Health-promoting behaviors in men age 45 and above.

The purpose of this research was to examine and explain health-promoting behaviors among middle-aged and older men. The Health Promotion Model (HPM) was used as the framework for this study. Surveys were used to measure perceived benefits, perceived barriers, self-efficacy, demographic factors, self-rated health (independent variables), and health-promoting behaviors (dependent variables). A multiple linear regression model revealed 66% of the variation in total health-promoting behaviors (HPB) accounted for by the independent variables. Models for exercise HPB and nutrition HPB showed 65% and 53% of the variance explained by the independent variables. Partial correlations determined that self-efficacy was the single most important predictor in all three models. Interventions to enhance men's self-efficacy for health-promoting behaviors should be developed and studies for impact on behavior and health carried out.

Keywords: men's health-promoting behaviors, self-efficacy, health-promotion model

**********

In the United States, there are significant differences between men's and women's health. Men suffer higher rates of acute and chronic conditions such as heart disease, cancer, and AIDS; are involved in more car crashes and occupation-related accidents; and are more likely than women to be overweight, drink heavily, and smoke cigarettes. Premature mortality (deaths before age 65) is 50% higher for men; men age 45-64 are three times more likely than women to suffer fatal coronary events.

Tendencies for higher male mortality persist or increase with aging, as men age 65-84 are twice as likely as women to have fatal coronary events. For cancer, the gender disparity in death rates widens as men age. While cancer death rates for men age 45-54 are only slightly higher than rates for women, at age 55-64, male cancer death rates are 25% higher; at 65-74, 50% higher, and at 75-84, 70% higher. Men have higher death rates in all of the 10 leading causes of death, resulting in a 5.5-year-lower life expectancy for men (National Center for Health Statistics, 2004). These differences constitute health disparities, and by definition render men as a vulnerable population in terms of health. According to de Chesney (2005), vulnerable populations are those with a greater-than-average risk of developing health problems by virtue of some factor, in this case male gender.

Why do men die sooner than women? The answers are likely related to the combination and interaction of behavioral, biological, social, and environmental factors (Jones, 2001; USDHHS, 2000). Possible explanations for gender differences in mortality include protective factors within female sex chromosomes, more environmental hazards for men, men's reluctance to seek healthcare, and poorer health behaviors among men (Loeb, 2004). Courtenay (2000) proposed that most gender differences in mortality can be explained by men's behaviors, although interaction between genetics and behavioral factors may account for most of these differences (Hazzard, 2000).

Little attention has been paid to health-promoting practices as men make the transition from the middle to the later years of their life. Middle age represents a time when individuals start to acknowledge their own aging, may face health problems for the first time, and begin to feel vulnerable. This developmental period can pose a crisis, with the potential for opportunity and threat (Levinson, 1986). Healthy behaviors in middle age, carried through to later years, have a profound impact on psychological, social, and physical well-being, potentially extending and improving quality of life (Fries, 1999).

Men's health literature documents differences in health behaviors, morbidity, and mortality but is largely devoid of a theoretical explanation (Denton & Waiters, 1999; Reed, Foley, White, Heimovitz, Burchfiel, & Masaki, 1998). The Health Promotion Model (HPM), originally developed in the 1980s by Nola Pender, served as the theoretical framework for this study of men's health-promoting behavior. According to the HPM, personal characteristics such as age, gender, or education interact with behavior-specific cognitions like self-efficacy, perceived benefits, or barriers to the behavior to influence an individual's commitment to engage in health-promoting behaviors. Balanced against this commitment are other immediate competing demands or preferences that may sidetrack an individual from carrying out the intended behavior (Pender, Murdaugh, & Parsons, 2002). Authors of the model contend that positive motivators, such as feeling better or looking better, may offer better explanation for long term health-promoting behaviors than avoidance-oriented motivators.

A number of studies based on the HPM have found higher ratings of self-efficacy to be associated with greater participation in health-promoting behaviors (Desmond, Conrad, Montgomery, & Simon, 1993; Duffy, Rossow, & Hernandez 1996; Misener, Phillips, & McGraw, 2000; Tapler, 1996). Higher perceptions of benefits have generally been associated with greater reported participation in health-promoting behaviors (Lusk, Ronis, Kerr, & Atwood, 1994; Jones & Nics, 1996); while more perceived barriers have been associated with lower levels of participation (Bungum, Orsak, & Chng, 1997; Jones & Nics, 1996; Stuifbergen, Seraphine, & Roberts, 2000; Tapler, 1996). Some studies have shown older individuals reporting more health-promoting behaviors (Bagwell & Bush, 2000; Loeb, 2003; Weitzel 1989), while others reveal no age-related differences (Desmond, Conrad, Montgomery, & Simon, 1993; Misener, Phillips, & McGraw, 2000). The purpose of this study was to use the Health Promotion Model as a framework to examine health-promoting behaviors among middle-aged and older men (ages 45-90).

