Share this article on:

00005768-199802000-0001500005768_1998_30_266_friedenreich_questionnaire_2miscellaneous-article< 128_0_14_14 >Medicine & Science in Sports & Exercise©1998The American College of Sports MedicineVolume 30(2)February 1998pp 266-274The Lifetime Total Physical Activity Questionnaire: development and reliability[Basic Sciences: Epidemiology]FRIEDENREICH, CHRISTINE M.; COURNEYA, KERRY S.; BRYANT, HEATHER E.Division of Epidemiology, Prevention and Screening, Alberta Cancer Board, 1331-29 St. N.W. Calgary, Alberta, T2N 4N2 CANADA; and Faculty of Physical Education and Recreation, University of Alberta, Alberta, CANADASubmitted for publication July 1996.Accepted for publication July 1997.This work was funded by a research grant from the Alberta Cancer Board's Research Initiative Program.C. M. Friedenreich is supported by a National Health Research Scholar Award from the National Health Research and Development Program of Health Canada.Address for correspondence: Division of Epidemiology, Prevention and Screening, Alberta Cancer Board, 1331-29 St. N.W. Calgary, Alberta, T2N 4N2. E-mail: To develop and test the intra-rater reliability of an interview-administered questionnaire that assesses lifetime patterns of total physical activity including occupational, household, and exercise/sports activities.Methods: The questionnaire was developed and pretested using cognitive interviewing techniques on a sample of women with and without previous breast cancer diagnoses. A pilot study was conducted with 115 women who were interviewed twice, 6 to 8 wk apart by interviewers trained in cognitive interviewing methods. Respondents used recall calendars to record their education, occupations, life events, and physical activity patterns before the interviews. Interviewers helped respondents recall their lifetime exposures, including their occupational, household, and exercise/sports activities, using these calendars and memory-probing strategies. Activity levels were estimated as the average number of hours of activity per week over different time periods. Means and correlation coefficients were estimated and compared for the two time periods.Results: The questionnaire was found to be highly reliable. The test-retest correlations for hours per week spent in total lifetime physical activity was 0.74, for lifetime occupational activity was 0.87, for household activity was 0.77, and for exercise/sports activities was 0.72.Conclusions: This is the first questionnaire to measure lifetime physical activity by collecting data on each type of physical activity separately over lifetime and by measuring frequency, intensity, and duration of each activity. It is also the first physical activity questionnaire to be developed, refined, and administered using cognitive-based methods employed in survey research. Respondents were able to reliably recall their lifetime physical activity patterns. This instrument can be used for any disease outcome for which physical activity may be a risk factor.Retrospective epidemiologic studies rely on the recall ability of respondents for estimates of past exposure. Despite the widespread use of retrospective studies, relatively few epidemiologic investigations have attempted to improve respondents' ability to recall(14). Improvements are possible by developing measurement instruments using cognitive techniques and by including memory probes and recall strategies(7,8,22,23). Recommendations have recently been made to use the cognitive model to develop, refine, and administer physical activity instruments (12). Since lifetime assessments of exposure are needed to evaluate fully the risk associated with these exposures, measurement instruments are needed that can provide such lifetime coverage. Although numerous physical activity measurement instruments have been designed (3), none of these instruments measures the frequency, duration, and intensity of all types of physical activity over a lifetime.A questionnaire designed by Kriska et al. (20), for use on a North American Indian population, is the only previously developed questionnaire that assesses physical activity for a time period longer than 1 yr. The Kriska questionnaire measures occupational and leisure activity historically (i.e., lifetime) for the past year and the past week, but it does not measure any household activity. This questionnaire was similar to one originally developed, by the same investigator, for a white postmenopausal female population that assessed only historical leisure physical activity(21). The main limitation of the Kriska questionnaire(20), for use in this breast cancer study, was the omission of household activities. Previous research has clearly shown that a major component of female activity, which accounts for most of the weekly energy expenditure, is household activity (31). Recommendations have been made to include household activites as a dimension separate from occupational and leisure activities in physical activity questionnaires (2,31). Given this limitation and because it was developed for a very specific population without the use of cognitive research methods known to improve accuracy of long-term recall(18,19,22,33), it was necessary to develop a questionnaire that met the criteria for our study. Thus, as part of an ongoing, population-based, case-control study of physical activity and breast cancer risk, a questionnaire was developed, refined, and tested to determine the intra-rater reliability of recall of lifetime occupational, household, and exercise/sports activity. This study received ethics approval, and all study participants signed an informed