Measurement of physical activity has become a very important issue in public health and clinical medicine, especially because of the association between physical activity levels and health. More specifically, physical activity has been shown to be inversely associated with a variety of chronic diseases in adults, such as coronary heart disease (3,4,18,24), hypertension (5,25), non-insulin-dependent diabetes mellitus (3,9-11,16,20), colon cancer (27), and osteoporosis (14,17). Measurement of physical activity can be essential to evaluate health benefits and risk factors (29). The most commonly used method of estimating the prevalence of physical activity in population studies is the self-report by questionnaire (6).
At its inception in 1984, the WHO-MONICA project focused on three risk factors: smoking, high blood pressure, and serum total cholesterol (MONICA = Monitoring trends and determinants of cardiovascular diseases). The World Health Organization (WHO) MONICA project is a 10-yr study that monitors deaths resulting from coronary heart disease, acute myocardial infarction, coronary care, and risk factors in men and women aged 35 to 64 yr in defined communities (28). However, as the study progressed, new scientific evidence linking physical activity to coronary heart disease (CHD) accumulated. Growing recognition of sedentary living as a risk factor for CHD prompted the MONICA project's principal investigators to authorize an optional study of physical activity in 1987. Several participating centers had independently developed and used their own physical activity questionnaires during the initial cross-sectional survey of risk factors. However, a standardized questionnaire and methodology appropriate for all centers interested in participating was required. Thus, the MONICA Optional study of Physical Activity (or MOSPA) was started.
The MOSPA-questionnaire was developed by the World Health Organization with two concerns in mind. First, it was developed for use by countries at varying stages of economic development. Questions encompassed not only leisure time physical activity but also occupational activity, transportation, and housework. It was expected that the contribution of each of these areas to overall physical activity might vary considerably among different countries and cultures. In addition, the questions were developed in an attempt to inquire about the physical activity pattern of the individual rather than to try to estimate activity based on occupational title or group membership. This was done because the activity required by members of the same occupation might vary considerably, depending upon the culture of the specific country. Second, it was realized that the questionnaire would probably be both self and interviewer administered.
Because it has been shown that self-reports can be inherently biased (15), it is important to assess a questionnaire's validity and reliability for every population in which it will be used (23). Therefore, the purpose of the present study was to evaluate the validity and reliability of the MOSPA questionnaire in a Flemish population.
Methods and subjects. The study design consisted of a mailed questionnaire, self-administered at home, followed some months later by a clinic visit during which a repeat questionnaire (22) was self administered, and physiological testing of indirect validation measures was performed. Questionnaire packages were mailed to 385 physical education or physical therapy alumni from the Vrije Universiteit Brussel, Belgium. Each package contained a medical history questionnaire and a MOSPA questionnaire. Of the total, 285 subjects returned a completed questionnaire and were subsequently invited for participation in the study. Of the 285 alumni, 167 entered the study while the remaining 118 subjects dropped out, mostly for practical reasons of time constraint. Subjects who reported changing job titles were excluded from the reliability study. The study was approved by the ethical committee of the Vrije Universiteit Brussel for studies involving human subjects. A written informed consent was obtained from all subjects.
Questionnaire. The physical activity questionnaire tested in this study was the Dutch language version of the MOSPA questionnaire (22) as used by the Belgian MONICA center in Ghent. This questionnaire can be self or interview administered and measures the average weekly time and energy expenditure spent in (a) one's occupation (WORK, KWORK); (b) transport to and from work, school, and shopping (TRANS, KTRANS); (c) household chores (HOUS, KHOUS); and (d) leisure time activity (LTA, KLTA). Subjects were instructed to make clear distinctions between each area to avoid possible overlap. For leisure time activities, the intensities were derived from the Compendium of Physical Activities (1). The average daily total energy expenditure (TEE, kcal·d1) and the average weekly energy expenditure for each area (KWORK, KTRANS, KHOUS, and KLTA, kcal·wk−1) were computed according to the scoring methodology provided by the WHO.
Physiological and biometrical measurements. Each subject performed a graded exercise test (GXT) and a biometrical examination including underwater weighing according to the classic method described by Goldman (8). V˙O2peak was measured by direct oxygen measurement using an open-circuit spirometry (Oxycon, Mannheim, Germany) during the GXT performed on a cycle-ergometer (Jaeger ER800, Wuerzburg, Germany). Basal metabolic rate (BMR) was calculated from the re-evaluated Harris-Benedict equation (26).
