Journal of Rehabilitation Research and Development
Vol. 39 No. 1, January/February 2002
Pages 95 - 103
Personal characteristics that influence exercise behavior of older adults
Lisa W. Boyette, MEd; Adrienne Lloyd, MEd; James E. Boyette, MSICS; Erica Watkins, BA; Lori Furbush, PhD; Sandra B. Dunbar, PhD; L. Jerome Brandon, PhD
Atlanta VA Medical Center, Rehabilitation Research & Development Center, Atlanta, GA; Division of Geriatric Medicine and Gerontology, Emory University School of Medicine, Emory University, Atlanta, GA; Veterans Affairs, Health Eligibility Center, Atlanta, GA; Nell Hodgson Woodruff School of Nursing, Emory University School of Nursing, Emory University, Atlanta, GA; Department of Kinesiology and Health, Georgia State University, Atlanta, GA
Abstract: Long-term exercise participation among older adults will result in healthier lifestyles and reduced need for health care. A better understanding, therefore, of what influences older individuals to start and maintain exercise plans would be beneficial. The twofold purpose of this study was (1) to create a knowledge base of determinants that influence exercise behavior in older adults and (2) to have health professionals prioritize determinants that affect exercise initiation and adherence in older adults. The expert panel examined nine determinants within the category of personal characteristics: age, gender, ethnicity, occupation, educational level, socioeconomic status, biomedical status, smoking status, and past exercise participation. The experts rated the determinants on importance for influencing exercise behavior of older adults. This expert panel concluded that older adults who are in good health and have a history of exercise activity might be more likely to participate in long-term exercise programs.
Key words: aging, determinants, exercise.
This material is based on work supported by the Department of Veterans Affairs, Rehabilitation Research and Development Services, #E825-RA.
Address all correspondence to Lisa W. Boyette, MEd, 1115 Applewood Circle, Signal Mountain, TN 37377; email: Lisazac3212@yahoo.com.
Many diseases and disabling conditions associated with aging can be prevented or delayed with regular exercise. In fact, two out of five adults aged 65 y or older report sedentary lifestyles, thereby increasing their death risk by 5 to 6 percent (1-3). Regular exercise is beneficial in preventing and managing hypertension, obesity, and diabetes mellitus in older females and males (4). These benefits can be attained with moderate exercise, such as a daily walk of 30 min and a 20-min strength training regimen 2 days a week. Continued exercise activity among older adults will result in healthier lifestyles and reduced need for medical care.
Despite advances in recruiting older adults into exercise interventions, recent studies have documented a 6 to 34 percent dropout rate, with the greatest number of dropouts occurring in the first 3 mo (5,6). To reduce dropout, particularly in older persons, one must understand the underlying factors producing it. In 1992, according to the Behavioral Risk Factor Surveillance System (BRFSS), only 23 percent of males and 15 percent of females over the age of 75 reported regular exercise at five times a week for 30 min or more per session (7). In order to create exercise programs to which older adults will continually adhere, health professionals must have a greater understanding of what determinants or exercise factors influence long-term exercise participation in this age group (8).
The major reason for further study of exercise determinants or factors that affect exercise behavior in older adults is to expand our knowledge in developing physical activity interventions for public health planning. A comprehensive determinant review conducted by King et al. (9) identified five major areas that influence exercise behavior in the general population: personal characteristics; knowledge, attitudes and beliefs; psychological/behavioral attributes; activity characteristics; and environmental characteristics. However, it is not known if the same determinants that influence exercise behavior in young and middle-aged adults also pertain to older adults.
Dishman (10) has stated that identification of exercise determinants and their importance for getting and keeping older adults active is necessary. Boyette, Sharon, and Brandon (11) emphasize the need for such studies on selected subgroups because previous determinant research focused primarily on the general population. It is unclear if the determinants in King's review (9) pertain to older people or if there are unique or more specific determinants that may influence exercise patterns in this age group. A better understanding of all determinants and processes involved in initiation of and adherence to exercise programs by older adults is a matter of significant necessity to the geriatric health profession (11,12).
