Preventive Medicine 28, 349–360 (1999) Article ID pmed.1998.0429, available online at http://www.idealibrary.com on Stage of Regular Exercise and Health-Related Quality of Life1 Robert G. Laforge, Sc.D.,*,2 Joseph S. Rossi, Ph.D.,* James O. Prochaska, Ph.D.,* Wayne F. Velicer, Ph.D.,* Deborah A. Levesque, Ph.D.,* and Colleen A. McHorney, Ph.D.† *University of Rhode Island, Kingston, Rhode Island; and †Department of Preventive Medicine, University of Wisconsin–Madison Medical School, Madison, Wisconsin Background. Research on cognitive factors and motivational readiness for exercise is important for increasing our understanding of behavior change among those with sedentary lifestyles. This study examines stage of change for regular exercise and self-perceived quality of life. Methods. Data are from 1,387 respondents to a random digit dial survey of health behaviors. Stage of change is assessed with a single item, and individuals are classified with respect to intention and exercise behavior. Quality of life is assessed with the SF-36, a multidimensional measure of health-related quality of life. Results. Exercise stage is associated with self-perceived quality of life. The three areas most strongly related were physical functioning, general health perceptions, and vitality. Physical functioning scores were lowest in precontemplation and highest in maintenance. Vitality and mental health scales were related to exercise behavior, but not to intention. Conclusions. Cognitions about self-perceived quality of life vary across the stages of change, with those who are least prepared to adopt regular exercise reporting the lowest levels of quality of life. These findings suggest that cognitive–motivational messages designed to emphasize quality of life benefits associated with exercise may be useful intervention strategies for people who are less motivationally ready to change. q1999 American Health Foundation and Academic Press Key Words: exercise; stage of change; quality of life; health surveys. INTRODUCTION There is a substantial literature indicating that physical activity and regular exercise can lead to improvements in quality of life along a broad range of physical 1 Supported in part by funds from the National Cancer Institute, Grants R01 CA28721 and P01 CA50087. 2 To whom correspondence and reprint requests should be addressed at, Cancer Prevention Research Center, University of and psychological dimensions [1,2]. Cross-sectional and longitudinal studies have found that increasing amounts of physical activity are associated with decreased overall mortality and better health and functioning [1,3], including a reduction in symptoms of coronary heart disease [4], cancer [5], osteoporosis [6], diabetes [7], anxiety [3], and depression [3]. However, millions of Americans do not engage in regular physical activity [8]. National public health objectives for the year 2000 include increasing the frequency and duration of light, moderate, and vigorous physical activity [9], and major initiatives are underway to promote exercise adoption and adherence in the general population [10]. Theoretical models that focus on cognitive and behavioral determinants of behavior change—such as self-efficacy [11], stages of change [12], decision making [13], and intention [14]—are increasingly being used to aid development of interventions designed to promote regular exercise [15–18]. An important intervention strategy of cognitive– behavioral models is to attempt to modify self-perceptions, attitudes, and intentions, which have been found to be determinants of exercise behavior [1,2]. For example, cognitive interventions that increase the awareness of health risks associated with sedentary lifestyles can enhance motivation to exercise [19]. Exercise interventions that classify populations by stage of motivational readiness for change can provide targeted cognitive messages appropriate to each individual’s stage of readiness to change [15,20]. This line of research goes beyond the simple distinction between physically active and sedentary individuals and attempts to understand the determinants of motivational readiness along an underlying continuum of stages of change, spanning from precontemplation (not intending to change) to maintenance (sustained regular exercise or physical activity over time). Stage-matched interventions have Rhode Island, 2 Chafee Road, Kingston, RI 02881. Fax: (401) 8745562. E-mail: WAA101@URIACC.URI.EDU. 349 0091-7435/99 $30.00 Copyright q 1999 by American Health Foundation and Academic Press All rights of reproduction in any form reserved. 350 LAFORGE ET AL. been shown to effectively increase physical activity adoption in a nonrandomized community setting [17] and in a recent randomized clinical trial [21]. Understanding how cognitions vary at different stages of change can provide the basis for continued progress in the development of stage-matched intervention strategies [22]. In this regard, related concepts from different theoretical models are increasingly being combined in an effort to better explain exercise acquisition and adherence [20–22]. Research into how stages of change for exercise is related to specific cognitions— such as perceived social pressure, intention, attitudes, decisional balance, self-efficacy, processes of change, perceived control, and perceived severity of the consequences of inactivity—has provided many new insights into how exercise promotion interventions using the stage of change construct can be made more effective [16,18,19,21–23]. Since modification of self-perception of the benefits of regular exercise is an important motivational strategy used in many exercise promotion interventions, it is meaningful to study whether, or to what degree, self-perception of quality of life varies across stage of change. This study extends the research literature on exercise behavior and quality of life by examining the relationship between motivational readiness and self-perceived health-related quality of life. Health-related quality of life is a multidimensional construct that represents a person’s overall satisfaction with life, and its assessment typically involves measures of functional status in the domains of physical, cognitive, emotional, and social health [24,25]. A better understanding of how the behavioral and intentional components of the stage construct are related to the various dimensions of health-related quality of life may be useful for evaluating existing stage-based intervention programs, as well as for its potential for providing new strategies for more effective matching of individual needs with exercise interventions. In sum, researchers have begun to consider classification by stage of motivational readiness as a meaningful method of differentiating sedentary and active populations. Although there is evidence that