed to provide a measure of total sedentary time. Participants also wore an activPAL device for the full day prior to completing the questionnaire and recorded their wake and sleep times in an activity log. Total waking sedentary time derived from the activPAL was used as the criterion measure. Correlation (intraclass correlation coefficient, ICC) and agreement (Bland–Altman plots) between PAST-U and activPAL sedentary time were examined. Results: Participants were sedentary (activPAL-determined) for 66% of waking hours. The correlation between PAST-U and activPAL sedentary time for the whole sample was ICC = 0.64 [95% confidence interval (CI) = 0.45, 0.77]; and higher for non-students (ICC = 0.78, 95%CI 0.52, 0.91) than students (ICC = 0.59, 95%CI 0.33, 0.77). Bland-Altman plots revealed that the mean difference between the two measures was 5 min although limits of agreement were wide (95% limits of agreement: −3.9 to 4.1 h). Conclusions: The PAST-U provides an acceptable measure of sedentary time in this population, which included students and adults with high workplace sitting time. © 2015 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
1. Background Time spent in sedentary behaviour (too much sitting as distinct from too little exercise) has been detrimentally associated with several health outcomes,1,2 and with premature mortality.3–5 Sedentary behaviours are defined as activities with an energy expenditure of less than 1.5 metabolic equivalents (METs) spent in a sitting or reclining position while awake.6 Sitting time, in particular, has been the target for contemporary research as adults are spending increasing amounts of time sitting each day in work, transport and leisure activities.7–9 Methods used to measure sedentary behaviour include the use of device-based measures, such as accelerometers and inclinometers, and self-report measures, such as questionnaires and 24-h recalls.10 While device-based measures provide a measure of posture and motion with good accuracy,10 they do not provide
∗ Corresponding author. E-mail address: [email protected] (B.K. Clark).
information on sedentary time in specific behaviours or contexts, and are often too expensive for large-scale studies. In contrast, selfreport questionnaires provide a low-cost and easy to use alternative for measuring sedentary behaviours, and are able to capture the type (e.g. computer use) and context (e.g. at work) of behaviour.10 These qualities make questionnaires a desirable tool for monitoring sedentary time in population health studies. However, self-report measures are susceptible to random error (i.e. inaccurate reporting) and systematic bias (e.g. social desirability).11,12 Thus, it is necessary to use high quality self-report measures to complement the information from device-based measures, in order to provide a more robust estimate of sedentary time,10 or to provide the best alternative when cost is an issue. Most commonly used sedentary behaviour questionnaires employ recall of past week or usual behaviour; however, these generally show poor validity against device-based measures. Recall of the past day may improve the accuracy of these measures.10 A recent study of the Past-day Adults’ Sedentary Time (PAST) questionnaire found it to be a valid, reliable and easily administered self-report measure.13 However, the study was conducted with a
http://dx.doi.org/10.1016/j.jsams.2015.02.001 1440-2440/© 2015 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model JSAMS-1144; No. of Pages 5
ARTICLE IN PRESS B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
2
female population diagnosed with breast cancer, a very specific group who may already be aware of their health habits. University communities include a wide range of participants, including students, research and administrative staff, who engage in activities that involve prolonged sitting.14,15 In view of the typically high sitting times in this population and specific concerns about weight gain in university students,16 the development of a self-report sedentary time measure in a university setting would provide a useful tool for future studies. The aim of this study is, therefore, to modify the PAST questionnaire for use in a university population and assess the validity of the modified version. 2. Methods The PAST questionnaire used in this study (see Appendix A) was a modification of the questionnaire used in the study by Clark and colleagues,13 which asks about sitting and lying down while awake on the previous day. The previous version of the PAST demonstrated acceptable reliability (intraclass correlation coefficient: 0.50, 95% confidence interval [CI] 0.32–0.64) and good validity (r = 0.57; 95% CI = 0.39–0.71 against activPAL) for measurement of sedentary behaviour in a population of women with a history of breast cancer.13 The modification process involved (1) discussion between the key researchers on the inclusion of new questions; and (2) interviews with university students and staff to give feedback on the questionnaire on two occasions. Students and staff who gave feedback on the questionnaire were recruited by word of mouth and were known to the research team; however, were not part of the team conducting this project. The revised version of the PAST had nine questions (Appendix A) about sedentary time for work, study, travelling, eating and drinking, watching television, using the computer, socialising and other daily activities. As it was modified for a university population it was named PAST-U. Times reported for each of the domains/behaviours were summed to create a total of self-reported time spent sitting and lying down (termed PAST-U sedentary time). Participants were recruited from the University of Queensland by convenience sampling, including word of mouth and an online university newsletter. Those who showed interest received an information sheet