consent.METHODDevelopment of instrumentsDesign of questionnaire. A questionnaire was designed to assess, in separate sections, lifetime occupational, household, and exercise/sports activity (Appendix 1). For occupational and household activity, the duration (in years), and the frequency (in months/year, days/week, and time/day) were reported. For exercise/sports activity, respondents were asked to report the frequency of their activity either per day, week, month, or year and the time for each activity in hours or minutes per exercise/sports session.Respondents were also asked to rate the intensity of each activity as sedentary (used for occupational activities only), light, moderate, or heavy. These activities were defined as follows: 1) sedentary are those activities requiring only sitting with minimal walking; 2) light are activities that require minimal physical effort such as standing and slow walking with no increase in heart rate and no perspiration; 3)moderate are activities that are not exhausting, that increase the heart rate slightly and may cause some light perspiration, such as those that require carrying light loads (5-10 lb.) or that have continuous walking; and 4) heavy are activities that are vigorous, increase the heart rate substantially and cause heavy sweating such as those that involve lifting, carrying heavy loads (>10 lb.), brisk walking, or climbing.Minimum levels of activity and a combination of frequency and duration of activity were assigned for each type of activity, and activities that were deemed to be done too infrequently to make a significant contribution to a respondent's total physical activity expenditure were not recorded. In addition, activities must have been done at least 10 times in a lifetime to be recorded.Recall calendars. Two recall calendars, one focused only on educational and occupational activities and the second on major life events, were developed as memory aids for the respondents since these have been proposed as a means for improving recall (23,24). The occupation calendar asked respondents to list their education and the jobs that they had held throughout their lives. For each job held, they were also asked to note the level of physical activity that was required for that position. Likewise, in the life events calendar respondents could record, at the appropriate ages, the major events in their lives (e.g., births, deaths, marriages, moves), menstrual and reproductive events (age at menarche, menopause, pregnancies), any hormone use history, and all exercise/sports activities that they undertook.Cognitive interviewing methods. Four interviewers were hired and provided with an intensive 2-d training workshop on the theory and methods of cognitive interviewing (32). These methods were used to pretest the questionnaire and recall calendars. Four cognitive processes are assessed when cognitive interviewing methods are used in pretesting questionnaires (30,33). The first process isquestion comprehension (do respondents understand the questions and is any rewording necessary?). The second cognitive process isinformation retrieval (can respondents remember the information being sought and do the recall calendars assist their recall?). The third cognitive process is judgment/estimation (how do respondents estimate the amount of time spent in various physical activities and how do they judge whether these estimates are accurate?). The final aspect of cognition assessed is response formulation (do respondents provide unbiased responses and are they confident in the accuracy of their responses?).Several interviewing techniques (18) were used to assess the respondents' ability to understand the questions, retrieve the information asked, estimate the time, amount, and intensity of physical activity undertaken, and formulate a response. These methods included: 1)concurrent think aloud techniques (the respondents think aloud when answering the questions and their responses are probed extensively); 2)paraphrasing (the respondents repeat the question in their own words, thereby permitting the interviewers to assess whether they understand the question); 3) probes (the interviewers use follow-up questions to gain more information about the respondents' strategies used for answering questions); 4) memory cues (the interviewers use methods, such as the recall calendar, to aid the respondents in recalling past activities); 5)response latency (the length of time used between the questions and the respondents' answers is measured to assess the difficulty of the question); and 6) confidence ratings (the interviewers ask the respondents to relate the degree of confidence they have in the accuracy of their answers).ProtestThe questionnaire and recall calendars were pretested, using cognitive interviewing methods, on 14 women, ages 34-65, about half of whom had a previous breast cancer diagnosis. The women were either members of a local breast cancer survivors group (Reach to Recovery) or volunteers at the university hospital or cancer clinic. In the pretesting, it was found that the respondents were able to answer the questions asked and that the recall calendars were particularly useful and effective in assisting them to situate the period of recall with respect to their own life events. Recall calendars are known to improve recall since cues about where an event took place, who was involved, and what happened help situate a respondent and thereby ease the recall process for the respondent(30). Some respondents initially felt that they would not be able to remember their past physical activities but that the recall calendars and probing techniques enabled