Statistical methods. The distribution of the physical activity data was checked for normality by means of the Kolmogorov Smirnov Goodness of Fit test. Since the physical activity measures were not normally distributed, non-parametric statistics were used in the examination of the data. An unpaired t-test was used to test the differences between men and women. The reliability analysis was performed using the intra-class correlation coefficient, and differences between the two administrations were tested using a Wilcoxon matched pairs signed-ranks test. For validation purposes, Spearman rank order correlation coefficients were calculated. Data were analyzed using STATISTICA for Windows 5.0 statistical software.
Descriptive statistics. Mean levels of descriptive and validation measures are presented in Table 1. There was no significant difference in age between men and women. Men had a higher body mass index than women and obtained a higher peak oxygen uptake during the GXT. Women had a higher percentage of body fat than men.
Although men on average worked more hours per week than women (46 ± 1 vs 32 ± 2 h·wk−1, P < 0.001), men were not more physically active during occupational time than women as can be observed from the mean weekly energy expenditure during physical active work-related activities (KWORK) (3322 ± 279 vs 2515 ± 354 kcal·wk−1; NS). Time and energy expenditure in transportational activities (TRANS and KTRANS) did not differ between men and women. Women reported to spend more time and energy in household chores than men (HOUS and KHOUS). Although there was no difference in time spent at leisure time physical activities (LTA, 205 ± 14 for men vs 225 ± 36 for women min·wk−1), men were characterized as more intensively physically active during leisure time than women, as can be observed from their energy expenditure in leisure time activities (KLTA) (1733 ± 138 vs 1215 ± 227 kcal·wk−1; P < 0.05)
Reliability. Table 2 presents the intra-class correlation coefficients (ICC) for the short- and long-term reproducibility of the MOSPA questionnaire and its component parts. ICC scores for the MOSPA did not differ between men and women (data not shown). The mean time interval between two administrations of the questionnaire was 146 ± 11 d. Short-term reliability (38 ± 3 d) ranged from r = 0.57 to 0.91 whereas long-term reliability ranged from r = 0.45 to 0.85 for repeated administrations of the questionnaire (240 ± 12 d). Generally, reliability at short-term administration (less than 3 months) was higher than reliability at long-term administration (greater than 3 months). Reliability was lower for household chores and transport-related activities than for occupational- and leisure time-related activities. Figure 1 shows test-retest values ± SEM for both administrations. KWORK and KLTA differed between the two administrations (P < 0.05), whereas KTRANS and KHOUS were not significantly different. The mean daily total energy expenditure (TEE) that is derived from KWORK, KTRANS, KHOUS, KLTA, and BMR did not differ between the two administrations.
Validity. Table 3 provides an overview of the correlation between MOSPA scores and validation criteria. Moderate to high correlations were found between cardiorespiratory fitness (V˙O2peak), lean body mass (LBM), body fatness, and MOSPA (KLTA and TEE). However, low correlations were obtained between household- and transport-related activities (HOUS, TRANS) and validation criteria. The occupational scores (WORK, KWORK) correlated moderately, but significantly, with cardiorespiratory fitness, LBM, and body fatness.
This study reports on the validity and reliability of the MOSPA-questionnaire in a Flemish population.
A moderate to high reliability was found for occupational and leisure time physical activities, and a low to moderate reliability for household- and transport-related activities. In the validation study, the highest correlations were found between the leisure time activities, and the cardiorespiratory (V˙O2peak) and biometrical measurements. The correlations found between MOSPA scores and validation criteria are in the same range as those found by others using similar criteria (2,7,23,29). In a simultaneous evaluation of 10 commonly used physical activity questionnaires, Jacobs et al. (12) found sex-adjusted correlation coefficients of 0.39-0.45 between the Minnesota Leisure Time Physical Activity questionnaire (MLTPA-Q) and cardiorespiratory fitness variables. In our study the MOSPA KLTA-score correlated 0.44 with the V˙O2peak measurement.
When the reliability of a questionnaire is examined, doubt can arise as to whether the consistency of the questionnaire or the consistency of the subjects' physical activity habits is being examined (19). In this study no evidence was found to suggest that subjects altered their physical activity levels during the study period.