Research studies designed to examine exercise determinants would provide a valuable resource for professionals in the health-related fields (13). Therefore, the purpose of this study was twofold: (1) to create a complete knowledge base of determinants that influence exercise behavior in older adults and (2) to have experts in health professions prioritize the determinants that affect exercise initiation and adherence in older adults.
METHODSDevelopment of Determinants Knowledge Base
A descriptive research design was used in this study. We developed a detailed and comprehensive knowledge base of determinants that affect beginning (initiation) and maintaining (adherence) an exercise plan in older adults. In this study, the initiation phase was defined as the first 6 mo of starting a new exercise program while the adherence phase was the continuance of the exercise routine after the initial 6-mo period. A search of the literature revealed no published studies that generated a knowledge base specific for exercise determinants relating to older adults. We, therefore, used a modification of the determinant framework of King et al. (9) and applied the same five categories as the outline for doing a literature search of studies specific to exercise determinants relating to older adults.
The studies used in the knowledge base consisted of older adults over the age of 60 who were medically stable with normal physical, mental, and social capabilities. The studies were published from 1980 to 1997 and consisted of 100 articles relating to many of the determinants identified by King et al. (9) plus additional determinants (e.g., body image) recognized by other researchers (11). The theoretical framework of King et al. provided a solid basis for selecting studies relating to older adults, with the use of pre-established determinants known for the general population.
Also integrated within the knowledge base were results from exercise program research conducted at the Atlanta Veterans Affairs Rehabilitation Research and Development Center that used the Physical Exercise Profile (PEP) Questionnaire (13). The PEP assessment tool was used to identify which determinants were most important for exercise initiation and adherence in two exercise intervention studies with older adults. The PEP assessed seven major determinants: current exercise level, past exercise level, exercise motivation, perceived health status, physical fitness knowledge, body image, and socioeconomic status. In one PEP data set, there were 76 subjects aged 61 to 87 y (mean age = 73.51 y; SD = 6.44 y). The PEP was used with another data set consisting of 63 subjects aged 61 to 87 y (mean age = 72.06 y; SD = 5.217 y).
The final knowledge base used five major categories with distinct determinants in each of them: (1) personal characteristics (nine determinants); (2) knowledge, attitudes, and beliefs (nine determinants); (3) psychological/behavioral attributes (four determinants); (4) activity characteristics (four determinants); and (5) environmental characteristics (three determinants). This resulted in a total of 29 exercise determinants previously identified as incentives and/or obstacles to exercise initiation and adherence (9,10,13).Validation of Determinants Knowledge Base
We then chose a panel of 18 experts (age range 41 to 70 y; mean age = 49.5 y) who consisted of 1 international expert, 13 national experts, and 4 local experts to participate in this project (names, titles, and affiliations provided in Appendix A, Table). All experts held post-graduate degrees and 50 percent were females. Experts were selected based on several criteria: (1) they had conducted research studies on exercise; (2) they had published extensively in geriatric research, and/or (3) they had clinical experience with the geriatric population.
The experts were professionals who frequently saw or interacted with older adults in their practices and provided them exercise counseling. Expert validation of the knowledge base was essential. Validation was done by reviewing the literature to determine if there was congruence between the knowledge base and their specific experiences with older adults, and then by prioritizing the importance of the determinants. Project staff individually mailed a copy of the knowledge base to each expert. Experts were asked to review it for organizational structure, completeness, and importance of the determinants. The experts provided knowledge in both the theoretical and applied aspects of exercise and aging throughout the project.
Because of the extensive amount of data collected for all five categories of determinants (a total of 29 determinants), we determined that to present these data effectively and comprehensively, each category should be presented separately. The "personal characteristics" category, with nine determinants, is the focus of this discussion. "Personal characteristics" represents those factors that create the unique demographic makeup of an individual that can sway his or her decisions about exercise behavior and are believed to be powerful determinants of exercise initiation and adherence. "Personal characteristics" includes age, gender, ethnicity, occupation, educational level, socioeconomic status, biomedical status, smoking status, and past exercise participation. We define biomedical status in this study as the rating of a patient's health, provided by a physician.