them to remember these activities quite easily. In addition, lists of recreational activities were used to help interviewers prompt respondents to remember all activities that they performed in the past.Pilot StudyFor the reliability study, four interviewers were each asked to interview 30 women twice over a 6- to 8-wk period. These women were all participants inScreen Test: Alberta Program for the Early Detection of Breast Cancer (the provincial breast screening program). Interviews began in June 1995, were completed by October 1995, and were conducted at a time and place convenient for the respondents. The average time between interviews was 7.4(SD = ±1.4) wk. A total of 203 women were sent letters of invitation and contacted by telephone to arrange an interview, and 119 women agreed for the first interview. Of these 119 women, 115 were interviewed a second time; 4 women were unavailable during the time period needed for the second interview because of holidays, they had moved, or they were too busy. Of the 115 women, 2 did not complete the entire physical activity questionnaire for both interviews; hence, the analysis is restricted to 113 respondents.Interviewers asked about past physical activity beginning with childhood and ending with the reference year activity (the year before the interview in this study). After completing each interview, they coded the questionnaires; the coding of the questionnaire was then verified by the research team, and corrections were made before all data were double entered. The physical activity questionnaire takes about 20-30 min to complete. Study participants did not find the questionnaire too long or too difficult to complete; indeed, some actually commented on how much they enjoyed remembering their lifetime activities. Respondent fatigue is minimized by administering this questionnaire before other questions on personal characteristics and health history that are more quickly and readily reported.Analytic methodsThe outcome variables were estimated as the number of hours spent in each type of activity for different time periods in a respondent's lifetime and at different intensity levels (Appendix 2). Sedentary activities were recorded for occupational activity since many jobs involve only sitting, and an intensity level had to be recorded for each job held. However, the hours spent in sedentary activity were not included in any estimates of total activity since these estimates are of activity as opposed to inactivity. Total physical activity was estimated as the sum of occupational, household, and exercise/sports activities in hours per week. It is also possible to convert these data into energy expended by multiplying the hours spent by the estimated metabolic cost of that activity. The resulting data would be denoted as MET-hours/week. A MET is defined as the ratio of the associated metabolic rate for a specific activity divided by the resting metabolic rate (5). For this pilot study, three preassigned intensity MET values were used for light, moderate, and heavy intensity activity (15). However, a more appropriate use for this questionnaire will be to apply MET levels from the Compendium of Physical Activities for each reported activity (4). Future research using this questionnaire is planned that will apply the compendium to these data.For each type of activity, estimates of activity were made for different time periods including total lifetime, the year before the interview(reference year), childhood and adolescence (≤17 yr), early adulthood(18-34 yr), middle adulthood (35-49 yr), and older adulthood (≥50 yr). To examine the test-retest reliability of the measurements of physical activity in the pilot study, Pearson correlation coefficients were estimated, andt- tests were performed on the difference in means for each pair of measurements.RESULTSThe study population consisted of middle-aged, mainly Caucasian, mostly married, moderately educated women who were all participants of the breast screening program and were free of breast cancer at the time of participation in this study (Table 1).TABLE 1. Sociodemographic characteristics of female participants for study of reliability of lifetime physical activity assessment (Alberta 1995,N = 113).The reliability of the physical activity assessment was first examined by estimating and comparing the means, standard deviations, and ranges for each of the two measurements of activity made in this study. Descriptive statistics were estimated for all comparisons examined, and those for lifetime activities are presented (Table 2). For this population, the average number of hours spent in total physical activity per week over their lifetimes, for the two measurements taken, was 55.7 h. The majority of this time per week was spent in household activity (29.8 h), less time was spent in occupational activity (22.3 h), and relatively little time was allocated to sports and exercise activities (3.7 h).TABLE 2. Means for repeat† measurements of lifetime physical activity assessment by type of activity (Alberta 1995, N = 113).The reliability of the physical activity instrument was further assessed by estimating the correlations between the two measurements of activity by type of activity over each time period (Table 3). The reliability of total lifetime activity was estimated as 0.74. Occupational activity had the highest correlation of any type of activity for all age periods estimated at 0.87. The lowest correlation for lifetime activity was for exercise/sports activities at 0.72.TABLE 3. Pearson correlation coefficients for reliability† of physical activity assessment, by type of activity and