explaining the study and the eligibility criteria; and an invitation to join the study via email. to be eligible for the study participants had to be over the age of 18 years, healthy and ambulatory. Eligible participants provided written informed consent prior to enrolling in the study and in return for their participation received a $20 cash gratuity. Ethical clearance was obtained from the Medical Research Ethics Committee of the University of Queensland. The protocol took place over three consecutive days from August to October 2013. The majority of data collection was completed during term time with regular lecture scheduling, although 11 participants provided data during mid-semester break (no regular lectures but continuing study and assignment commitments). Day 1 involved a visit to the Human Movement Studies department. At that session, participants self-completed a questionnaire to provide demographic information including age, gender, marital status and education. Instruction on activPAL inclinometer use was then provided before the device was attached. Participants were instructed to wear the activPAL continuously until the next visit on Day 3, allowing complete 24 h wear on Day 2. Participants were also provided with an activity log to record their waking and sleeping times for Day 2 and any periods the activPAL device was removed. The second visit took place on Day 3. The activPAL and the completed activity log were collected and the participants completed the modified PAST questionnaire, which was interviewer administered. As Day 3 was scheduled for either Wednesday or Friday, the recall of the previous day was always a weekday.
The activPAL device (Version 3, Pal Technologies Ltd, Glasgow, UK) is a thigh-worn inclinometer accelerometer, which continuously records posture and movement (time spent sitting/lying, standing or stepping). The device was sealed with a nitrile finger cot and a layer of opsite and attached to the skin with a transparent film (TegadermTM Roll, 3MTM ) in order to provide a waterproof barrier. The attachment was made to the right thigh (midline on the anterior aspect), which was prepared with an alcohol swab and a patch of hypafix to minimise skin irritation. The activPALs were initialized (default settings used) and data were downloaded using activPalTM Professional Software, v6.1.2 Research Edition (Pal Technologies Ltd, 2010). Estimates for time spent sitting/lying were derived from the event file, which includes time intervals per day in seconds. Wake and sleep times recorded in the activity log were used to extract activPAL data for the times participants were awake on Day 2 and continuing to bed time if past midnight on Day 2. Participants’ activPAL data were considered valid if they reported wearing the device for all waking hours with less than 30 min removal. The time spent sitting or lying down recorded by the device while awake was termed PAL sedentary time. While not strictly a gold standard measure of sitting time, the activPAL has shown good agreement with direct observation of sitting time (99% agreement17 ; and mean bias of 7.7 min per day18 and 0.19%19 ) and so was used as the criterion measure for assessing the validity of the PAST questionnaire. Data were analysed using SPSS version 21 (IBM Corporation) with statistical significance set at p < 0.05. Findings are presented for the overall sample; and for students and non-students separately to provide information for researchers undertaking research in solely student or non-student working populations. Descriptive statistics (N, mean, standard deviation [SD], median and interquartile range [IQR]) were used to describe the characteristics of the sample. Intraclass correlation coefficients (ICC) with 95% confidence intervals (CI) were used to determine the correlation between PAST-U sedentary time and PAL sedentary time; with the addition the of Pearson’s correlation coefficients (r) so as to compare to existing validity studies for sedentary behaviour questionnaires. Bland-Altman plots20 were used to examine differences between PAST-U–PAL sedentary time and the average of the two measures, with mean difference (MD) and 95% limits of agreement (LoA; ±1.96 SD) reported. Statistical difference between sedentary times from the PAST-U and the activPAL were examined using Student’s t-tests. 3. Results Modifications made for the university version of the PAST questionnaire were the addition of questions on sitting time for study, for socialising and for meals and the incorporation of sitting for hobbies into the ‘all other sitting’ question. These modifications were based on consultation with the research team (addition of a question on sitting for study); and two sets of interviews with students and staff. At the first set of interviews, seven university students provided critical feedback. The main recommendations were to have ‘time spent sitting while socialising’ as a standalone question and to have ‘time spent on hobbies’ combined into the ‘all other activities’ question. On the second occasion, feedback from six university students and two members of staff was received. An additional suggestion was to have ‘time spent sitting for meal times’ as a standalone question, and to have examples for each question so as to be clear on the definition of activities in the various questions (e.g. paid work would include activities such as babysitting, administrative/clerical work; use of computer for leisure would include activities such as Facebook, YouTube, Skype or online shopping). Fifty-eight participants consented to take part in the study (38 students and 20 staff). All 58 participants wore the activPAL device
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model JSAMS-1144; No. of Pages 5
ARTICLE IN PRESS B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
3
Table 1 Attributes of the study sample.