physical activity can have a positive influence on different dimensions of health-related quality of life [26,27], there have been few studies that have used the multidimensional measurement approach [1,2]. This study is unique because it examines the relationship between stage of readiness to exercise and an established multidimensional measure of self-perceived quality of life (the SF-36) with data from a large general population sample. Further it explores how cognitions concerning self-perceived health-related quality of life are associated with stage of change for regular exercise, and these results are considered in the context of the growing literature on cognitive processes involved in exercise behavior that have been found to be related to stage of change. METHODS Sample The study sample consisted of 1,387 respondents ages 18–75 to a random digit dial telephone survey conducted in Rhode Island in 1992. The next-birthday method of sample selection within a household was used to minimize respondent bias [28]. The survey was administered over the phone by interviewers trained in standardized interviewing techniques [29], and the survey procedures were approved by a University Institutional Review Board. The household response rate was 75.1%. MEASURES Stage of Change for Exercise Stage of change for exercise behavior is a theoretical construct that combines self-reported intention and behaviors to allow classification of populations with respect to readiness for behavioral change. Stage of change is the temporal dimension of a more general theoretical model, the Transtheoretical model [20]. This model also includes other constructs, such as the processes of change, decisional balance, and self-efficacy, that describe how human behavior changes. Assessment of stage of readiness to engage in, or adopt, regular exercise offers a profile of the population with respect to the degree of intention or current adherence to a regime of regular physical exercise. Measurement of the stage of change construct for exercise has been generally found to be valid and robust across various methods of assessment using different item and response formats [30]. These include a “ladder” format with descriptive anchors [31], a 6-item true/ false response format [32], a 32-item descriptive statement scale [33], an algorithm for each stage that uses a 5-choice Likert response format [34], a 1-item algorithm with 5 distinct response choices, [22,30], and other approaches that vary in the theoretical consistency of their stage definitions [35,36]. Because of its simplicity for use in survey research, we used the 1-item algorithm with 5 distinct response choices. This was the recommended format reported in a recent comparison of several staging methods [30]. In this study, stage of change for regular exercise was assessed with a single item with five mutually exclusive response categories. The question stated “Do you consistently get regular exercise, that is, 3 times a week, for 20 minutes each time?” The quantity and frequency of exercise are based on recommendations developed by the American College of Sports Medicine [37]. Respondents were required to select the category that best STAGE OF EXERCISE AND QUALITY OF LIFE described their current condition, and the five response categories were used to classify stage of change. The response categories are (1) “Yes, and I have been doing so for more than 6 months” (maintenance); (2) “Yes, but I have been doing so for less than 6 months” (action); (3) “No, but I intend to in the next 30 days” (preparation); (4) “No, but I intend to in the next 6 months” (contemplation); and (5) “No, and I don’t intend to in the next 6 months” (precontemplation). This short fivechoice staging algorithm has been found to reliably assess stage of regular exercise in several large population studies [30,38,39]. It is an adaptation of the original stages of change measure developed for smoking cessation [12]. This exercise staging approach does not provide differentiation by degree of exercise intensity— e.g., vigorous or moderate—nor does it provide examples such as running, swimming, and so forth, to indicate intensity level. Rather, it reflects an increasing degree of immediacy of intention to engage in the target criterion of exercise across the first three stages (precontemplation to preparation), and increasing duration of time engaged in the target behavior—consistent regular exercise at least 3 times a week for 20 minutes each time—across the last two stages (action and maintenance). This algorithm differs slightly in its definition of the preparation stage from the short-item exercise stage algorithms employed in some other research based on the Transtheoretical model, where the preparation stage is indicated by an item that reflects making small behavioral changes [22,34]. In our stage algorithm the preparation stage does not explicitly require that small behavioral changes have been made, but rather represents more proximal intention to engage in the target behavior (“in the next 30 days”). It is likely, however, that increasing intention to engage in consistent regular exercise at the criterion level also involves engaging in physical activity, but at subcriterion levels. The similarity in the stage distributions found using this algorithm [30,38,39] compared with brief exercise stage algorithms reported in other studies [22,30,34] suggests that the stage construct is robust, and the differences in the preparation stage estimates are relatively minor. Brief instruments can be reliable and valid indicators of exercise stage [22,30] and exercise behavior [40] and are particularly important for use in complex epidemiological studies, which often require long questionnaires and respondent burden to be minimized. Although exercise stage has been assessed using a variety of different methods [30,31,35,34], there is consistent evidence of construct validity across various populations, with the stages found to be associated with increasing amounts of physical activity [22,30–32,34,41,42]. For example, Cardinal [41] reported that exercise behavior, body mass index, and cardiorespiratory fitness increased 351 from precontemplation to maintenance, after controlling for demographics and social desirability. Acceptably high levels of test–retest reliability (k $ 0.78) have been reported for the simple stage measures [22,34]. A recent comparison of several methods of assessing exercise stage found that the single-item exercise stage algorithm used in this study outperformed other staging algorithms in predicting hours of exercise performed, a decisional balance score for exercise, and a measure of confidence (self-efficacy) in the ability to exercise [30]. Quality of Life Assessment—The Medical Outcome Study (MOS) SF-36 Over the past 2 decades, there have been