Reliability was low for moderate and low intensive physical activities (e.g. HOUS) which corroborates with other studies (12). The latter can be explained by the fact that this type of activity is not stable over time and, consequently, is not easy to recall by comparison with intensive (e.g., LTA) or structured, frequently repeated activities (e.g., WORK). Although the TEE for both administrations of the questionnaire was the same, KWORK and KLTA differed significantly. This could be attributed to the fact that high intensive physical activity is often overestimated in self-reports (15,21). The KLTA for the first administration was significantly higher than for the second administration. The fact that in our study population KLTA included mostly high intensive physical activities such as running, soccer, and aerobics might explain the observed difference between the two administrations. Because of the background of the population studied (physical education and physical therapy), it is assumed that the subjects are aware of the importance of the effects and benefits of moderate physical activity levels. We presume that this attitude contributes to the bias of underestimating nonintensive physical activity and overestimating high intensive physical activity. Finally, the second administration of the questionnaire was performed in a laboratory setting, and this may have been the reason that the questions were more thoroughly answered. The subjects were also conscious of the fact that after completing the questionnaire they would be tested for physical fitness status. Knowledge of their physical fitness condition might have influenced them to respond more objectively to the questions regarding their physical activities during leisure time. Work-related activities, mostly low intensive physical activities, were less frequently reported at the first administration compared with those reported at the second administration. Since these activities were generally of low intensity, the observed differences could be explained again by the aforementioned reasons.
Validation is much more difficult to establish since there is no consensus on how a questionnaire, covering such a long time period, should be validated. In most studies concurrent validity has been measured by administering questionnaires with a time-recall at different intervals throughout the study period in such a way that these recalls when summarized, validate the questionnaire which covers the whole study period (5,7,12,13,29). Another method that has been proposed to validate questionnaires is to obtain diaries from the studied population (21). In our study the aforementioned methods were not used because of practical limitations. Instead, validation was accomplished by comparing the estimates of physical activity (derived from the questionnaire) to objectively obtained physiological and biometrical measurements.
Although the association between the objective measures and reported physical activity was in the expected direction, many of the correlates between the questionnaire and fitness measures were weak and not significant. Measures of fitness or lean body mass may not be the most appropriate validation criteria for lower-intensity activities such as occupational activity or housework. However, as long as we do not have a gold standard for validating low intensive physical activity measurements, the validation of the MOSPA questionnaire is subject to limitations imposed by the criterion measures of physical activity.
In conclusion, the questionnaire examined in this study was found to be acceptably valid and reliable. The MOSPA-questionnaire can be easily administered to large numbers of individuals with minimal cost and burden to the subjects. It is comprehensive in that both leisure time and occupational activity can be assessed. This is especially meaningful when evaluating physical activity in countries where funding for scientific research is often limited. Understanding validity, reliability, and limitations of such questionnaires that are designed to assess physical activity in various cultures is important. This information may help to understand physical activity differences in various cultures and allow researchers to compare activity levels in these various cultures to illuminate various disease patterns that could be attributed to differences in activity levels. Further research is warranted to clarify the validity of the MOSPA-questionnaire in other populations and cultures of the MONICA project.
1. Ainsworth, B. E., W. L. Haskell, A. S. Leon, et al. Compendium of Physical Activities: classification of energy costs of human physical activities. Med. Sci. Sports Exerc.
2. Baranowski, T. Validity and reliability of self report measures of physical activity: an information-processing perspective. RQES
3. Barnard, R. J. and S. J. Wen. Exercise and diet in the prevention and control of the metabolic syndrome. Sports Med.
4. Berlin, J. A. and G. A. Colditz. A meta-analysis of physical activity in the prevention of coronary heart disease. Am. J. Epidemiol.
5. Bijnen, F. C., C. J. Caspersen, and W. L. Mosterd. Physical inactivity as a risk factor for coronary heart disease: a WHO and International Society and Federation of Cardiology position statement. Bull. World Health Organ.
6. Booth, M. L., N. Owen, A. E. Bauman, and C. J. Gore. Retest reliability of recall measures of leisure-time physical activity in Australian adults. Int. J. Epidemiol.
7. Chasan-Taber, S., E. B. Rimm, M. J. Stampfer, et al. Reproducibility and validity of a self-administered physical activity questionnaire for male health professionals. Epidemiology
8. Goldman, R. F. and E. R. Buskirk. Body volume measurement by under-water weighing: description of a method. In: Techniques for Measuring Body Composition,
J. Brozek and A. Henschel (Eds.). Washington, D.C.: National Academy of Sciences, National Research Council, 1961, pp. 78-89.