Age, median (interquartile range) (years) Female, n (%) Body mass index, mean ± SD (kg m−2 ) Education: completed post-high school education, n (%) Awake time, mean ± SD (h) Sedentary time, activPAL mean ± SD (h)
Overall (n = 57)
Student (n = 37)
26 (23, 31) 27 (47%) 23.82 (3.05) 52 (91%) 16.09 (1.90) 10.65 (2.42)
24 (22, 27) 17 (46%) 23.42 (3.11) 34 (92%) 16.09 (2.03) 10.76 (2.46)
Non-student (n = 20) 33 (30, 36) 10 (50%) 24.56 (2.87) 18 (90%) 16.11 (1.68) 10.43 (2.03)
Time awake and sedentary is in hours on the day recalled. Standard deviation (SD).
for all waking hours on the day recalled and completed the PASTU questionnaire. For one participant the activPAL did not record all waking hours on the day of wear so their data were excluded from the analysis. The characteristics of the 57 participants who were included in this analysis are presented in Table 1. Most participants had completed post-high school education. There were close to even numbers of men and women. Participants were awake for 16.1 h (SD: 1.9) on average on the day of monitoring, and of this time an average of 10.7 h (SD: 2.4) was recorded as sedentary time on the activPAL (66% of awake time). The student group was younger than the non-student group but there were no other differences between the groups in terms of demographic characteristics or activPAL measures of sedentary time (Table 1). Table 2 shows the total sedentary time reported in the PAST-U questionnaires overall and for each group (students and nonstudents). Participants reported on average over 10 h of sitting or lying while awake (10.72 ± 2.42 h). Mean PAST-U sedentary time was similar in the student group (10.74 ± 2.65 h) and the nonstudent group (10.70 ± 1.98 h). Table 2 also shows the reported sitting times for the individual items in the PAST-U questionnaire. Highest median time reported for the whole sample was sitting for study, followed by television viewing, using the computer and eating. The lowest reported times were for reading as only 10 participants reported any reading for leisure time. The most frequently reported item was sitting for eating (56 participants) followed by leisure time computer use (46 participants). For the student population, highest median sitting time for an individual item was sitting for study, followed by using the computer and then eating; and for non-students highest median sitting time was sitting for work followed by television viewing. There was a moderate correlation between PAST-U sedentary time and PAL sedentary time (ICC = 0.64, 95%CI 0.45, 0.77; r = 0.63, 95%CI = 0.44–0.76). A similar correlation was found for the student population (ICC = 0.59, 95%CI 0.33, 0.77; r = 0.58, 95%CI = 0.32–0.76), though the strength of this correlation was lower than for the non-student population (ICC = 0.78, 95%CI 0.52, 0.91; r = 0.78, 95%CI = 0.51–0.91). The Bland-Altman plot showing agreement between PAST-U and PAL sedentary time for the total sample is shown in Fig. 1. Differences between sedentary time on the PAST-U and activPAL were normally distributed. The mean difference between the two measures was small (0.08 h, SD 2.04 h), indicating that the PAST-U questionnaire produced estimates of sedentary time that were about 5 min more than the sedentary time recorded by the activPAL. However, 95% limits of agreement were wide (−3.92 to 4.07 h), indicating that differences between measures for some individuals were large. Most participants (84%) were within two hours difference between PAST-U and PAL sedentary time. When the agreement was assessed separately for students (−0.02 h, SD 2.35 h) and non-students (0.26 h, SD 1.34 h) mean differences were similar; however, limits of agreement were wide, although greater for students (−4.63 to 5.59 h) than non-students (−2.37 to 2.89 h) (plots not shown). T-tests showed that differences between the PAST-U sedentary time and PAL sedentary time