Socioeconomic status (SES) is defined as the combination of educational level, occupation, and income level. This term is a more global term, but the published studies have also examined the components of SES. We therefore, took an extra step in having the experts not only review SES as a global term but also review educational level, occupation, and income level as separate components to assess their impact on exercise behavior.Ratings and Analyses of Determinants Within "Personal Characteristics"
Experts independently rated each of the nine determinants in the "personal characteristics" category separately, for both initiation and adherence, according to its importance for exercise in older adults. The experts used a scale of 1 to 10, with 10 being most important (Appendix B, Example). Any determinant could get the same rating as any other determinant on this importance scale. Experts provided their rating information through the mail.
Data analyses included basic descriptive statistics on the experts' ratings of the nine determinant variables. The determinant ratings were evaluated for means, standard deviations, and lower and upper 95 percent confidence intervals. If there was no visual overlap of confidence intervals when the nine determinants within the "personal characteristics" category were examined, the determinants were considered different (14).
Means and confidence intervals were presented in graphic form for both the initiation and adherence phases (see Figures 1 and 2 ), enabling the reader to visually inspect for overlap of confidence intervals (14). Confidence intervals were reported in this article as the upper and lower limits around the mean.
The data should only be used descriptively and do not imply statistical significance, given the error rate involved with multiple pair-wise comparisons. We considered determinants with the higher means the more important factors, based on the opinions of the experts.
RESULTSDeterminants Knowledge Base
After receiving feedback on the knowledge base from the experts, the project staff made appropriate changes to reflect these suggestions or comments. Only minor editing was suggested and some categories were expanded to include more references, especially in the area of self-efficacy. The experts neither added more determinants than those already listed nor wanted to exclude any determinants. Their valuable input regarding expansion of some determinant information and their editorial comments were used to complete the final knowledge base. The experts reviewed and approved the final knowledge base with no further suggestions, edits, or changes.Ratings and Analyses of "Personal Characteristics" Determinants
All 18 experts completed the importance rating of the nine determinants within the category of "personal characteristics." For the exercise initiation phase (Figure 1), determinants for biomedical status, past exercise participation, and educational level were rated notably higher than for occupation, ethnicity, and gender, as indicated by nonoverlap of confidence intervals. Specific data for each initiation phase determinant are shown in the Appendix B, Table, along with ranking order of determinants. The biomedical status mean was 8.3, with 95 percent confidence intervals (CI95%) of 7.4 to 9.2. Past exercise participation received the second highest rating, (mean = 7.8, CI95% = 6.8 to 8.8), while educational level was rated as the third highest initiation determinant (mean = 7.5, CI95% = 6.7 to 8.3). SES was rated as the fourth highest determinant (mean = 7.2, CI95% = 6.2 to 8.2) and was rated higher than occupation and gender. The other determinants received mean scores from 6.0 to 4.7, with gender receiving the lowest rating.
For the exercise adherence phase, biomedical status and past exercise participation were rated higher than smoking status, ethnicity, age, occupation, and gender (Figure 2), as indicated by nonoverlap of confidence intervals. SES was rated higher than ethnicity, age, occupation, and gender. Specific data for each adherence phase determinant are shown in the Appendix B, Table , as is the determinant ranking order. The adherence data indicated that biomedical status continued to be the most important determinant, having the highest mean rating of 8.4, CI95% = 7.4 to 9.4 (Appendix B, Table). Past exercise participation received the second highest rating with a mean of 7.9, CI95% = 7.0 to 8.8. The SES determinant was rated as third highest, receiving a mean value of 7.4, CI95% = 6.4 to 8.4. The other adherence factors received mean ratings of 6.6 or less, with gender again rated as least important for this age group.
This study provided support for prioritizing which determinants within the "personal characteristics" category are most important when designing an exercise plan for older adults. During the initiation and adherence phases of exercise, our experts rated biomedical status as the most important determinant. They strongly agreed that the state of health of older adults has an impact on how likely they are to initiate and adhere to exercise, which is consistent with other findings in the literature (15,16). It has been consistently reported that healthy adults are more active than persons with medical complications (15,17). A thorough evaluation of the illnesses, injuries, health conditions, and symptoms of an older person should be documented.
Monitoring the biomedical status of an older adult is important when working with this age group because status can quickly change. If the older adult continues to have few illnesses and injuries, she or he is more likely to stay in the exercise program.