considerable advances in the measurement of health-related quality of life for both general and clinically ill populations. The short-form 36-item (SF-36) is a recent addition to a host of generic measures [25,43]. The SF-36 instrument was designed to provide a valid measure of health functioning that is easy to use in surveys of large populations. It has been developed for self-administration by persons 14 years or older and for administration over the telephone. The SF-36 was constructed for use in monitoring the MOS, a longitudinal study of health outcomes among the chronically ill, but it has been widely adopted by researchers conducting studies of clinical and nonclinical populations [43]. For example, it has been used in studies of health services, methodology, clinical trials [43,44], chronic conditions in the general population [45], and treatment outcome studies across a variety of health conditions [43,44]. Ware and colleagues reported in 1994 that the measurement model of the SF-36 had been validated in 62 crosssectional studies and 27 longitudinal studies [43]. The SF-36 is here employed to provide a comparison of health-related quality of life of a general population sample of adults across each of the five stages of exercise adoption. The SF-36 taps both physical and mental health aspects of quality of life by using the respondents’ perspective on their health and functional status. Factor analytic studies of the SF-36 have found eight distinct physical and mental health dimensions in patients participating in the MOS [46] as well as in the general U.S. population [47]. Ware et al. [43] have shown that from 80 to 85% of the reliable variance in the eight scales is accounted for by two hierarchical components, a physical component scale (PCS) and a mental health component scale (MCS). The item content of the eight separate scales and their corresponding PCS and MCS hierarchical components are displayed in Fig. 1. The separate scales assess physical functioning (PF), role functioning due to limitations in physical problems (RP), bodily pain (BP), general health perceptions (GH), general mental health 352 LAFORGE ET AL. FIG. 1. SF-36 measurement model. Adapted from Ref. [43]. (MH), vitality (VT), role limitations due to emotional problems (RE), and social functioning (SF). One final SF-36 item asks respondents to rate the amount of change in their health status over the past year, but it is not used to score any of the eight scales and is not reported on here. The response choices for most items are Likert scales. Items that assess problems in role functioning due to either physical health or emotional problems are dichotomous (yes/no) responses [25]. The SF-36 scales were scored using the standard SF-36 scoring algorithms to allow comparison across the eight dimensions [47]. The eight scale scores range from 0 to 100, with 100 representing optimal health and functioning. In contrast, scoring for the PCS and MCS summary measures used the norm-based scoring methods recommended by Ware and colleagues, who provide extensive evidence of the validity and reliability of the PCS and MCS, as well as a valuable discussion of the usefulness of the SF-36 measurement model [43]. One beneficial feature of using the PCS and MCS is that they have a direct interpretation in relation to the distribution of scores in the general U.S. population, which are defined to have a mean of 50 and a standard deviation of 10. STAGE OF EXERCISE AND QUALITY OF LIFE For example, a score of 55 on the PCS indicates that the study group has better physical health functioning than the general U.S. population and that the difference in the PCS score obtained represents a medium effect size of 0.5 standard deviation [48]. Procedure and Analysis Adult respondents to a random digit dial telephone survey were administered a questionnaire that assessed stage of change for several health behaviors, demographics, and the SF-36. Bivariate analyses involved t tests and one-way ANOVA comparisons of scores on the SF-36 subscales. One-way MANOVA was performed on the SF-36 scales with and without statistical adjustment for age, gender, and years of education. Mean scores on the eight quality of life scales and the two component summary measures are presented across five categories of exercise stage and for age, gender, and education. Post hoc analyses of ANOVA results used the Tukey HSD procedure for adjustment for multiple comparisons. P values for differences across stages are reported, but the Bonferoni adjustment for multiple comparisons across the eight scales should be assumed, resulting in the univariate criterion for statistical significance of P 5 0.006. The (v 2 estimate of the standardized effect size is presented as a measure of the strength of association between exercise stage and the SF-36 components and scales. It represents the proportion of the variance in the SF-36 explained by exercise stage. RESULTS Table 1 presents sample characteristics for age, gender, education, and exercise stage. The results for age TABLE 1 Selected Sample Characteristics Age LT 30 30–40 41–55 GT 55 Gender Male Female Education LT high school High school College Graduate school Exercise stage Precontemplation Contemplation Preparation Action Maintenance N % 381 347 318 337 27.5 25.1 23.0 24.4 515 870 37.0 62.5 187 484 554 158 13.5 35.0 40.1 11.4 250 164 287 82 604 18.0 11.8 20.7 5.9 43.5 353 and education are comparable to the 1990 U.S. Census statistics for the state of Rhode Island [49], suggesting the sample is fairly representative on these characteristics. The sample overrepresents females, but contains a sufficiently large sample of males to make stable estimates of relationships between variables. Nevertheless, gender was not found to be predictive of quality of life and therefore posed little threat to external validity of this sample. The distribution for exercise stage found in this sample is similar to that found in five other surveys conducted in the United States and Australia that used the identical stage item [39], including a sample of approximately 19,000 HMO members [30,38]. Over half of the sample did not report consistent regular exercise, with 18% reporting no intention to consistently engage in regular exercise (the precontemplation stage) in the next 6 months, 12% intending to begin in the next 6 months (contemplation stage), and 21% reporting that they were intending to begin regular exercise in the next 30 days (the preparation stage). Six percent reported consistent regular exercise for less than 6 months (action), and 44% of the sample declared that they have consistently engaged in regular exercise for more than 6 months (maintenance). As expected, scores on the SF-36 indicate that the Rhode Island general population sample was functioning at the high end of the health-related