Measurement of Adults' Sedentary Behaviour by Questionnaire

Bronwyn Clark (2011). Measurement of Adults' Sedentary Behaviour by Questionnaire PhD Thesis, School of Population Health, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s3141889_phd_finalthesis.pdf Final Thesis application/pdf 1.88MB 0
Author Bronwyn Clark
Thesis Title Measurement of Adults' Sedentary Behaviour by Questionnaire
School, Centre or Institute School of Population Health
Institution The University of Queensland
Publication date 2011-11-01
Thesis type PhD Thesis
Supervisor Neville Owen
David Dunstan
Takemi Sugiyama
Genevieve Healy
Total pages 213
Total colour pages 6
Total black and white pages 207
Language eng
Subjects 111706 Epidemiology
111799 Public Health and Health Services not elsewhere classified
110699 Human Movement and Sports Science not elsewhere classified
Abstract/Summary Background Sedentary behaviour, defined by both posture (sitting or lying) and low energy expenditure (≤1.5MET), is a highly prevalent behaviour in adults. There is evidence that time spent in sedentary behaviour is detrimentally associated with several health outcomes, even after adjusting for leisure time physical activity. Much of this evidence-base has been generated from questionnaire measures. Existing questionnaires assess sedentary behaviours in different domains, utilising different recall periods and administration methods, and have been compared to various criterion measures. However, research has not provided enough information about how best to measure sedentary behaviour with a questionnaire and whether improvements in the accuracy of these are possible is largely unknown. Aims This Thesis aims to examine the reliability and validity of existing questionnaire measures of adults’ sedentary behaviour; and, to inform the development of potential new questionnaire measures for future epidemiological studies and population surveillance. Methods and Results A comprehensive literature review examined the reported reliability and validity of existing questionnaire measures of sedentary behaviour in adults. Sedentary behaviour questionnaires predominantly have used single-item questions of overall sitting or TV viewing time with past week or typical day recall as the reference period. The review found that for most existing questionnaires the test-retest reliability was good. However, validity studies showed low to moderate correlations with device-based criterion measures (ρ=-0.01 to 0.61). There was limited information on other leisure-time sedentary behaviours, and the measurement of sedentary behaviour within the domains of work and transport, particularly in terms of validity. Based on the results of the literature review, three empirical studies were undertaken in this Thesis. The literature review showed that TV viewing time was the most commonly assessed sedentary behaviour and was often the only measure of sedentary behaviour. Therefore, the first empirical study in the Thesis examined the relationship between self-reported TV viewing time and overall device-measured sedentary time, to determine if TV time could be used as a surrogate measure of sedentary time. Data were derived from the National Health and Nutrition Examination Survey (n=5738 adults in the USA), which included a TV viewing time question and an accelerometer measure (used to determine sedentary time, <100 counts per minute). Findings showed that TV viewing time, despite being positively associated with overall sedentary time, was not a strong marker of sedentary time (ρ=0.22). This study indicates that questionnaires should assess sitting undertaken within other behaviours both within the leisure domain and in other domains, as well as TV viewing time, to better capture adults’ overall sedentary time. Importantly, there was no relationship between TV viewing time and overall sedentary time for full-time workers. Workplace sitting may be particularly important for this group. Although workplace has been identified as a key setting for health behaviour interventions, considerable gaps in research on the validity of workplace sitting questionnaires remain. The second empirical study examined the validity of a question on workplace sitting (total time in the past week, data from Stand Up Australia, n=121 Australian adults) using an accelerometer measure obtained during work hours as the criterion. The validity correlation was r=0.39, which was similar to findings for measures of overall sitting when compared to device-based criterion measures. In order to assess workplace sitting more accurately with a questionnaire, changes in the recall period and the use of a more appropriate criterion measure were recommended. These improvements were addressed in the final study of this Thesis. The final study examined the reliability and validity of a questionnaire on sitting during the past day (seven items, time summed for overall sitting) using the activPAL (inclinometer measure of sit/lie time) device as the validity criterion. Data from the Living Well after Breast Cancer study (n=85) were used for the study. Past-day recall of overall sitting time showed fair to good reliability (ICC=0.54). The validity correlation was ρ=0.68, which was higher than previous findings for the validity of overall sedentary time questionnaires. The approach of using past-day recall of overall sitting shows promise for future epidemiological and surveillance studies. Conclusions This Thesis provides new insights into the measurement of sedentary behaviour by examining the reliability and validity of existing questionnaires and identifying new measures for use in epidemiological and surveillance studies. The literature review showed that existing questionnaires may not provide an adequate measure of sedentary behaviour. Findings from the empirical studies in the Thesis indicate that future questionnaires should ask about multiple sedentary behaviours in a range of domains, not solely leisure. Additionally, while the reliability of currently-used questionnaires is good, validity could be improved by employing a past-day recall method. Future research should extend this work by examining how new and existing questionnaires perform in diverse populations, using the best available criterion measures of sedentary time, including their ability to detect responsiveness to change in sitting.
Keyword Measurement
Additional Notes Colour printing p150, 206, 208, 209, 211, 212 Landscape p113-114, 155-185

Citation counts: Google Scholar Search Google Scholar
Created: Tue, 01 May 2012, 22:15:50 EST by Bronwyn Clark on behalf of Library - Information Access Service