METHODS

DESIGN

A cross-sectional research design was employed in this study using survey research for efficiency, specificity, generalizability, and potential for verification through replication (Babbie, 1997; Rea & Parker, 1997). Age 45 was selected as the youngest cut-point in the study because 45-64 is the age category recognized by the U.S. Census Bureau (2001) as middle age. Dependent variables were overall and specific health-promoting behaviors (healthy eating, exercise, stress management, interpersonal relations, health responsibility, and spiritual growth). Independent variables were delimited to constructs demonstrating explanatory power in other published reports and included demographics, self-rated health, self-efficacy, perceived benefits, and perceived barriers to health-promoting behaviors.

INSTRUMENTS

The 52-item Health-Promoting Lifestyles Profile II (HPLP) was used to measure health-promoting behaviors. The recently adapted HPLPII has produced subscale alpha coefficients ranging from .793 to .861 and overall .943 (personal communication, Walker, 2002). Self-efficacy was measured with the Self-Rated Abilities for Health Practices (SRAHP), a scale specific to health-promoting behaviors (Becker, Stuifbergen, Oh, & Hall, 1993). This 28-item instrument also has subscales for nutrition, physical activity, health responsibility, and stress management. Totaling items produces subscale and overall self-efficacy scores. This scale has been shown to be reliable with alpha coefficients of .94 for the entire scale and .81-.92 for the various subscales. Murdaugh and Hinshaw's (1986) 20-item BES scale was used to measure benefits. This scale focuses primarily on benefits of healthy eating and exercise. Previous studies have yielded Cronbach's alpha of .80 and a two-week, test-retest coefficient of .64 (Murdaugh & Hinshaw, 1986). The Barriers to Health Promoting Activities for Disabled Persons (Becker, Stuifbergen, & Sands, 1991) scale was developed from other instruments and from interviews with disabled college students. The internal consistency coefficient was .82 (Becker, Stuifbergen, & Sands, 1991). Although this scale was developed for people with disabilities, a review of the questions suggested the items were appropriate for nondisabled individuals.

Additional questions for age, income, education, marital status, race, and self-rated health were added to the instrument. The overall reading level for the instrument was estimated to be at grades 8-10 (Doak, Doak, & Root, 1995).

PILOT TESTING

A pilot test with 83 male firefighters and municipal workers was conducted to determine the internal consistency of the various scales. Cronbach's alpha coefficients for each of the scales and subscales were similar to those reported by other researchers (.77-.95 for the various subscales). Respondents in the pilot test represented a convenience sample and were not included in the final analysis.

POPULATION AND SAMPLING STRATEGY

The minimum number of respondents needed was determined, a priori, to be 120, although additional responses would provide more statistical stability (Nunnally & Berstein, 1994). The sample was drawn from male employees and retirees from a Midwestern metropolitan university and from the membership of an American Legion post. These subject pools were selected because of the potentially wide range of education, income levels, and likelihood of recruiting men 45 and older. Permission was obtained from the university's institutional review board to sample male faculty, staff, and retirees for the study. Permission to sample Legionnaires was granted from the post board of directors. A random number was used to begin systematic (nth) selection from the university directory and the Legion member list. If any selected name was determined to be a woman's name, then the next man's name on the list was selected, in all, 641 men were selected to receive surveys. Surveys were distributed through campus mail to 267 male employees and were mailed to 74 male retirees and 300 male Legion members.

RESPONSE RATE

Measures taken to optimize the response rate included a pre-survey announcement, a cover letter explaining the purpose, the voluntary and anonymous nature of the study, and a response card to inform the researcher of those who had responded without violating anonymity. Another copy of the survey was sent to men who did not return response cards within two weeks of the initial mailing. The response cards also served as an entry for a cash drawing and a mechanism for participants to request a synopsis of the survey results as additional incentives (Babbie, 1997; Torabi, 1991).

The overall response rate was 39.3%, based on the number of surveys presumably reaching respondents. Estimated response rates for the three groups were similar (36-43%). Rates were based on returned response cards, although more surveys than response cards were returned. Twenty-seven surveys were returned as undeliverable or were returned by a household member, indicating the intended respondent was deceased or ill. The 34 respondents younger than the cutoff point for the research (age 45) were excluded from analysis as were 17 surveys containing less than 75% completed items for any subscale. Mean values (for individual items) were inserted in surveys having some missing items but at least 75% complete for each subscale (Babbie, 1997; Nunnally & Bernstein, 1994). A total of 191 surveys were suitable for analysis.

DATA ANALYSIS

Calculations and statistical analyses were performed using the statistical software packages SPSS for Windows Version 11.0 and Statistical Analysis System for Unix (2000). The accuracy of data entry was verified by reexamining a random sample of 10% of the surveys. Data entered were found to be 99.7% accurate. Surveys from Legionnaires, university employees, and retirees were pooled for analysis rather than sorted by category of respondent. Differences between groups were not of interest in this study.