quality of life scales (Table 2). Scores on the SF-36 scales were generally in the high-functioning range, with 100 indicating optimal functional health status. All eight scales were found to have good to excellent internal reliability, as measured by Cronbach’s a. There was little evidence of floor effects in this sample, but several scales (PF, RP, BP, RE, and SF) exhibited significant ceiling effects, that is, had a high percentage of respondents who scored at the highest end of the scoring range. The results reported in Table 2 are remarkably similar to those recently reported for another general population survey using the SF-36, which was also conducted using telephone interviewing [50]. Table 3 presents the SF-36 scores and summary measures for the total population and bivariate tests of association with age, gender, and years of education. The sample population norms for total SF-36 scale scores were very similar to those published for the general U.S. population [43]. Overall the sample scored highest on the PF scale (M 5 85.8) and lowest on the VT scale (M 5 63.5). Likewise, the PCS and MCS indicate that the scores for our Rhode Island sample were quite similar to that of the general U.S. population, differing by less than one-tenth of a standard deviation. The slightly higher scores in the Rhode Island sample— i.e., an average 1.3 points higher on the SF-36 scales than the national norms—may be a consequence of collecting the data by telephone, which has been shown 354 LAFORGE ET AL. TABLE 2 Descriptive Statistics for the SF-36 Scales Scale Mean SD % Floor a % Ceilingb Coefficient a 85.8 80.3 77.9 73.7 63.5 75.3 83.3 84.6 22.4 34.1 25.8 20.7 20.9 18.8 30.3 22.0 0 11 1 0 1 0 7 1 46 69 41 9 3 7 72 53 0.92 0.88 0.80 0.74 0.76 0.78 0.75 0.76 Physical functioning (PF) Role limitations due to physical problems (RP) Bodily pain (BP) General health (GH) Vitality (VT) Mental health (MH) Role limitations due to emotional problems (RE) Social functioning (SF) Note. a, Cronbach’s a, measure of internal consistency reliability of the scale. a % Floor, percentage of respondents with lowest possible score of 0. b % Ceiling, percentage of respondents with highest possible score of 100. to result in slightly higher scores than mail surveys [50]. As expected, increasing age was inversely associated with measures of physical health-related quality of life, as indicated by all four SF-36 scales assessing PF, GH, BP, and RP (Table 3). In contrast, age was not significantly associated with any of the SF-36 scales measuring mental health, but was significantly and directly related to the MCS summary measure of mental healthrelated quality of life. Although the effect sizes for differences were not large, younger people tended to score slightly lower on MH and RE than did older people. Females scored significantly higher than males on the VT scale, but otherwise did not differ from males with respect to health-related quality of life. Education was strongly associated with the two summary measures of PF and MH, as well as with five of the scales measuring PF, GH, RP, RE, AND SF. In general, those with more education scored higher on PF and MH measures of quality of life. The results of analyses with exercise stage reveal a generally consistent pattern indicating that quality of life scale scores increase across the stages from precontemplation to maintenance. The unadjusted MANOVA TABLE 3 Selected Sample Characteristics by SF-36 Scale Scores and Summary Measures Physical health (PCS) scalesa Variable Age LT 30 30–40 41–55 GT 55 Gender Male Female Education LT high school High school College Grad school Total sample Difference from U.S. normsc a Physical Role functioning physical Bodily pain Mental health (MCS) scalesa General PCS MCS health summary Mental Role Social summary perceptions measure b Vitality health emotional functioning measure b 91.3* 91.7 84.0 75.6 86.0* 85.7 79.1 69.3 83.2* 77.9 76.7 73.2 76.5* 78.3 72.2 67.3 53.4* 53.7 49.8 45.8 63.9 65.3 62.4 62.1 74.1 75.2 74.9 76.9 81.7 83.9 83.2 84.4 84.6 84.9 84.7 84.3 49.1* 50.1 50.7 52.6 84.8 87.6 79.6 81.4 76.5 80.2 73.7 73.9 51.2 50.2 61.8* 66.5 74.3 77.0 83.7 82.5 84.3 85.2 51.0 50.3 76.0* 84.4 88.3 94.1 85.8 11.2 71.7* 80.9 80.2 89.3 80.3 20.9 74.3 78.3 77.4 83.5 77.9 12.4 65.6* 74.5 74.3 79.7 73.7 11.5 47.6* 50.5 50.9 53.5 50.6 10.6 69.5 74.9 76.2 80.1 75.3 10.5 74.5* 84.3 83.2 90.7 83.3 12.0 78.5* 86.2 84.0 89.8 84.6 11.0 48.3* 50.9 50.5 52.5 50.6 10.6 60.6 63.8 63.0 67.6 63.5 12.4 SF-36 scale scores range from 0 to 100, with 100 indicating optimal health and functioning. PCS and MCS summary measures are standardized to the US. general population with mean 5 50 and SD 5 10. c Source of U.S. population norms for SF-36 scale and summary scores: [43,47]. * p , 20.006. b 355 STAGE OF EXERCISE AND QUALITY OF LIFE results revealed that exercise stage was significantly related to the linear composite of the eight scales of health-related quality of life (Hotellings T 2 5 0.113, df 5 32; Wilks l 5 0.897, P , 0.0005). (A MANOVA conducted using the PCS and MCS summary scales was similarly related to exercise stage; Hotellings T 2 5 0.042, df 5 8; Wilks l 5 0.959, P , 0.0005.) Age and education remained significant covariates, but gender failed to achieve statistical significance in the multivariate model. Statistical adjustment of the mean scores for age, gender, and education did not change the relationships found in the unadjusted analyses, and resulted in only small, insignificant changes from the observed scores. For the SF-36 scale scores, the amount of variability in the discriminant score attributable to between-group differences in exercise stage was 28.5% in the unadjusted model and 23.9% in the adjusted model. Since statistical adjustment contributed little to the overall relationship with exercise stage, only the unadjusted univariate ANOVA follow-up tests are reported below. The results of the ANOVA and post hoc tests for exercise stage and indices of health-related quality of life are presented in Table 4. The early stages (precontemplation, contemplation, and preparation) represent self reports of increasing degrees of intention to exercise, while the later stages (action and maintenance) reflect increasing duration of time spent engaged in consistent regular exercise. Exercise stage was significantly associated with physical and mental health-related quality of life, including all of the SF-36 scales except the SF scale. Stage of exercise was linearly related to most measures of health-related quality of life, although differences between stages did not achieve statistical significance on all of the scales, as indicated by the results of Tukey HSD tests. The standardized effect sizes (v 2) ranged from medium to small as described by Cohen [48] and are comparable to those in other published literature