time period (Alberta 1995, N = 113).Since each type of activity was also reported in this study, the means and standard deviations (Table 4) and the correlation coefficients (Table 5) by intensity levels for each type of activity and for total activity and by age periods are also presented. The means presented are for the first measurement of physical activity. The respondents spent, on average over their lifetimes, 18.8 h per week in light activity, 22.4 h in moderate activity, and 5.8 h in heavy activity for all types of activity combined. The remainder of time per week would have been spent in sedentary or resting activities that were not included in these measurements of activity. The total hours per week spent in all activities combined peaked during the middle age groups (18-50 yr) largely because of light and moderate household activities.TABLE 4. Means and standard deviations for first measurements of physical activity, by intensity level and by time period (Alberta 1995, N = 113).TABLE 5. Correlation coefficients* for reliability† of physical activity, by intensity level and by time period (Alberta 1995,N = 113).The highest correlations were found for occupational activity, followed by household and then by exercise/sports activities. Within each type of activity, the highest correlations were generally observed for heavy activity and lowest for light activity.For each type of activity, the lifetime reports of activity had the highest correlations and the reference year the lowest. A clear pattern of decreasing or increasing correlations with length of recall period was not observed. Correlations were sometimes lower for most distant time periods; however, this was not a consistent finding across type of activity. Indeed, correlations for different age periods were fairly comparable across age categories.The t-tests of the differences in means for the two physical activity assessments revealed few large differences in means, no systematic over- or underestimation from the first to second assessment, and only one set of assessments that were statistically significantly different. These were for light household activities for ages 18-50+ (data not shown). Similarly, total light lifetime physical activity assessments were significantly different for ages 18-34 and 35-50. No other important differences in means were observed.DISCUSSIONThis study tested the reliability of reporting of lifetime physical activity patterns with a newly designed interview-administered questionnaire. This questionnaire, developed using the cognitive model, was administered using a recall calendar and cognitive interviewing methods to improve the participants' ability to report their past activity patterns. The questionnaire was found to have good test-retest reliability as assessed by examining the correlations for different types of activity, time periods, and intensity levels. Of most interest for epidemiologic studies are the psychometric properties of this instrument for measuring total lifetime physical activity patterns. For these properties, this questionnaire was most acceptable.The questionnaire, as developed, must be interviewer-administered rather than self-administered and needs interviewers trained in cognitive interviewing methods. We found that interviewers could be trained during a 2-d intensive workshop to use these methods effectively. Furthermore, given the level of detail and the amount of information being recalled, we felt that a lifetime assessment using a self-administered questionnaire would be much more difficult to obtain. In its present format, the questionnaire was easy to administer and to complete.Correlations for the total hours per week are those that are etiologically most relevant in epidemiologic studies since these combine the information on intensity, duration, and frequency of activity into one measure. Of these correlations, those for total physical activity in each time period are of most interest and, likewise, lifetime activity represents the best estimate of the respondents' physical activity patterns over their entire lives. Hence, it is of interest that the correlations for lifetime activity were among the highest correlations measured regardless of the type of activity examined. The lowest correlations were those for the reference year, suggesting that lifetime, or generic, memory is better than the episodic memory that is needed for the reporting of a specific time period such as the reference year. Studies of cognitive processes during dietary recall, which require similar types of recall as for physical activity assessments, have shown that respondents base their reports largely on their generic knowledge of their diets rather than on specific memories (29). Indeed, Smith et al. (29) have suggested that epidemiologists should be asking respondents to use their generic memory since measurement errors exist for any recall period that is longer than the previous few hours. Cognitive survey research has shown that individuals report usual patterns of events more accurately than events that occur irregularly since these patterns can be more readily recalled from their generic memory (19). Since this questionnaire is focused on patterns of activity that are constant for specified time periods, generic memory is more important than episodic memory (19).Within types of activities, the highest correlations were found for occupational, then household and exercise/sports activities. Occupational histories are fairly easy to recall and describe for most respondents. These findings are supported by an evaluation done of the accuracy of occupational activity