ORIGINAL PAPER
Measurement properties of the CHAMPS physical activity questionnaire in a sample of older Australians

https://doi.org/10.1016/j.jsams.2006.03.001Get rights and content

Summary

Background

The effective evaluation of physical activity interventions for older adults requires measurement instruments with acceptable psychometric properties that are sufficiently sensitive to detect changes in this population.

Aim

To assess the measurement properties (reliability and validity) of the Community Healthy Activities Model Program for Seniors (CHAMPS) questionnaire in a sample of older Australians.

Methods

CHAMPS data were collected from 167 older adults (mean age 79.1 S.D. 6.3 years) and validated with tests of physical ability and the SF-12 measures of physical and mental health. Responses from a sub-sample of 43 older adults were used to assess 1-week test–retest reliability.

Results

Approximately 25% of participants needed assistance to complete the CHAMPS questionnaire. There were low but significant correlations between the CHAMPS scores and the physical performance measures (rho = 0.14–0.32) and the physical health scale of the SF-12 (rho = 0.12–0.24). Reliability coefficients were highest for moderate-intensity (ICC = 0.81–0.88) and lowest for vigorous-intensity physical activity (ICC = 0.34–0.45). Agreement between test-retest estimates of sufficient physical activity for health benefits (≥150 min and ≥5 sessions per week) was high (percent agreement = 88% and Cohen's kappa = 0.68).

Conclusion

These findings suggest that the CHAMPS questionnaire has acceptable measurement properties, and is therefore suitable for use among older Australian adults, as long as adequate assistance is provided during administration.

Introduction

The number of Australians aged over 65 years and, consequently, the number of people with chronic disease and/or age-related disorders is set to rapidly increase over the next 20 years.1 Regular physical activity (PA) can prevent or postpone the development of many chronic health conditions, and in older adults it can ameliorate the declines in physical function commonly associated with ageing.2, 3, 4 To prevent the course of unhealthy ageing, research attention is beginning to focus on the development and implementation of strategies for increasing PA among older adults. To enhance the evidence base, programs must be rigorously evaluated. Measurement tools that provide valid and reliable estimates of PA in this population need to be identified.

In applied research settings, PA is usually assessed using self-report methods (e.g. survey and diaries). Population-based data collected using the Active Australia (AA) questionnaire suggest that only 44% of Australian adults aged 60–75 years are sufficiently active for health benefit and that almost one in five older Australians are completely sedentary.5 The reliability and validity of the AA questionnaire have been established6, 7 but were found to be lowest among those aged over 60 years.6 Whilst the AA questionnaire is the preferred instrument for monitoring national population levels of PA over time, it may not be sensitive enough to detect small but significant changes in PA following planned interventions in smaller samples of older Australians.

Self-report data are limited by a number of factors. Among older adults in particular, self-report data may be plagued by recall bias8, 9 and by the ‘floor’ effects resulting from low levels of PA in this population.8, 10 To address these issues, several PA questionnaires have been developed overseas to assist older adults recall and report PA.11, 12, 13, 14, 15 Typically, these surveys emphasise leisure and household activities, although the Physical Activity Scale for the Elderly (PASE)14 and the Community Healthy Activities Model Program for Seniors (CHAMPS) questionnaire15 also measure time spent in paid or volunteer work. Within leisure and household domains, the PASE, the Yale Physical Activity Survey (YPAS),13 and the CHAMPS questionnaire15 cover the widest spectrum of activities. In terms of capturing small changes in PA over time, the CHAMPS questionnaire has been found to be more sensitive to change than the YPAS or the PASE.8

Stewart and colleagues15 developed the CHAMPS questionnaire as an outcome measure for a PA intervention study, because, at that time there were no accurate self-report instruments for assessing PA in under-active older adults. The questionnaire was specifically designed to minimise social desirability (by including non-physical activities such as hobbies) and recall bias (use of recognition memory). The CHAMPS questionnaire assesses the weekly frequency and duration of activities typically undertaken by older adults at light (e.g. leisurely walking, light gardening), moderate (e.g. cycling, heavy housework) and vigorous (e.g. singles tennis, jogging) intensities. A MET value is assigned to each activity, using the compendium,16 but with adjustment for the likely intensity of each activity among older adults.15

To date, two studies have evaluated the psychometric properties of the CHAMPS questionnaire among community-dwelling older adults in the United States. They reported Intraclass Correlation Coefficients (ICC) with moderate 6-month stability (ICC = 0.58–0.67)15 and good test-retest reliability over a 2-week period (ICC = 0.62–0.76).8 In these studies, the CHAMPS data demonstrated modest but acceptable predictive validity, including associations between CHAMPS scores and both physical functioning and health-related quality of life measures.8 CHAMPS data also showed moderate correlations with data from the PASE and the YPAS,8 and were able to discriminate between less active and more active older adults.8, 15

To date, no studies have explored the measurement properties of the CHAMPS survey in older Australian adults. The aim of the present study was therefore to examine the predictive validity and test-retest reliability of the CHAMPS questionnaire in a sample of older Australians.

Section snippets

Participants

A sample of 167 older adults (132 women and 35 men) aged over 60 years was recruited from nine independent-living retirement villages located in the Brisbane metropolitan area. These residents had volunteered to participate in an exercise program, due to commence at the completion of this study. Participants were provided with verbal and written information about the project, and informed consent forms were signed prior to participation. Approval for this study was obtained from the Medical

Results

Complete predictive validity and test-retest reliability data were received from 167 and 43 participants, respectively. Participants were aged between 65 and 96 years. The majority of participants were retired women, and 72% reported three or more chronic health problems. The reliability sample was significantly younger, included fewer participants with osteoporosis, and included more participants classified as sufficiently active than the validity sample (see Table 1). Thirteen of those who

Discussion

This study examined the measurement properties (predictive validity and test-retest reliability) of the CHAMPS questionnaire, to determine its suitability for use with older Australians. To date, the psychometric properties of this questionnaire have only been evaluated with samples of older Americans.8, 15

In this study, the length of time required to complete the CHAMPS questionnaire was comparable to previous reports involving samples of community-dwelling older adults.8, 15 However,

Acknowledgements

This project was funded by the Australian Government (Department of Health and Ageing—Office for an Ageing Australia) and Blue Care (Uniting Care Queensland). This research was carried out whilst Ms Cyarto was holding an International Postgraduate Research Scholarship at The University of Queensland. The authors wish to acknowledge the individuals who took part in this study.

References (26)

  • C. Tudor-Locke et al.

    Challenges and opportunities in measuring physical activity in sedentary adults

    Sports Med

    (2001)
  • C. Caspersen et al.

    The prevalence of selected physical activities and their relation with coronary heart disease risk factors in elderly men: the Zutphen study, 1985

    Am J Epidemiol

    (1991)
  • L. Voorrips et al.

    A physical activity questionnaire for the elderly

    Med Sci Sports Exerc

    (1991)
  • Cited by (46)

    View all citing articles on Scopus
    View full text