examining the association of variables with exercise stage [22,23]. The linear relationship was most direct for physical health, as measured by the PF, BP, and GH scales. In these scales, persons in the maintenance stage for regular exercise scored higher than did those in action, who in turn scored higher than those in the contemplation stage, who in turn scored higher than those in the precontemplation stage for regular exercise. A nonlinear relationship was found between exercise stage and the SF-36 scales that tap mental, emotional, and role functioning, such as the scales that assess MH and RE. With respect to MH functioning the strongest differences appeared for those who engaged in regular TABLE 4 Exercise Stage of Change and Health-Related Quality of Life Scores (Standard Deviations) Exercise stage of changeb SF-36 scalea n Physical functioning (PF) Role limitations due to physical problems (RP) Bodily pain (BP) General health (GH) Vitality (VT) Mental health (MH) Role limitations due to emotional problems (RE) Social functioning (SF) Physical health (PCS) c Mental health (MCS) c a PC 250 C 164 P 287 A 82 M 604 Tukey HSD* F P value v2 77.2 (28.3) 72.5 (39.1) 73.0 (29.3) 67.0 (24.2) 57.7 (24.5) 73.8 (21.5) 82.9 (31.5) 82.2 (25.3) 46.9 (12.0) 50.6 (10.6) 83.0 (23.9) 74.8 (38.3) 73.3 (29.3) 69.4 (22.2) 59.8 (19.8) 74.3 (18.3) 80.6 (31.2) 81.7 (23.6) 48.5 (11.7) 49.9 (10.1) 85.8 (20.1) 79.5 (32.7) 77.3 (25.8) 71.5 (19.6) 59.7 (19.7) 72.1 (20.1) 78.9 (33.3) 84.3 (21.4) 50.7 (9.8) 48.8 (10.1) 87.3 (21.5) 86.6 (29.2) 80.4 (19.8) 75.5 (18.1) 66.2 (16.9) 78.6 (15.2) 91.5 (20.2) 87.6 (19.8) 51.1 (8.7) 52.7 (7.0) 90.3 (18.9) 84.7 (30.8) 81.4 (23.2) 78.7 (18.3) 68.5 (19.4) 77.3 (17.3) 85.1 (29.0) 86.3 (20.6) 52.6 (8.5) 51.3 (9.1) PC,P,A,M C,M PC,C ,M PC,A PC,C ,M ,0.00005 0.043 ,0.00005 0.019 ,0.00005 0.017 PC,C,P,M PC,A PC,C,P,M PC,A PC ,A,M ,0.00005 0.049 ,0.00005 0.048 0.0006 0.011 P,A,M 0.0038 0.008 ns 0.0250 0.005 PC,,P,A,M C,M P,A,M ,0.00005 0.043 0.0008 0.011 SF-36 scale scores range from 0 to 100, ruth 100 indicating optimal health and functioning. PC, precontemplation; C, contemplation; P, preparation; A, action; M, maintenance. c PCS and MCS summary measures are standardized to the US. general population with mean 5 50 and SD 5 10. * Significant at adjusted 0.05 level. b 356 LAFORGE ET AL. exercise behavior versus those who did not, and differences in the degree of intention across stages—as represented by the earlier (precontemplation, contemplation, and preparation) stages—were not significantly different. For example, although the VT scale had a moderately strong association with exercise stage (v 2 5 0.048), the earlier stages did not differ significantly from each other, but were each significantly different from the action and/or maintenance stages. The SF scale did not discriminate among the stages of exercise. DISCUSSION In this study, we examined the association of selfperceived quality of life with a measure of stage of change for adoption of consistent regular exercise, defined as 3 or more times per week for 20 minutes per occasion. Studies of the characteristics of people who do, and do not, exercise regularly can provide important insights to aid the effort to enhance both the efficacy and impact of health promotion initiatives. Classification of the population by stage of motivational readiness to exercise extends this body of research by further differentiating the sedentary and active populations into meaningful stages along an underlying continuum of change. Enhancing self-perceptions and attitudes about the health consequences of exercise behavior is an important strategy proposed for increasing physical activity adoption and adherence in populations [1,19]. It is important, therefore, to understand if stage of change classifications for exercise are influenced by, or influence, self-perceived quality of life and to consider whether this relationship has practical implications for stage-matched intervention messages. The results demonstrate that cognitions about selfperceived quality of life vary significantly across the stages of change for regular exercise. In this large representative sample of adults, the three areas that showed the strongest relationship with stage of exercise were GH, VT, and PF. Linear relationships were found between the stages of change for exercise and the four scales most associated with physical health, as well as for the composite measure of physical health. On the PF scale, for example, precontemplators scored significantly lower than those in the preparation, action, or maintenance stages. Contemplator’s scores were significantly lower than those in maintenance. Precontemplators differentiated from people in other stages on six of the eight dimensions of quality of life. Further, those in the precontemplation stage had the lowest quality of life score on each of the four physical health and functioning scales. This finding of differentiation across stages of exercise is consistent with previous research on other cognitive constructs drawn from a variety of theoretical models. Self-efficacy, decisional balance, processes of change, attitudes, intention, perceived severity, and control beliefs have all been found to discriminate across the stages of change for exercise [22,23,32,34,51]. In one of the few longitudinal studies in this area, Marcus and colleagues found that use of the cognitive process of change labeled “dramatic relief”—which assesses affective aspects of change and is measured by items such as “warnings about the health hazards of inactivity move me emotionally”—increased significantly across the stages of change among persons who adopted exercise and decreased significantly among persons who relapsed from exercise [32]. Similarly, Courneya found that a related construct “perceived severity of the health consequences of inactivity” was directly associated with exercise stage [23]. These studies are compatible with our finding that self-perceived quality of life increases across the stages of exercise. It may be that the perception of positive health benefits of exercise increase with increasing physical activity and may diminish when one relapses from regularly exercising. One potential implication of these results is that modification of an individual’s beliefs in the benefits of regular exercise, or the severity of health consequences due to inactivity, may be useful strategies to help move early stage people to seriously consider becoming more physically active. Focusing intervention messages on altering the perception of the positive benefits of regular exercise (and the negative consequences of a sedentary lifestyle) is also suggested by studies that have shown a generally consistent relationship between stage of exercise and measures of the pros and cons (decisional balance) of exercise [30,36,51,52]. Longitudinal studies and intervention research should continue to explore these relationships to determine whether intention to change exercise behavior can be influenced by changes in self-perception of the positive and negative health consequences of engaging in exercise. Exercise stage was weakly associated with the composite measure of mental health-related quality of life (the MCS summary scale) and demonstrated a significant, but nonlinear, relationship with three of the four mental health scales of the SF-36. Further analysis revealed that engaging in exercise behavior, and not the intention to exercise, was primarily responsible for these associations. Most of the variance found in the exercise stage–mental health status association is explained by the VT scale. Vitality is a complex construct which is correlated moderately with both mental and physical health functioning [25,43]. It measures the degree of energy, pep, or tiredness experienced. Neither intention nor regular exercise behavior above the criterion level were associated with the SF-36 SF scale in this study. The SF scale represents the extent and amount of time that physical or emotional problems have interfered with normal social activities, such as visiting friends. STAGE OF EXERCISE AND QUALITY OF LIFE The literature on the psychosocial health benefits of physical activity and regular exercise is not as consistent as the literature on the physical health benefits of exercise. Numerous studies have found that exercise is associated with feelings of well-being and reduced symptoms of depression, stress, and anxiety [3,53]. Other studies, however, have found little evidence of mental health benefits from exercise [54,55]. Two large cohort studies examined the dose–response relationship between physical activity and symptoms of depression and reported that the benefits of regular physical activity were evident only in the portion of the population at the highest activity levels and not among persons exercising at lower levels [56,57]. In this study, people in the preaction stages (precontemplation, contemplation, and preparation) reported not to be exercising at the specified criterion level (3 or more times a week for 20 minutes), and therefore, may not have approached the threshold of physical activity needed to demonstrate a measurable benefit on the SF-36 MH scales. However, those who reported exercising at or above the criterion—those in the action and maintenance stages—did exhibit measurable increases in three out of four of the mental-health-related quality of life scales. Population studies of the stages of motivational readiness to change exercise behavior can have important implications for the social marketing of health promotion initiatives [58]. It is notable that the distributions and pattern of results on both the exercise stage construct and the SF-36 scores appear to be representative of the general population of adults. The demographic differences found for the PCS and MCS scales in this sample are remarkably consistent with those reported by Ware and colleagues [43]. The distribution of the exercise stage construct is similar to other studies that have used the same or similar staging item [22,30,34,38]. The finding that exercise stage can discriminate between the physical health and mental health components of the SF-36 provides additional evidence of the external validity of the stage measure. This suggests that the relationships found for exercise stage in this study may have wide applicability in monitoring trends in exercise health behavior and intention, as well as for the development of exercise interventions that can reach the majority of the population at risk for the consequences of sedentary lifestyles. There are several limitations to this study that must be considered. The SF-36 has been used extensively with clinical populations and is sensitive to changes in functional health due to clinical conditions. Like the present study, other studies of general population samples have reported significant ceiling effects on some of the SF-36 scales [43]. This is presumably due to the lower frequency of dysfunction in the general population. The PCS and MCS summary component measures 357 are thought to be less affected by ceiling effects than are the SF-36 scales [43]. Ceiling effects can diminish the discriminatory power of a scale. Despite these limitations, the SF-36 was still found to discriminate across the stages of regular exercise on most indices of physical and mental functioning. One consequence of the ceiling effects found in this study is that the effect sizes reported for the stage variable in Table 4 may underestimate the size of the true effect of exercise stage on selfperceived quality of life. These data are cross-sectional and cannot provide evidence of the causal direction of the relationship between exercise stage and quality of life. It is not possible, for example, to determine whether sedentary lifestyle caused the diminished self-perceived quality of life or whether people with lower levels of health-related quality of life did not exercise, or express the intention to exercise, because they have poor health. People who have good physical and mental health-related quality of life may simply be more likely to be physically active. There are, however, many studies that provide strong evidence of a causal relationship between increased levels of exercise and improved health status [1,2,9,10,27]. It will be useful for future longitudinal research to investigate whether the observed relationship between stage of exercise and self-perceived quality of life is due primarily to ill health, to low levels of physical activity, or to other cognitive or lifestyle factors that influence exercise intention and behavior. There are, of course, limitations to our understanding of the validity of the exercise stage construct. In survey research there is always a trade-off between the need for brevity of the questionnaire and the ability to provide detailed data on a topic area. Considerations of cost and respondent burden often give brevity a commanding hand in many large-scale population surveys. There is evidence, however, that the simple stage measure used in this study has construct validity. Approximately one-third of the population was in the precontemplation or contemplation stage, which is consistent with Caspersen and colleagues’ estimate that approximately one-third of the general population is sedentary [59]. This survey did not collect data on exercise behaviors beyond the stage measure, and cannot, therefore, distinguish the degree of exercise intensity, such as vigorous or moderate. It has been shown, however, that progression through the exercise stages is related to increasing amounts of physical activity [23,31– 33,41,42]. Although