9. Helmrich, S. P., D. R. Ragland, R. W. Leung, and R. S. Paffenbarger, Jr. Physical activity and reduced occurrence of non-insulin-dependent diabetes mellitus. N. Engl. J. Med.
10. Helmrich, S. P., D. R. Ragland, and R. S. Paffenbarger, Jr. Prevention of non-insulin-dependent diabetes mellitus with physical activity. Med. Sci. Sports Exerc.
11. Horton, E. S. Exercise and decreased risk of NIDDM. N. Engl. J. Med.
12. Jacobs, D. R., B. E. Ainsworth, T. J. Hartman, and A. S. Leon. A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med. Sci. Sports Exerc.
13. Kelder, S. H., C. L. Perry, K. I. Klepp, and L. L. Lytle. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am. J. Public Health
14. Kirchner, E. M., R. D. Lewis, and P. J. O'Connor. Effect of past gymnastics participation on adult bone mass. J. Appl. Physiol.
15. Klesges, R. C., L. H. Eck, M. W. Mellon, W. Fulliton, G. W. Somes, and C. L. Hanson. The accuracy of self-reports of physical activity. Med. Sci. Sports Exerc.
16. Kriska, A. M., W. C. Knowler, R. E. Laporte, et al. Development of a questionnaire to examine relationship of physical activity and diabetes in Pima Indians. Diabetes Care
17. Kriska, A. M., R. B. Sandler, J. A. Cauley, R. E. Laporte, D. L. Hom, and G. Pambianco. The assessment of historical physical activity and its relation to adult bone parameters. Am. J. Epidemiol.
18. Lakka, T. A., J. M. Venalainen, R. Rauramaa, R. Salonen, J. Tuomilehto, and J. T. Salonen. Relation of leisure-time physical activity and cardiorespiratory fitness to the risk of acute myocardial infarction in men. N. Engl. J. Med.
19. Lamb, K. L. and D. A. Brodie. The assessment of physical activity by leisure-time physical activity questionnaires. Sports Med.
20. Mayer, E. J., B. W. Alderman, J. G. Regensteiner, et al. Physical activity assessment measures compared in a biethnic rural population: the San Luis Valley diabetes study. Am. J. Clin. Nutr.
21. Montoye, H. J., H. C. G. Kemper, W. H. M. Saris, and R. A. Washburn. Questionnaires and interviews. In: Measuring Physical Activity and Energy Expenditure.
H. J. Montoye, H. C. G. Kemper, W. H. M. Saris, and R. A. Washburn (Eds.), Champaign, IL: Human Kinetics, 1996, pp. 42-71.
22. Pereira, M. A., S. J. Fitzgerald, E. W. Gregg, et al. A collection of physical activity questionnaires for health-related research: The Monica Optional Study of Physical Activity (MOSPA). Med. Sci. Sports Exerc.
23. Pols, M. A., P. H. M. Peeters, H. B. Bueno-De-Mesquita, et al. Validity and repeatability of a modified Baecke questionnaire on physical activity. Int. J. Epidemiol.
24. Powell, K. E., P. D. Thompson, C. J. Caspersen, and J. S. Kendrick. Physical activity and the incidence of coronary heart disease. Ann. Rev. Public Health
25. Rauramaa, R., S. B. Väisänen, T. Rankinen, et al. Inverse relation of physical activity and apolipoprotein A1 to blood pressure in elderly women. Med. Sci. Sports Exerc.
26. Roza, A. M. and H. M. Shizgal. The Harris-Benedict equation reevaluated: resting energy requirements and the body cell mass. Am. J. Clin. Nutr.
27. Slattery, M. L., M. C. Schumacher, K. R. Smith, D. W. West, and N. Abd-Elghany. Physical activity, diet, and risk of colon cancer in Utah. Am. J. Epidemiol.
28. Tunstall-Pedoe, H., K. Kuulasmaa, P. Amouyel, D. Arveiler, A.-M. Rajakangas, and A. Pajak. Myocardial infarction and coronary deaths in the World Health Organization MONICA project. Circulation
29. Wolf, A. M., D. J. Hunter, G. A. Colditz, et al. Reproducibility and validity of a self-administered physical activity questionnaire. Int. J. Epidemiol.
Keywords:© Williams & Wilkins 1998. All Rights Reserved.
ENERGY EXPENDITURE; EXERCISE; REPEATABILITY