for the whole sample or for the students and non-students were non-significant. 4. Discussion The aim of this study was to modify the PAST questionnaire, recalling sedentary time on the previous day, to suit a university population and validate it against a device-based measure. The findings suggest that the PAST-U has acceptable levels of validity when compared to sedentary time from the activPAL, in terms of correlation (ICC = 0.64) and agreement (mean difference: 5 min) at the group level; however, estimates were poor at an individual level, as suggested by wide limits of agreement. This questionnaire may, therefore, be most appropriate for use in large-scale studies rather than for studies requiring estimates of an individual’s sedentary time. The strength of the correlation found between the PAST-U and activPAL derived sedentary time was similar to that found in the original validity study of the PAST questionnaire (r = 0.57).13 In comparison with another study which used an activPAL as criterion measure, the correlation for the PAST-U was higher than for the International Physical Activity Questionnaire (IPAQ) on weekdays (r = 0.41) and higher than the Domain Specific Questionnaire for weekdays (r = 0.30).21 Although there is no recommended standard for the acceptable strength of correlations between sedentary behaviour questionnaires and criterion measures, physical activity researchers have set a standard of r > 0.5 (against the accelerometer as criterion) for questionnaires measuring moderate and vigorous physical activity.22 Our findings show correlations above this level for the whole sample and for both the student and non-student subsamples. Very few other studies have used activPAL devices as the criterion measure for self-report measures of sedentary time. One other study that used this criterion measure examined the validity of an interviewer-administered, web-based time-use 24-h Physical Activity Recall. When sedentary time from the 24-h recall was validated against the activPAL, correlations were higher (men, = 0.81; women, = 0.81), than found for the university version of the PAST.23 In terms of agreement, the mean difference (men, MD = 0.72 h, women, MD = 0.75 h) was greater than for the PASTU, but the limits of agreement were smaller (95% LoA men = −2.61 to 4.05; women, −2.21 to 3.71).23 The stronger correlations and smaller limits of agreement for the 24-h recall suggest that it may provide better estimates of sitting time at the individual level than the PAST-U. However, the longer time spent completing such detailed recalls puts a higher burden on participants compared to questionnaires, which also provide information on the time spent in specific sedentary behaviours and can be completed in a shorter time (the original PAST questionnaire was completed in a median time of 7 min)13 . Given the lower participant burden and greater accuracy at the group level (evidenced by a small mean difference on the Bland Altman Plots) the PAST-U may be more suited to large scale surveillance studies. The PAST-U provides information on sedentary time in specific contexts for both students and staff. For students, the most
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model
ARTICLE IN PRESS
JSAMS-1144; No. of Pages 5
B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
4
Table 2 Sedentary time (h) reported on the PAST-U items.
Study Work Transport TV viewing Computer/electronic device use Reading Eating Socializing Others Composite sedentary time PAST, mean ± SD
Overall (n = 57)
Student (n = 37)
Non-student (n = 20)
2.50 (0, 5.50) 0 (0, 6.00) 0.67 (0.13, 1.08) 1.00 (0, 2.00) 1.00 (0.33, 1.75) 0 (0, 0) 1.00 (0.50, 1.50) 0.17 (0, 1.0) 0 (0, 0.50) 10.72 (2.42)
4.00 (2.00, 6.50) 0 (0, 0.29) 0.50 (0.04, 1.00) 1.00 (0, 2.00) 1.00 (0.50, 2.00) 0 (0, 0) 1.00 (0.58, 1.50) 0.33 (0, 1.00) 0 (0, 0.50) 10.74 (2.65)
0 (0, 0) 6.00 (5.13, 7.00) 0.88 (0.50, 1.23) 1.75 (0.06, 2.00) 0.50 (0, 1.00) 0 (0, 0.19) 0.75 (0.38, 1.00) 0.08 (0, 0.50) 0 (0, 0.69) 10.70 (1.98)
Data for PAST-U sedentary times are presented as median (interquartile range) except where indicated. Standard deviation (SD).