In a sample of over 6,000 older adults, aged = 70 y, Wolinsky, Stump, and Clark found that persons with orthopedic limitations are less likely to exercise (15). According to the study, heavier body mass also negatively affects psychosocial behaviors such as motivation and body image. Other studies also report that overweight individuals are less likely to participate in exercise than normal weight individuals (9,16). According to Dishman, overweight individuals are less likely to stay with a vigorous exercise plan and might respond better to moderate activities, such as walking (18). In a group of women of ages 55 to 96 y, biomedical conditions that affect exercise participation were examined (19). The only significant chronic biomedical condition that differed between the exercisers and nonexercisers was a previous diagnosis of osteoporosis.
Past exercise participation was rated as the second highest determinant for exercise initiation and adherence. Other studies have found similar results when examining past exercise participation. In the study by Boyette, Sharon, and Brandon, past exercise level was a significant determinant of exercise adherence in a sample of 76 older adults (11). Older adults who survived the 50 percent attrition rate during the first 6 mo of exercise were likely to continue to exercise 1 y later (20). According to McAuley, Lox, and Duncan, once a regular routine of exercise was established, this habitual activity became a major predictor of future exercise maintenance (21).
Learning the history of past exercise participation of the older adult, including past leisure activity, structured exercise activity, and advanced activities of daily living, will help identify strategies to optimize future exercise behavior. Exercise history over the past year, and especially within the past month, should be assessed. It is advisable to determine what previous exercise activities the client liked or disliked, as well as to examine his or her previous exercise incentives and barriers.
The experts rated educational level as the third most important determinant during the initiation phase. The positive association between exercise participation and level of education has been established in a number of other studies (17,22-23). Clark conducted a study that examined educational differences in physical activity levels of adults 70 and older while controlling age, sex, race, and income (24). The participants were asked if they "get as much exercise as needed" and if they "have a regular exercise routine." Those individuals with 8 y or less of education were found to be less physically active than those with 9 y or more of education. In another study conducted by Booth et al., more exercise generally decreased with increasing age, but increased with the education level when age was controlled (25).
An educational component should be part of the exercise plan, with explanations provided for how exercise may impact health conditions and symptoms. The health practitioner should also explain major concepts of exercise to the older participant, such as target heart rate, perceived exertion, and body composition. It would be worthwhile to use written handouts with exercises and pictures that are appropriate for the educational level of the individual. The health practitioner could heighten exercise awareness using methods tailored to the preferred learning mode(s) of the individual (i.e., group discussions, fun and educational games, interview, and demonstration and practice with the machines and equipment).
SES was rated by our experts as being the third most important determinant during the adherence phase, instead of education. In this study, SES was defined as the combination of an individual's income, education, and occupation. Few studies have examined the relationship between different levels of SES and exercise. Ford et al. reported that women of lower SES spent significantly less time in physical activity than women of higher SES (26). In contrast, the amount of time spent per week in physical activity reported by men of upper and lower SES was nearly the same. Both Ford et al. and Schoenborn have also found that men and women with higher SES spent more time engaged in leisure-time physical activity than did lower SES men and women (26,27). King maintains that the success of some community exercise intervention programs are of substantial interest to lower socioeconomic groups and need not be limited to the middle and upper class (28). For example, the Community Health Assessment and Promotion Project (CHAPP) targeted inner-city residents of Atlanta and found that this community segment was quite interested in exercise (29). Therefore, to promote exercise within this group, security escorts were furnished for groups walking in dangerous neighborhoods, curtains were added to the exercise classrooms to increase privacy, and methods for arranging transportation were provided.
SES is also important during the adherence phase, because the income level of an older adult might be changing as he or she enters or remains in a retirement life phase. Therefore, when the health professional is updating the exercise plan during the adherence phase, he or she needs to know how much money, if any, the client is willing to pay for exercise. Perhaps, after 6 mo of initiating an exercise plan, even though an individual knows that she or he needs to continue exercising, there are now questions about the expense and priority of the program, in light of other commitments. For example, if the older adult can continually afford the exercise plan and sets aside time to exercise after a working day, there will be a positive correlation for continued exercise maintenance.