collected from seven survey instruments (1). In this evaluation, the test-retest reliability was high for most occupational questions (r ≥ 0.63). Occupational activities are constant over a longer time period, are salient to the respondent, and therefore are easier to remember and quantify. The respondents also had a good understanding of household activities that they regularly performed and could recall these quite well. The poorest recall was for exercise/sports activities, since these are much more subject to variation than are the other two types of activity. Cognitive research has indicated that activities that are repetitive and regular are more easily recalled than those that are occasional and irregular(1).Similarly, when intensity of activities was examined, heavy activity was more salient and more easily remembered and reported. Light activities are the most difficult to recall and are also the most common. For all of these assessments, there was generally an increase in the correlations from light to heavy activities. These findings are in agreement with previous research that indicates that light activities are the most difficult to report and heavy activities are most easily recalled(6,17,28).Although research has shown that recall accuracy decreases with the increase in the length of time since the event (10), this study did not show any differences in the reliability of recall by the length of time elapsed. These results suggest that the recall calendars used were effective in increasing the reliability of dating the recall for specific time periods in the respondents' lives.Our questionnaire's psychometric characteristics can be compared with the Kriska lifetime physical activity questionnaire developed for the Pima Indians(20). That questionnaire achieved test-retest reliability coefficients of 0.94 for historical leisure and occupational activity combined for a subsample of the population (N = 18) that were 21-36 yr of age at the time of data collection. These results may be an overestimate of the reliability of the questionnaire since the small study sample was likely quite conscientious, and the data were collected 1-3 wk apart. In other studies that examined test-retest reliability of physical activity questionnaires that measured long-term recall (up to 1 yr), correlations were generally in the range of 0.60-0.90 for leisure and household activities(13,16,17,20,25,26). Thus, given that the current questionnaire collected data 6-8 wk apart and that lifetime activity, rather than just 1 yr, was being reported, this instrument can be considered reliable, to have content and face validity, and, therefore, to be a potential tool for assessing physical activity patterns in etiologic studies of any chronic disease.The limitations of this study must also be considered. First, the study sample was restricted to women who were of the same target age group as the respondents for the main study of breast cancer that is ongoing. Hence, the results of this pilot study cannot be generalized to men or to women who were not of the same age group (50-75 yr). However, since the questionnaire has an open format, it should be equally effective for different age groups and for male respondents. In addition, the respondents in this study may have been more health conscious since they were all participants in the breast screening program. Consequently, they may have been more willing to participate and may have had better recall ability.Second, since this was a pilot study operating under time and cost constraints, the sample size was relatively small. The correlation coefficients were, consequently, less stable and precise as could be expected from a larger study.Third, the inter-rater reliability of this instrument was not examined for time constraint reasons but should be considered in future research on this questionnaire. Likewise, the validity of long-term recall of physical activity patterns was not tested since we did not have the methods available to make this assessment. To examine validity, either a cohort of respondents is necessary for whom self-reported past physical activity was available or for whom past physical fitness measures had been taken. Physical activity would need to have been measured using similar parameters as reported in this questionnaire (i.e., type of activity, intensity, frequency, and duration of activity). Since such a cohort was not available at the time of this pilot, such an assessment was not possible. The determinants of recall ability were also not examined in this study but should be considered in future research to refine the questionnaire even further.Fourth, the correlations observed for this reliability study may have been underestimated for several reasons. The first reason for a possible underestimation was that the testing of this questionnaire coincided with a training and learning period for the interviewers. Second, the physical activity questions were only a part of a larger questionnaire on risk factors for breast cancer, and the interview was fairly long (1½-2 h). Third, these sports and exercise correlations are probably underestimates of those achievable for a male population since men generally engage in more vigorous activities with greater intensity(9,11,24,27). These activities are often structured exercise activities (e.g., team sports) that are more readily recalled.In summary, the value of this questionnaire is that it is the first to measure lifetime total physical activity by collecting data on each type of physical activity separately over lifetime and by measuring frequency, intensity, and duration of each