the stage construct appears to be relatively robust, future research may require comparison of alternative staging methods with objective measures of exercise behavior to determine the most appropriate method for the particular study or intervention approach. An important limitation of this study is that we cannot disentangle the effect of increasing physical activity 358 LAFORGE ET AL. from the possible effect of increasing degrees of intention across the stages. This is an important area to explore in future research. It is likely that the single measure of exercise stage used in this study exhibits an optimistic bias, that is, more people say they are intending to exercise or are in maintenance than really are. It will be useful to know whether the 44% of the population who reports consistent regular exercise for more than 6 months is actually performing at that level. However, even without detailed knowledge of these issues, this paper has found that exercise stage is a brief and reliable construct that can be useful for classifying the population with respect to exercise intention and behavior. CONCLUSIONS This study found that the sedentary portion of the population who report being the least prepared to adopt regular exercise as part of their lifestyle had the lowest self-reported quality of life for physical health. These results imply that health promotion interventions for precontemplators may benefit from targeted messages that call attention to improvements in quality of life that can be expected with increased physical activity. Additionally, the results of this study suggest that it is important to study whether sedentary groups, or those with low levels of physical activity, are experiencing more emotional distress and may be more in need of programs to assist them to adopt regular exercise as a method of managing such distress. There are many personal and social reasons why maintenance of regular exercise is a desirable, yet elusive, objective of health promotion. Existing research has only scratched the surface in the effort to understand the factors involved with regular exercise adoption. This study extends previous research with the finding that self-perception of health-related quality of life is associated with an individual’s stage of motivational readiness to engage in regular physical activity. Research into the causal nature of this relationship could provide information important to the development of more effective stage-matched interventions. The study is consistent with other studies that have demonstrated that meaningful classification of the general population can be made using a simple method of self-report for stage of change of regular exercise. Stage of change for exercise is one construct in a broader set of cognitive and behavioral constructs that have been found to be useful for studying motivational issues related to exercise behavior change [19,23,35,52]. The study of the characteristics of individuals across the distribution of motivational readiness is an important step in the development of social marketing strategies for health promotion which can be applicable to broad segments of the population. ACKNOWLEDGMENTS An earlier version of this paper was presented at the 14th Annual Meeting of the Society for Behavioral Medicine, San Francisco, CA, March 10–13, 1993. The authors thank Claudio Nigg for his thoughtful review and suggestions on an earlier draft of the manuscript. REFERENCES 1. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Physical activity and health: a report of the Surgeon General, Atlanta, 1996. 2. Bouchard C, Shepard RJ, Stephens T, Sutton JR, MCPherson BD (editors). Exercise, fitness and health: a consensus of current knowledge. Champaign, IL: Human Kinetics, 1990. 3. Stephens T. Physical activity and mental health in the United Stages and Canada: evidence from four population surveys. Prev Med 1988;17:35–47. 4. Blair, SN. Physical activity, fitness, and coronary heart disease. In: Bouchard C, Shepard RJ, Stephens T, editors. Physical activity, fitness, and health; international proceedings and conscensus statement. Champaign, IL: Human Kinetics, 1994:579–90. 5. Giovannucci E, Ascherio A, Rimm EB, Colditz GA, Stampfer M, Willett WC. Physical activity, obesity, and risk of colon cancer and adenoma in men. Ann Intern Med 1995;122:327–34. 6. Dalsky GP, Stocke SS, Ehsani AA, Slatopolsky E, Lee WC, Birge SJ. Weight-bearing exercise training and lumbar bone mineral content in postmenopausal women. Ann Intern Med 1988;108: 824–28. 7. Helmrich S, Ragland D, Leung R, et al. Physical activity and reduced occurrence of non-insulin-dependent diabetes mellitus. N Eng J Med 1991;325:147–52. 8. Centers for Disease Control and Prevention. Prevalence of sedentary lifestyle—behavioral risk factor surveillance system, United States, 1991. MMWR Mob Mortal Wkly Rep 1993:42:576–79, 9. U.S. Department of Health and Human Services. Healthy People 2000. National health promotion and disease prevention objectives, Rockville, MD: U.S. DHHS, Public Health Services, 1991. [DHHS Publication No. 91-50212] 10. Pate RR, Pratt M, Blair SN, Haskell WL, Macera CA, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA, 1995;273(5):402–07. 11. Bandura, A. Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, NJ: Prentice–Hall, 1986. 12. Prochaska, JO, DiClemente CC. The stages and process of selfchange in smoking: towards an integrative model of change. J Consult Clin Psychol 1983;51:390–95. 13. Janis, IL, Mann, L. Decision making: a psychological analysis of conflict, choice, and commitment. New York: Free Press, 1977. 14. Ajzen, I, Fishbein, M. Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ, Prentice–Hall, 1980. 15. Cardinal, BJ. The transtheoretical model of behavior change as applied to physical activity and exercise: a review. J Phys Educ Sport Sci 1995:8:32–45. 16. McAuley, E, Courney, KS, Rudolph, DL, Lox, CL. Enhancing exercise adherence in middle-aged males and females. Prev Med 1994;23:498–506. 17. Marcus, BH, Banspatch, SW, Lefebvre, RC, Rossi, JS, Carleton, RA, Abrams, DB. Using the stages of change model to increase the adoption of physical activity among community participants. Am J Health Promot 1992; 6:424–29. 18. Dunn AL, Marcus BH, Kampert JB, Garcia ME, Kohl HW, Blair STAGE OF EXERCISE AND QUALITY OF LIFE SN. Reduction in cardiovascular disease risk factors: 6-month results from Project Active. Prev Med 1997;28:883–92. 19. McCauley, E. Courneya, KS. Adherence to exercise and physical activity as health-promoting behaviors: attitudinal and self-efficacy influences. Appl Prev Psychol 1993;2;65–77. 20. Prochaska, JO, Marcus, BH. The Transtheoretical model: application to exercise. In: Dishman RK, editor. Advances in exercise adherence. Champaign, IL, Human Kinetics, 1994. 