Based on this sampling of expert opinions, age, gender, ethnicity, occupation, and smoking status were not rated as important as other factors for habitual exercise in older adults. However, these demographic factors are necessary pieces of information for obtaining an accurate profile of the older individual. Obviously, the age of the client is important to consider because older individuals may have a greater appreciation of exercise activities that provide health-related fitness benefits as opposed to performance-related benefits. Older individuals with different ethnic backgrounds have different life experiences and, therefore, might feel more comfortable with certain exercise activities than with others. The work status of the older person is relevant because he or she may still work part-time or have a volunteer position, making their work location and/or time commitments pertinent information. If the older adult smokes, the health practitioner may prescribe a lower intensity level during the initial stages of exercise. While smoking is a difficult habit to change, the older adult might be able to overcome it with the help from the health professional.
This study prioritized determinants within the "personal characteristics" category that are most important when designing an exercise plan for older adults. Expert health professionals identified biomedical status, past exercise participation, and education, in order of decreasing priority, as most important during the initiation phase. During the adherence phase, the most important determinants were prioritized as biomedical status, then past exercise participation, and finally, SES. These findings suggest that healthier older adults who have a history of exercise are more likely to start and maintain an exercise plan. Additionally, educating older adults on the benefits of exercise increases the likelihood of them initiating and adhering to an exercise prescription. Older adults, in conjunction with their healthcare professional, should identify strategies that can remove barriers to exercise and create incentives to optimize the habit of regular exercise in their lifestyle.
Panel of Experts
The following table is the panel of 18 experts (age range 41 to 70 y; mean age = 49.5 y) who consisted of 1 international expert, 13 national experts, and 4 local experts to participate in this project. The experts were professionals who frequently saw or interacted with older adults in their practices and provided them with exercise counseling. They provided expert validation of the determinants knowledge base that was essential. These experts provided knowledge in both the theoretical and applied aspects of exercise and aging throughout the project.
Type of Expertise
Name, Professional Degree(s)
International experts (n = 1)
Sarah Elizabeth Lamb, MSc, MCSP, SRP
Harkness Fellow in Public Policy, The Commonwealth Fund of New York.
(n = 13)
1. Judy Beamer, BS
Director, Cecile Cox Quillen Exercise Research Program, East Tennessee State University.
2. Steven N. Blair, PED
Director, Epidemiology and Clinical Applications; Cooper Institute for Aerobics Research in Dallas, TX.
3. Carl J. Caspersen, PhD, MPH
Physical Activity Epidemiologist, Physical Activity and Health Branch, Division of Nutrition and Physical Activity, National Center for Chronic Disease, Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA.
4. Robert P. Cunningham, MD
Corporate Medical Director, retired; Bell South, Atlanta, GA.
5. Barbara de Lateur, MD
Professor, Director and Lawrence Cardinal Shehan Chair, Department of Physical Medicine and Rehabilitation, The Johns Hopkins Hospital, School of Medicine. Joint Professor of Health Policy and Management, School of Hygiene and Public Health, Baltimore, MD.
6. Barbara J. Fletcher, RN, MN, FAAN
Clinical Associate Professor, University of North Florida, College of Health, Department of Nursing.
7. Ronald C. Hamdy, MD, FACP, FRCP
Associate Chief of Staff, Extended Care & Geriatrics, Veterans Affairs Medical Center, Mountain Home, Tennessee. Holder of the Cecile Cox Quillen Chair of Excellence in Geriatric Medicine & Gerontology at James H. Quillen College of Medicine, East Tennessee State University, Johnson City, Tennessee. Director of the East Tennessee State University Osteoporosis Center.
8. Priscilla G. MacRae, PhD
Professor of Sports Medicine, Department of Sports Medicine and Physical Education, Pepperdine University, CA.
9. Edward McAuley, PhD
Professor of Exercise Psychology, Department of Kinesiology, University of Illinois at Urbana-Champaign.
10. Miriam C. Morey, PhD
Director, GEROFIT, Geriatric Research, Education and Clinical Center, VA Medical Center, Assistant Research Professor, Department of Medicine, Center on Aging and Human Development, Duke Medical Center, Durham, NC
11. Scott Sherman, MD, MPH
Assistant Professor Medicine, UCLA/San Fernando Valley Program. Chief, PACE Research, Evaluation, and Faculty Development, Sepulveda VA Medical Center.