activity. It is also the first physical activity questionnaire to be developed, refined, and administered using cognitive-based methods employed in survey research. With memory-probing strategies and a recall calendar, respondents were able to report their lifetime patterns of occupational, household, and exercise/sports activities and their recall was reliable. Another advantage of this questionnaire is that the time periods for reporting are specified by the respondent rather than by the investigator. Different time periods can, therefore, be assessed as appropriate for the particular investigation. Thus, this instrument is innovative, flexible, comprehensive in the detail of physical activity assessed, and especially designed to maximize a respondent's ability to report his/her lifetime physical activity patterns. Since physical activity is a potential means for the primary prevention and rehabilitation for a number of chronic diseases, this instrument should be of value for numerous studies.The authors thank Kathleen Douglas, Laura Godard, and Zeva Mah for assistance in the data collection for the project and Shelley Cooper, Marilyn Dickson, Shirley Baker, and Theresa Esteves for interviewing the study subjects.REFERENCES1. Ainsworth, B. E., D. R. Jacobs, Jr., A. S. Leon, M. T. Richardson, and H. J. Montoye. Assessment of the accuracy of physical activity questionnaire occupational data. J. Occup. Med. 35:1017-1027, 1993. [Full Text] [Medline Link] [Context Link]2. Ainsworth, B. E., M. Richardson, D. R. Jacobs, Jr., and A. S. Leon. Gender differences in physical activity. Women Sport Physiol. Act. J. 2:1-16, 1993. [Context Link]3. Ainsworth, B. E., H. J. Montoye, and A. S. Leon. Methods of assessing physical activity during leisure and work. In: Physical Activity, Fitness and Health, Ch. 9, C. Bouchard, R. J. Shephard, and T. Stephens (Eds.). Champaign, IL: Human Kinetics, 1993, pp. 146-159. [Context Link]4. Ainsworth, B. E., W. L. Haskell, A. S. Leon, H. J. Montoye, J. F. Sallis, and R. S. Paffenbarger, Jr. Compendium of physical activities: classification of energy costs of human physical activities.Med. Sci. Sports Exerc. 25:71-80, 1993. [CrossRef] [Full Text] [Medline Link] [Context Link]5. Anshel, M. H., P. Freedson, J. Hamill, K. Haywood, M. Horwat, and S. A. Plowman. Dictionary of the Sports and Exercise Sciences. Champaign IL: Human Kinetics, 1991. [Context Link]6. Blair, S. N., M. Dowda, R. R. Rate, et. al. Reliability of long-term recall of participation in physical activity by middle-aged men and women. Am. J. Epidemiol. 133:266-275, 1991. [Context Link]7. Bradburn, N. M., L. J. Rips, and S. K. Shevell. Answering autobiographical questions: the impact of memory and inference on surveys.Science 236:157-161, 1987. [CrossRef] [Medline Link] [Context Link]8. Brewer, M. B., V. T. Dull, and J. B. Jobe. Social cognition approach to reporting chronic conditions in health surveys.Vital Health Stat. 6:1-13, 1989. [Context Link]9. Canadian Fitness Lifestyle Research Institute.Progress in Prevention. Ottawa, 1995. [Context Link]10. Coughlin, S. S. Recall bias in epidemiologic studies.J. Clin. Epidemiol. 43:87-91, 1990. [CrossRef] [Medline Link] [Context Link]11. Deem, R. Unleisured lives: sport in the context of women's leisure. Women's Studies 10:423-432, 1987. [Context Link]12. Durante, R. and B. E. Ainsworth. The recall of physical activity: using a cognitive model of the question-answering process.Med. Sci. Sports Exerc. 28:1282-1291, 1996. [CrossRef] [Full Text] [Medline Link] [Context Link]13. Folsom, A. R., C. J. Caspersen, H. L. Taylor, et al. Leisure time physical activity and its relationship to coronary risk factors in a population-based sample: the Minnesota heart survey. Am. J. Epidemiol. 121:570-579; 1985. [Context Link]14. Friedenreich, C. M. Improving long-term recall in epidemiologic studies. Epidemiology 5:1-4, 1994. [CrossRef] [Full Text] [Medline Link] [Context Link]15. Godin, G. and R. J. Shephard. A simple method to assess exercise behavior in the community. Can. J. Appl. Sport Sci. 10:141-146, 1985. [Context Link]16. Jacobs, D. R., Jr., L. P. Hahn, W. L. Haskell, P. Pirie, and S. Sidney. Validity and reliability of short physical activity history: CARDIA and the Minnesota Heart Health Program. J. Cardiopulm. Rehab. 9:448-459, 1989. [CrossRef] [Full Text] [Context Link]17. Jacobs, D. R., Jr., B. E. Ainsworth, T. J. Hartman, and A. S. Leon. A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med. Sci. Sport Exerc. 25:81-91, 1993. [CrossRef] [Full Text] [Medline Link] [Context Link]18. Jobe, J. B. and D. J. Mingay. Cognitive research improves questionnaires. Am. J. Public Health 79:1053-1055, 1989. [CrossRef] [Medline Link] [Context Link]19. Jobe, J. B., R. Tourangeau, and A. F. Smith. Contributions of survey research to the understanding of memory. Appl. Cognit. Psychol. 7:567-584, 1993. [Context Link]20. Kriska, A. M., W. C. Knowler, R. E. LaPorte et al. Development of questionnaire to examine relationship of physical activity and diabetes in Pima Indians. Diabetes Care 13:401-411, 1990. [CrossRef] [Medline Link] [Context Link]21. Kriska, A. M., R. B. Sandler, J. A. Cauley, R. E. LaPorte, D. L. Horn, and G. Pambianco. The assessment of historical physical activity and its relationship to adult bone parameters. Am. J. Epidemiol. 127:1053-1063, 1988. [Context Link]22. Lessler, J., R. Tourangeau, and W. Salter. Questionnaire design in the cognitive research laboratories: results of an experimental prototype. Vital Health Stat. 6:1-31, 1989. [Context Link]23. Means, B., A. Nigam, M. Zarrow, E. F. Loftus, and M. S. Donaldson. Autobiographical memory for health-related events: enhanced memory for recurring incidents. Vital Health Stat. 6:1-22, 1989. [Context Link]24. Means, B. and E. F. Loftus. When personal history repeats itself: decomposing memories for recurring events. Appl. Cognit. Psychol. 