21. Marcus, BH, Emmons, KM, Simkin-Silverman, L, Linnan, LA, Taylor, ER, Bock, BC, Roberts, MB, Rossi, JS, Abrams, DB. Evaluation of motivationally-tailored versus standard self-help physical activity interventions at the workplace. Am J Health Promot 1998;12(4):246–53. 22. Courneya, KS. Understanding readiness for regular physical activity in older individuals: an application of the theory of planned behavior. Health Psychol 1995;14(1):1–8. 23. Courneya, KS. Perceived severity of the consequences of physical inactivity across the stages of change in older adults. J Sport Exerc Psychol 1995;17:447–57. 24. Kaplan, RM, Bush, JW. Health-related quality of life measurement for evaluation research and policy analysis. Health Psychol 1982;1:61–80. 25. Ware, JE, Sherbourne, CD. The MOS 36-Item Short-Form Health Survey (SF-36): I. Conceptual framework and item selection. Med Care 1992;30(6):473–83. 26. Rejeski, WJ, Brawley LR, Schumaker, SA. Physical activity and health-related quality of life. Exerc Sport Sci Rev 1996;24: 71–108. 27. Stewart, AL, Hays, RD, Wells, KB, Rogers, WH, et al. Long-term functioning and well being outcomes associated with physical activity and exercise in patients with chronic conditions in the Medical Outcomes Study. J Clin Epidemiol 1994;47(7):719–30. 28. Salmon, CT, Nichols, JS. The next-birthday method of respondent selection. Public Opin Q. 1983;47:207–76. 29. Fowler, FJ, and Mangione, TW. Standardized survey interviewing: minimizing interviewer-related error. Newbury Park: Sage, 1990 30. Reed, GR, Velicer, WF, Prochaska, JO, Rossi, JS, Marcus, BH. What makes a good staging algorithm: examples from regular exercise. Am J Health Promot 1997;12(1):57–66. 31. Wyse, J, Mercer, T, Ashford, B, Buxton, K Gleeson, N. Evidence for the validity and utility of the stages of exercise behavior change scale in young adults. Health Educ Res 1995;10(3): 365–77. 32. Marcus, BH, Simkin, LR. The stages of exercise behavior. J Sports Med Phys Fitness 1993;63:60–6. 33. Barke, CR Nichols, PPR. Physical activity in older adults: the stages of change. J Appl Gerontol 1990;9:216–30. 34. Marcus, BH, Selby, V, Niaura, RS, Rossi, JS. Self-efficacy and the stages of exercise behavior. Res Q Exerc Sport 1992;63:60–6. 35. Armstrong, CA, Sallis, JF, Hovell, FM Hofstetter, CR. Stages of change, self-efficacy, and adoption of vigourous exercise: a prospective analysis. J Sport Exerc Psychol 1993;9:216–23. 36. Myers, RS, Roth, DL. Perceived benefits and barriers to exercise and stage of exercise adoption in young adults Health Psychol 1997;16(3):277–83. 359 37. American College of Sports Medicine. Position statement on the recommended quantity and quality of exercise for developing and maintaining cardioresiratory and muscular fitness in healthy adults. Med Sci Sport Exerc 1990;22:265–74. 38. Rossi JS, Laforge RG, Prochaska JO. Readiness to change 10 health behaviors in smokers, ex-smokers and non-smokers. Proceedings of the 13th Annual Scientific Sessions of the Society of Behavioral Medicine. San Diego, CA: 1995. 39. Laforge, RG, Velicer, WF, Richmond, RL, Owen, N. Stage distributions for five health behaviors in the USA and Australia. Prev Med 1998; in press. 40. Schechtman, KB, Barzlai, B, Rost, K, Fisher, EB. Measuring physical activity with a single question. Am J Pub Health 1991; 81(6):771–73. 41. Cardinal, BJ. Construct validity of stages of change for exercise behavior. Am J Health Promot 1997;12(1):68–74. 42. Emmons, KM, Marcus, BH, Linnan, L, Rossi, JS, Abrams, DB. Mechanisms in multiple risk factor interventions: smoking, physical activity, and dietary fat intake among manufacturing workers. Prev Med 1994;24(4):481–89. 43. Ware, JE, Kosinski, M, Keller, SD. SF-36 physical and mental health summary scales: a user’s manual. Boston, MA: The Health Institute, 1994. 44. Rampal, P, Martin, C, Marquis, P, Ware, JE, Bonfils, S. A quality of life study in five hundred and eighty-one duodenal ulcer patients: maintenance versus intermittent treatment with nozatidine. Scand J Gastroenterol 1994;206 Suppl 29:44–51. 45. Stewart, AL, Greenfield, S, Hays, RD, Wells, KB, Rogers, WH, Berry, SD, McGlynn, EA, Ware, JE. Functional status and wellbeing of patients with chronic conditions: results from the Medical Outcomes Study. JAMA, 1989;267(7):907–13. 46. McHorney, CA, Ware, JE, Raczek, AE. The MOS 36-Item ShortForm Health Status Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs Med Care, 1993;31:247. 47. Ware, JE, Snow, KK, Kosinski, M, Gandek, B. SF-36 health survey manual and interpretation guide. Boston, MA: The Health Institute, New England Medical Center Hospitals, 1993. 48. Cohen, J. Statistical power analysis for the behavioral sciences. Second Ed. Hillsdale, NJ: Lawrence Erlbaum, 1988. 49. State Data Center. 1990 census of population and housing, summary tape file for Rhode Island. Amherst, MA: Massachusetts Institute for Social and Economic Research, 1990. 50. McHorney, CA, Kosinski M, Ware, JE. Comparisons of the costs and quality of norms for the SF-36 Health Survey collected by mail versus telephone interview: results from a national survey. Med Care, 1994;32:551–67. 51. Marcus, BH, Eaton, CA, Rossi, JS, Harlow, LL Self-efficacy, decision-making, and stages of change: an integrative model of physical exercise. J Appl Soc Psychol 1994;24:489–508. 52. Prochaska, JO, Velicer, WF, Rossi, JS, Goldstein, MG, Marcus, BH, Rakowski, W, Fiore, C, Harlow, LL, Redding, CA, Rosenbloom, D, Rossi, SR. Stages of change and decisional balance for 12 problem behaviors. Health Psychol 1994;13(1):39–46. 53. Martinsen, EW, Stephens, T. Exercise and mental health in clinical and free-living populations. In: RK Dishman, editor. Advances in exercise adherence. Champaign, IL: Human Kinetics, 1994; 55–72. 54. Brown DR, Wang Y, Ward A, Ebbeling CB, Fortlage L, Puelo E, et al. Chronic psychological effects of exercise and exercise plus cognitive strategies. Med Exerc Nutr Health 1992;3:125–42. 55. King, AC, Taylor, CB, Haskell, WL. Effects of differing intensities and formats of 12 months of exercise training on psychological outcomes in older adults. Health Psychol 1993;12:292–300. 360 LAFORGE ET AL. 56. Camacho, TC, Roberts, RE, Lazarus, NB, Kaplan, GA, et al. Physical activity and depression: evidence from the Alameda County Study. Am J Epidemiol 1991;134:220–31. 58. Donovan, RJ, Owen, N. Social marketing of population interventions. In: Dishman RK, editor. Advances in exercise adherence. Champaign, IL: Human Kinetics, 1994. 57. Paffenbarger, RS, JR, Lee I-M, Leung, R. Physical activity and personal characteristics associated with depression and suicide in American college men. Acta Psych Scand Suppl 1994;377: 16–22. 59. Caspersen, CJ, Merritt, RK, Stephens, T. International physical activity patterns: a methodological perspective. In: Dishman RK, editor. Advances in exercise adherence. Champaign, IL: Human Kinetics, 1994.