12. Frank Whittington, PhD
Professor of Sociology and Director of the Gerontology Center at Georgia State University, Atlanta, GA.
13. Jeffrey C. Rupp, PhD
Associate Professor and Chairman, Department of Kinesiology & Health, Georgia State University, Atlanta, GA.
(n = 4)
1. Carol E. Coogler, ScD
Assistant Professor, Emory University School of Medicine & Center for Rehabilitation Medicine, Emory University, Atlanta, GA.
2. Sandra B. Dunbar, RN, DSN
Professor, Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta, GA.
3. Dale Strasser, MD
Chief of Rehabilitation Medicine, Wesley Woods Geriatric Hospital, & Interim Chair, Department of Rehabilitation Medicine, Emory School of Medicine, Atlanta, GA.
4. Robert A. Zorowitz, MD, FACP
Medical Director of Geriatrics Services, DeKalb Regional Healthcare System, Decatur, GA.
Sample Initiation Importance Rating Sheet
Please rate each of the determinants independently, on a scale of 1 to 10, based on how important you think each determinant is in getting healthy, older adults (aged 65 to 85 y) to initiate exercise programs. Importance of determinants will vary between individuals; however, for the purpose of this rating, consider the determinants in a general sense for all healthy older adults.
A rating of 1 means that you think the determinant is not important at all for initiation, while a rating of 10 means that you think the determinant is most important for initiation. Independently rating each determinant means that you assign a score of 1 to 10 to each determinant, so determinants that you think are of equal importance to exercise initiation are given the same score. If you added some determinants to the knowledge base, please use the blank spaces in the right column to fill in the additional determinants and their associated ratings.Personal Characteristics
___ Socioeconomic status
___ Biomedical status
___ Smoking status
___ Past exercise participation
The initiation phase is the first 6 mo of starting a new exercise program, and the adherence phase is the continuance of the exercise routine after the initial 6-mo period.Back to text
The determinant ratings were evaluated for lower and upper 95 percent confidence level intervals. Confidence intervals were reported as the upper and lower limits around the mean. If there was no visual overlap of confidence intervals when the nine determinants within the "personal characteristics" category were examined, then the determinants were considered to be rated differently from each other.
Ratings of determinants by experts.
Back to text
Mean + 1 SD
Mean + 1 SD
5.9 + 2.1
2 - 10
4.9 + 2.1
2 - 10
8.3 + 1.8
4 - 10
8.4 + 2.0
4 - 10
7.5 + 1.7
3 - 10
6.6 + 1.9
3 - 10
5.1 + 2.6
1 - 10
5.1 + 2.5
2 - 10
4.7 + 1.8
2 - 8
4.1 + 2.4
1 - 9
5.1 + 2.1
2 - 8
4.7 + 2.2
2 - 8
7.8 + 2.1
4 - 10
7.9 + 1.8
4 - 10
6.0 + 2.0
2 - 9
5.9 + 1.7
2 - 8
7.2 + 2.1
3 - 10
7.4 + 2.0
3 - 10
SD = one standard deviation
- McGinnis JM. The public health burden of a sedentary lifestyle. Med Sci Sports Exerc 1992;24:S196-S200.
- Jette AM, Lachman M, Giorgetti MM, Assman SF, Harris BA, Levenson C, et al. Exercise-it's never too late: The strong-for-life program. Am J Public Health 1999;89:66-72.
- Ettinger WH, Burns R, Messier SP, Applegate W, Rejeski WJ, Morgan T, et al. A randomized trial comparing aerobic exercise and resistance exercise with a health education program in older adults with knee osteoarthritis. The Fitness Arthritis and Seniors Trial (FAST). J Am Med Assoc 1997;277:25-31.
- Burnham JM. Exercise is medicine: Health benefits of regular physical activity. J La State Med Soc 1998;150(7):319-23.
- Powell KE, Blair SN. The public health burdens of sedentary living habits: Theoretical but realistic estimates. Med Sci Sports Exerc 1993;26:851-6.