5:297-318, 1991. [Context Link]25. Pols, M. A., P. H. Peeters, H. C. Kemper, and H. J. Collette. Repeatability and relative validity of two physical activity questionnaires in elderly women. Med. Sci. Sports Exerc. 28:1020-1025, 1996. [CrossRef] [Full Text] [Medline Link] [Context Link]26. Richardson, M. T., A. S. Leon, D. R. Jacobs, Jr., B. E. Ainsworth, and R. Serfass. Comprehensive evaluation of the Minnesota leisure time physical activity questionnaire. J. Clin. Epidemiol. 47:271-281, 1994. [CrossRef] [Medline Link] [Context Link]27. Shaw, S. M. Gender and leisure: inequality in the distribution of leisure time. J. Leisure Res. 17:266-282, 1985. [Medline Link] [Context Link]28. Slattery, M. L. and D. R. Jacobs, Jr. Assessment of ability to recall physical activity of several years ago. Ann. Epidemiol. 5:292-296, 1995. [Context Link]29. Smith, A. F., J. B. Jobe, and D. J. Mingay. Retrieval from memory of dietary information. Appl. Cognit. Psychol. 5:269-296, 1991. [Context Link]30. Tourangeau, R. Cognitive science and survey methods. In: Cognitive Aspects of Survey Methodology: Building Bridges between Disciplines, T. B. Jabine, M. L. Straf, J. M. Tanur, and R. Tourangeau(Eds.). Washington DC: National Academy Press, 1984, pp. 73-100. [Context Link]31. Wilbur, J. E., A. Miller, A. J. Dan, and K. Holm. Measuring physical activity in midlife women. Public Health Nurs. 6:120-128, 1989. [CrossRef] [Medline Link] [Context Link]32. Willis, G. B. Cognitive interviewing and questionnaire design: a training manual. Cognitive Methods Staff Working Paper Series, No. 7, Office of Research and Methodology, National Center for Health Statistics, 1994. [Context Link]33. Willis, G. B., P. Royston, and D. Bercini. The use of verbal report methods in the development and testing of survey questionnaires.Appl. Cognit. Psychol. 5:251-267, 1991. [Context Link]APPENDIX 1. LIFETIME TOTAL PHYSICAL ACTIVITY QUESTIONNAIREThe next section will be about your physical activity patterns over your lifetime. Specifically, I will be asking you about your occupational, household and exercise/sports activities. Starting with your occupational activities, please tell me what jobs (paid or volunteer) that you have doneat least 8 hours a week for four months of the year over your lifetime. We will start with your first job and end with the job that you had in your reference year.Please describe the job that you had, the age that you started working at this job and the age when you ended doing this particular job. For each job we also need to know the number of years, the number of months per year, the number of days per week, the number of hours per day and the intensity of the job.Household ActivitiesNow I am going to ask you to report what household and gardening activities that you have done over your lifetime. Again, we will start with your past activity and then continue up to your reference year. Please include only those activities that you have done at least 7 hours per week for 4 months of the year. It may help you to consider what a typical day is for you. Then think about how many hours of household and gardening or yard work you do in a typical day. For seasonal activities, such as gardening, you can report those separately from all other household activities that are done all year.Exercise/Sports ActivitiesNow I would like to know all your exercise or sports activities that you did during your lifetime starting with childhood and continuing to your reference year. Please report the activities that you have done at least2 hours per week for at least 4 months of the year. Please tell us what exercise and sports activities you have done at least 10 times during your lifetime. Besides sports and exercise, we are also interested in knowing whether you walked or biked to work. If you have done this, please report all the information as for the other sports activities. Please begin by telling me the activities that you did during your school years including your physical education (gym) classes. TableTable Table [Context Link]No Caption Available.No Caption Available.No Caption Available.APPENDIX 2: ESTIMATION OF OUTCOME VARIABLESThe outcome variables that were used to examine the reliability of the questionnaire were estimated as follows:a) Average number of hours per week spent in occupational activity over lifetime = Equation The average number of hours per week spent in occupational activity over a lifetime was estimated separately for sedentary, light, moderate, and heavy occupational activity.Equation 1Ab) Average number of hours per week spent in household activity over lifetime = Equation Average number of hours per week spent in household activity over lifetime was estimated separately for light, moderate, and heavy household activity.Equation 1Bc) Average number of hours per week spent in exercise/sports activities over lifetime =if respondent reported per day: EquationEquation 1Cif respondent reported per week: EquationEquation 1Dif respondent reported per month: EquationEquation 1Eif respondent reported per year: Equation [Context Link]Equation 1F BREAST CANCER; MEASUREMENT; METHODS; PHYSICAL ACTIVITY; RELIABILITY;|00005768-199802000-00015#xpointer(id(R1-15))|11065404||ovftdb|00005122-199310000-00012SL00005122199335101711065404P61[Full Text]|00005768-199802000-00015#xpointer(id(R1-15))|11065405||ovftdb|00005122-199310000-00012SL00005122199335101711065405P61[Medline