prevalent behaviour was studying, followed by using the computer and electronic devices. The finding that studying was the primary purpose of sedentary time in students is unsurprising and has been found previously in another study on sedentary behaviour patterns of university students.14 Long hours of sitting for work by non-student participants (median 5.53 h) was also expected and similar to findings for working participants in the previous study of the PAST13 and other studies of occupational sitting.24–26 Although absolute agreement between the measures was similar in the student and non-student population, correlation was better for the non-students suggesting this measure may be useful in working populations with high sedentary time. Participants in this study were sedentary for 66% of their waking hours which is a similar finding to other studies that have measured sedentary time using the activPAL in office workers (67% waking hours sedentary21 and 67% of working hours sedentary26 ) but more than the proportion of sedentary time found in the previous study on the PAST in breast cancer survivors (58%).13 The university population, therefore, appears to represent a population that experiences high sedentary time, similar to office workers. The use of the activPAL device as a criterion measure for the PAST-U questionnaire is a major strength of the study. The activPAL has shown high agreement with direct observation of sitting17–19 and is, therefore, a good criterion for evaluating correlation and agreement in self-report measures of sedentary time. The inclusion of multiple items assessing sedentary time in various contexts is also a strength. Composite measures of sitting time have been shown to provide better correlations with device-based measures
of sedentary time than single item sitting time questions.10,18,27 However, one of the issues with composite measures compared to single item sitting time questions is increased participant burden. A suggestion to further refine the PAST-U questionnaire is to remove or combine certain questions with very low level responses (e.g. reading), thus shortening the completion time. One limitation of this study is that the estimates of sedentary time spent in individual behaviours or domains may not perfectly reflect the proportion of time spent sedentary for each purpose, due to the possibility of multi-tasking and overlapping of time. For example, the component item ‘television viewing’ was asked earlier in the questionnaire, so the time reported could have included sedentary time from other components experienced simultaneously such as time spent eating, because participants were told not to report time for subsequent components if they had already reported it. This could affect the true value of the sedentary times for some components but does not affect estimates of total sedentary time. Another limitation is the single administration of the PAST-U. Behaviours and schedules vary daily and usual time spent sedentary may not be accurately assessed by a single pastday recall. Assessing multiple administrations of the PAST-U over several weekdays and weekend days is an area for improvement in future studies. The final limitation of this study is the sample. As expected in a University sample, the participants had all completed high school education. For broader use of the PAST-U questionnaire validity should also be assessed in other populations including those with lower levels of education. Participants were also possibly more
Fig. 1. Bland–Altman plot for overall PAST-U sedentary time and PAL sedentary time. The solid line represents the mean difference (MD) between the two measures and the dashed lines are the 95% LoA. Students are represented by solid grey diamonds (
), non-students are clear diamonds (
).
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
G Model JSAMS-1144; No. of Pages 5
ARTICLE IN PRESS B.K. Clark et al. / Journal of Science and Medicine in Sport xxx (2015) xxx–xxx
aware of their sitting time as they had been informed the study would measure sitting time. 5. Conclusion The PAST-U questionnaire shows acceptable validity for estimating sedentary time in a student and working population and can provide useful contextual information about sedentary time in different domains. This supports earlier research that past-day recall of sedentary time provides a viable alternative to commonly used recall of past week or usual behaviour. The PAST-U shows promise for use in population surveillance studies and to complement device-based measures of sedentary time. Practical implications • The PAST-U questionnaire shows validity properties that are similar to the original version. • The PAST-U shows acceptable validity for estimating sedentary time in student and working populations at a group level. • The accuracy of the PAST-U is not suitable for measurement of individual’s sedentary time. Acknowledgement The study was funded by a University of Queensland (UQ) Startup Grant (UQ no. 2012002122). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jsams.2015.02.001. References 1. Proper K, Singh A, van Mechelen W et al. Sedentary behaviors and health outcomes among adults: a systematic review of prospective studies. Am J Prev Med 2011; 40:174–182. 2. Williams DM, Raynor HA, Ciccolo JT. A review of TV viewing and its association with health outcomes in adults. Am J Lifestyle Med 2008; 2(3):250–259. 3. Chau JY, Grunseit AC, Chey T et al. Daily sitting time and all-cause mortality: a meta-analysis. PLoS ONE 2013; 8(11):e80000. 4. Dunstan DW, Barr ELM, Healy GN et al. Television viewing time and mortality: the Australian diabetes, obesity and lifestyle study (AusDiab). Circulation 2010; 121:384–391. 5. Pavey TG, Peeters GG, Brown WJ. Sitting-time and 9-year all-cause mortality in older women. Br J Sports Med 2015; 49:95–99.