- Blair SN, Kohl HW, Barlow CE, Paffenbarger RS, Gibbons LW, Macera CA. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. J Am Med Assoc 1995;273:1093-8.
- U.S. Department of Health and Human Services. Physical activity and health: A report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion 1996.
- Allison MJ, Keller C. Physical activity maintenance in elders with cardiac problems. Ger Nurs 2000;21(4):200-3.
- King AC, Blair SN, Bild DE, Dishman RK, Dubbert PM, Marcus, BH, Oldridge NB, Paffenbarger RS, Powell KE, Yeager KK. Determinants of physical activity and interventions in adults. Med Sci Sports Exerc 1992;24(6):S221-S236.
- Dishman RK. Exercise adherence research: Future directions. Am J Health Promot 1988;3(1):52-6.
- Boyette LW, Sharon BF, Brandon LJ. Exercise adherence for a strength training program in older adults. J Nutr Health Aging 1997;1(2):93-7.
- Schmidt JA, Gruman C, King MB, Wolfson LI. Attrition in an exercise intervention: A comparison of early and later dropouts. J Am Ger Soc 2000;48(8):952-60.
- Boyette LW, Cannella KC, Archea C, Sharon B, Del' Aune W. Reliability and validity of the physical exercise profile. Med Sci Res 1995;23(7):499-501.
- Velleman PF, Hoaglin DC. Applications, basics and computing of explor data analysis. Belmont, CA: Duxbury Press; 1981.
- Wolinsky, FD, Stump TE, Clark DO. Antecedents and consequences of physical activity and exercise among older adults. The Gerontologist 1995;35(4):451-62.
- U.S. Department of Health and Human Services. Centers for Disease Control and Prevention (CDC). MMWR Morbid Mortal Wkly Rep 1999;48: No. SS-8.
- Rhodes RE, Martin AD, Taunton JE, Rhodes EC, Donnelly M, Elliot J. Factors associated with exercise adherence among older adults. An individual perspective. Sports Med 1999; 28(6):397-411.
- Dishman RK, Increasing and maintaining physical activity and exercise. Beh Therapy 1991;41:3-15.
- Richter DL, Macera CA, Williams H, Koerber M. Disincentives to participation in planned exercise activities among older adults. Health Values 1993;17(5):51-5.
- Dishman RK, Sallis, JF, Orenstein DR. The determinants of physical activity and exercise. Public Health Rep 1985;100:158-71.
- McAuley E, Lox C, Duncan TE. Long-term maintenance of exercise, self-efficacy, and physiological change in older adults. J Gerontol B Psychol Sci Soc Sci 1993;48(4):218-24.
- Folsom AR, Caspersen CJ, Taylor HL. Leisure-time physical activity and its relationship to coronary risk factors in a population-based survey. Am J Epidemiol 1985;121:570-9.
- Stephens T, Jacobs, DR, White CC. A descriptive of leisure-time physical activity. Public Health Rep 1985;100(2):147-58.
- Clark DO. Racial and educational differences in physical activity among older adults. The Gerontologist 1995;35(4):472-80.
- Booth ML, MacAskill P, Owen N, Oldenburg B, Marcus BH, Bauman A. Population prevalence and correlates of stages of change in physical activity. Health Ed Q 1993;20(3):431-40.
- Ford ES, Merritt RK, Heath GW, Powell KE, Washburn RA, Kriska A, Haile G. Physical activity behaviors in lower and higher socioeconomic status populations. Am J Epidemiol 1991;133:1246-56.
- Schoenborn CA. Health habits of U.S. adults: The "Alamaeda 7" revisited. Public Health Rep 1986;101:571-80.
- King AC, Haskell WL, Taylor CB, Kraemer HC, Debusk RF. Group- vs. home-based exercise training in healthy older men and women: A community-based clinical trial. J Am Med Assoc 1991;266:1535-42.
- Lasco RA, Curry RH, Dickson VJ, Powers J, Menes S, Merritt RK. Participation rates, weight loss, and blood pressure changes among obese women in a nutrition-exercise program. Public Health Rep 1989;104:640-6.
Submitted for publication December 12, 2000. Accepted in revised form March 6, 2001.
Go to TOP.