Link]|00005768-199802000-00015#xpointer(id(R4-15))|11065213||ovftdb|00005768-199301000-00011SL000057681993257111065213P64[CrossRef]|00005768-199802000-00015#xpointer(id(R4-15))|11065404||ovftdb|00005768-199301000-00011SL000057681993257111065404P64[Full Text]|00005768-199802000-00015#xpointer(id(R4-15))|11065405||ovftdb|00005768-199301000-00011SL000057681993257111065405P64[Medline Link]|00005768-199802000-00015#xpointer(id(R7-15))|11065213||ovftdb|SL00007529198723615711065213P67[CrossRef]|00005768-199802000-00015#xpointer(id(R7-15))|11065405||ovftdb|SL00007529198723615711065405P67[Medline Link]|00005768-199802000-00015#xpointer(id(R10-15))|11065213||ovftdb|SL000050771990438711065213P70[CrossRef]|00005768-199802000-00015#xpointer(id(R10-15))|11065405||ovftdb|SL000050771990438711065405P70[Medline Link]|00005768-199802000-00015#xpointer(id(R12-15))|11065213||ovftdb|00005768-199610000-00012SL00005768199628128211065213P72[CrossRef]|00005768-199802000-00015#xpointer(id(R12-15))|11065404||ovftdb|00005768-199610000-00012SL00005768199628128211065404P72[Full Text]|00005768-199802000-00015#xpointer(id(R12-15))|11065405||ovftdb|00005768-199610000-00012SL00005768199628128211065405P72[Medline Link]|00005768-199802000-00015#xpointer(id(R14-15))|11065213||ovftdb|00001648-199401000-00001SL0000164819945111065213P74[CrossRef]|00005768-199802000-00015#xpointer(id(R14-15))|11065404||ovftdb|00001648-199401000-00001SL0000164819945111065404P74[Full Text]|00005768-199802000-00015#xpointer(id(R14-15))|11065405||ovftdb|00001648-199401000-00001SL0000164819945111065405P74[Medline Link]|00005768-199802000-00015#xpointer(id(R16-15))|11065213||ovftdb|00008483-198911000-00003SL000084831989944811065213P76[CrossRef]|00005768-199802000-00015#xpointer(id(R16-15))|11065404||ovftdb|00008483-198911000-00003SL000084831989944811065404P76[Full Text]|00005768-199802000-00015#xpointer(id(R17-15))|11065213||ovftdb|00005768-199301000-00012SL000057681993258111065213P77[CrossRef]|00005768-199802000-00015#xpointer(id(R17-15))|11065404||ovftdb|00005768-199301000-00012SL000057681993258111065404P77[Full Text]|00005768-199802000-00015#xpointer(id(R17-15))|11065405||ovftdb|00005768-199301000-00012SL000057681993258111065405P77[Medline Link]|00005768-199802000-00015#xpointer(id(R18-15))|11065213||ovftdb|SL00000469198979105311065213P78[CrossRef]|00005768-199802000-00015#xpointer(id(R18-15))|11065405||ovftdb|SL00000469198979105311065405P78[Medline Link]|00005768-199802000-00015#xpointer(id(R20-15))|11065213||ovftdb|SL0000345819901340111065213P80[CrossRef]|00005768-199802000-00015#xpointer(id(R20-15))|11065405||ovftdb|SL0000345819901340111065405P80[Medline Link]|00005768-199802000-00015#xpointer(id(R25-15))|11065213||ovftdb|00005768-199608000-00012SL00005768199628102011065213P85[CrossRef]|00005768-199802000-00015#xpointer(id(R25-15))|11065404||ovftdb|00005768-199608000-00012SL00005768199628102011065404P85[Full Text]|00005768-199802000-00015#xpointer(id(R25-15))|11065405||ovftdb|00005768-199608000-00012SL00005768199628102011065405P85[Medline Link]|00005768-199802000-00015#xpointer(id(R26-15))|11065213||ovftdb|SL0000507719944727111065213P86[CrossRef]|00005768-199802000-00015#xpointer(id(R26-15))|11065405||ovftdb|SL0000507719944727111065405P86[Medline Link]|00005768-199802000-00015#xpointer(id(R27-15))|11065405||ovftdb|SL0001161819851726611065405P87[Medline Link]|00005768-199802000-00015#xpointer(id(R31-15))|11065213||ovftdb|SL000066201989612011065213P91[CrossRef]|00005768-199802000-00015#xpointer(id(R31-15))|11065405||ovftdb|SL000066201989612011065405P91[Medline Link]2813255The Lifetime Total Physical Activity Questionnaire: development and reliabilityFRIEDENREICH, CHRISTINE M.; COURNEYA, KERRY S.; BRYANT, HEATHER E.Basic Sciences: Epidemiology230InternalMedicine & Science in Sports & Exercise10.1249/MSS.0b013e318186b1b12009412336-343FEB 2009Physical Activity Questionnaire Comprehension: Lessons from Cognitive InterviewsALTSCHULER, A; PICCHI, T; NELSON, M; ROGERS, JD; HART, J; STERNFELD, B & Science in Sports & Exercise10.1249/01.mss.0000210196.94952.8b2006384715-719APR 2006Reliability of a Measure of Prediagnosis Physical Activity for Cancer SurvivorsDUNN, J; AITKEN, JF; LYNCH, BM; OWEN, N; NEWMAN, B; PAKENHAM, K; LEGGETT, B & Science in Sports & Exercise2003355862-866MAY 2003Distant Past Exercise in Women: Measures May Be Reliable, But Are They Valid?DAWSON, J; JUSZCZAK, E; THOROGOOD, M; FOSTER, C; MARKS, S; DODD, C; FITZPATRICK, R & Science in Sports & Exercise10.1249/01.mss.0000222841.96885.a820063861121-1131JUN 2006Development and Reproducibility of the Bone Loading History QuestionnaireDOLAN, SH; WILLIAMS, DP; AINSWORTH, BE; SHAW, JM & Science in Sports & Exercise2001336S370-S378JUN 2001Quantifying energy expenditure and physical activity in the context of dose responseLAMONTE, MJ; AINSWORTH, BE & Science in Sports & Exercise19993191307-1312SEP 1999The Physical Activity for Total Health (PATH) Study: rationale and designMcTIERNAN, A; ULRICH, CM; YANCEY, D; SLATE, S; NAKAMURA, H; OESTREICHER, N; BOWEN, D; YASUI, Y; POTTER, J; SCHWARTZ, R & Science in Sports & Exercise20013391538-1545SEP 2001Relation between intensity of physical activity and breast cancer risk reductionFRIEDENREICH, CM; COURNEYA, KS; BRYANT, HE & Science in Sports & Exercise20043691559-1566SEP 2004Multivariate Genetic Analysis of Lifetime Exercise and Environmental FactorsSIMONEN, R; LEVÄLAHTI, E; KAPRIO, J; VIDEMAN, T; BATTIÉ, MC Journal of Cardiovascular Prevention & Rehabilitation2005122102-114APR 2005How to assess physical activity? How to assess physical fitness?Vanhees, L; Lefevre, J; Philippaerts, R; Martens, M; Huygens, W; Troosters, T; Beunen, G