5
6. Sedentary Behaviour Research Network. Standardized use of the terms “sedentary” and “sedentary, behaviours”. Appl Physiol Nutr Metab 2012; 37: 540–542. 7. Owen N, Bauman A, Brown W. Too much sitting: a novel and important predictor of chronic disease risk? Br J Sports Med 2009; 43(2):81–83. 8. Chau J, Merom D, Grunseit A et al. Temporal trends in non-occupational sedentary behaviours from Australian time use surveys 1992, 1997 and 2006. Int J Behav Nutr Phys Act 2012; 9(1):76. 9. Church T, Thomas D, Tudor-Locke C et al. Trends over 5 Decades in U.S. occupation-related physical activity and their associations with obesity. PLoS ONE 2011; 6:e19657. 10. Healy GN, Clark BK, Winkler AE et al. Measurement of adults’ sedentary time in population-based studies. Am J Prev Med 2011; 41(2):216–227. 11. Affuso O, Stevens J, Catellier D et al. Validity of self-reported leisure-time sedentary behavior in adolescents. J Negat Results Biomed 2011; 10(1):2. 12. Bassett Jr DR, Fitzhugh EC. Establishing validity and reliability of physical activity assessment instruments. 3, in Epidemiological methods in physical activity studies, Lee IM, editor, New York, NY, Oxford University Press, 2009. 13. Clark BK, Winkler E, Healy GN et al. Adults’ past-day recall of sedentary time: reliability, validity, and responsiveness. Med Sci Sports Exerc 2013; 45(6):1198–1207. 14. Rouse PC, Biddle SJH. An ecological momentary assessment of the physical activity and sedentary behaviour patterns of university students. Health Educ J 2010; 69(1):116–125. 15. Alkhajah TA, Reeves MM, Eakin EG et al. Sit-stand workstations: a pilot intervention to reduce office sitting time. Am J Prev Med 2012; 43(3): 298–303. 16. Vella-Zarb RA, Elgar FJ. The ‘freshman 5’: a meta-analysis of weight gain in the freshman year of college. J Am Coll Health 2009; 58(2):161–166. 17. Ryde GC, Gilson ND, Suppini A et al. Validation of a novel, objective measure of occupational sitting. J Occup Health 2012; 54(5):383–386. 18. Kozey-Keadle S, Libertine A, Lyden K et al. Validation of wearable monitors for assessing sedentary behavior. Med Sci Sports Exerc 2011; 43(8):1561–1567. 19. Grant P, Ryan C, Tigbe W et al. The validation of a novel activity monitor in the measurement of posture and motion during everyday activities. Br J Sports Med 2006; 40(12):992–997. 20. Bland JM, Altman GA. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986:307–311. 21. Kozey-Keadle S, Libertine A, Staudenmayer J et al. The feasibility of reducing and measuring sedentary time among overweight, non-exercising office workers. J Obes 2012:1–10. 22. van Poppel MN, Chinapaw MJ, Mokkink LB et al. Physical activity questionnaires for adults: a systematic review of measurement properties. Sports Med 2010; 40(7):565–600. 23. Matthews CE, Keadle SK, Sampson J et al. Validation of a previous-day recall measure of active and sedentary behaviors. Med Sci Sports Exerc 2013; 45(8):1629–1638. 24. Healy GN, Eakin EG, Lamontagne AD et al. Reducing sitting time in office workers: short-term efficacy of a multicomponent intervention. Prev Med 2013; 57(1):43–48. 25. Thorp AA, Healy GN, Winkler E et al. Prolonged sedentary time and physical activity in workplace and non-work contexts: a cross-sectional study of office, customer service and call centre employees. Int J Behav Nutr Phys Act 2012; 9:128. 26. Ryde GC, Brown HE, Peeters GM et al. Desk-based occupational sitting patterns: weight-related health outcomes. Am J Prev Med 2013; 45(4):448–452. 27. Clemes SA, David BM, Zhao Y et al. Validity of two self-report measures of sitting time. J Phys Act Health 2012; 9:533–539.
Please cite this article in press as: Clark BK, et al. Past-day recall of sedentary time: Validity of a self-reported measure of sedentary time in a university population. J Sci Med Sport (2015), http://dx.doi.org/10.1016/j.jsams.2015.02.001