Multiple regression analysis revealed variance in health-promoting behaviors explained by the independent variables. The relationship between personal factors (age, income, education, self-rated health) and behavior-specific cognitions (self-efficacy, perceived benefits, perceived barriers) was described by simple correlation coefficients as were the relationships between personal factors and health-promoting behaviors. Partial correlations were calculated to determine which personal factors and behavior-specific cognitions best predict health-promoting behaviors.

RESULTS

DESCRIPTION OF RESPONDENTS

The mean age of participants was 66.1 years of age (Range 45-90, SD = 11.2). More than half of the respondents (53%) reported holding at least a bachelor's degree. Respondents reported 8-22 years of education (mean = 15.9 years, SD = 3.6). The median reported household income was $50,000-$59,999. A comparison of education and income level of the respondents to adults in the state where the study was conducted revealed that the respondents tended to be better educated and earn more than the overall population. The sample overrepresented men age 65-84 and underrepresented men age 45-54 (U.S. Department of Commerce, 2001). A summary of demographic data is displayed in Table 1.

HEALTH-PROMOTING BEHAVIORS AND CORRELATES

HPLP items are scored on a four-point, Likert-type scale with higher numbers representing a greater degree of health-promoting behavior. Subscales scores are arithmetic means of items on the subscale. Subscale scores were highest in spiritual growth and lowest in physical activity. The mean HPLP II score of all 52 items was 2.71 (SD = .43). Scores for the HPLP II and subscales are displayed in Table 2. Relationships between all subscale scores were significant with the strongest relationship between spiritual growth and interpersonal relations (r = .761, p < .01).

The mean self-efficacy score was 101.62, (SD = 18.38). The highest subscale score was for health responsibility, the lowest for physical activity. The mean benefits score was 87.96 (SD = 14.08). Ten of the 20 items relating specifically to the benefits of exercise were totaled to create an exercise benefits subscale. Another six items relating specifically to nutrition were totaled to create a nutrition subscale. Subscales scores cannot be compared to each other since the number of items is unequal and the scores are obtained by totaling item responses. On the barriers scale, higher scores represent greater barriers to health-promoting behaviors. The range of scores was 18-72 (M = 27.70, S.D = 8.51). All of these results are displayed in Table 2.

Multiple regression equations were calculated to examine variance in health-promoting behaviors explained by components of the HPM. Results revealed 66% of the variance in overall health-promoting behaviors was explained by age, education, income, self-reported health, total self-efficacy, benefits, and barriers to health-promoting behaviors ([R.sup.2] = .663, df =12, p < .001).

Separate multiple regression equations were created with physical activity and nutrition as the dependent variables. For physical activity, 65% of the variance was accounted for by age, education, income, self-reported health, exercise self-efficacy, benefits of exercise, and barriers ([R.sup.2] = .650, df = 12, p < .001). For nutrition, 52% of the variance was due to age, education, income, self-rated health, self-efficacy for healthy eating, barriers, and benefits of healthy nutrition ([R.sup.2] = .524, df = 12, p < .001). The multiple regression data are displayed in Table 3.

Originally race and marital status data were collected from respondents. However, in the final analysis, these variables were excluded because most respondents were married (86.4%) and white (91.6%). Correlation coefficients were calculated to examine relationships between the personal factors. Older respondents tended to have less education (r = -167, p < .05) and lower levels of income (r = -.415, p < .01). Men with more education reported higher income (r = .618, p < .01) and being in better health (r = .218,p < .01).

Relationships between personal factors and other predictor constructs (self-efficacy, self-rated health, barriers, and benefits) were examined. Self-reported health was significantly correlated with all aspects of self-efficacy, benefits, and barriers. Those with better self-rated health reported higher overall self-efficacy and self-efficacy for specific dimensions of health-promoting behaviors. Individuals with better self-rated health also reported fewer barriers to health-promoting behaviors (r = -.319, p < .01) and perceived more benefits to health-promoting behaviors (r = .326, p < .01). These data are displayed in Table 4.

Correlation coefficients were calculated to explore the relationship between personal factors and health-promoting behaviors. Advancing age was associated with better scores of stress management (r = .229, p < .01) and health responsibility (r = .204, p < .01). Associations with age and total health-promoting behavior score were not significant. There was a significant positive correlation between education and total health-promoting behaviors (r = .245, p < .01) as well as income and HPB (r =. 164, p < .05). Relationships between income, nutrition, stress management, health responsibility, and interpersonal relationship scores were not significant. Relationships were also noted between self-rated health and HPB (r = .362, p < .01).

In an attempt to determine which independent variables were the best predictors of health-promoting behaviors, Type III sums of squares were calculated. Partial correlation for total health-promoting behavior revealed that total self-efficacy and age were the only independent variables offering significant unique contributions to the variance in total health-promoting behaviors. Self-efficacy accounted for the greatest unique shared variance with total health-promoting behaviors (F [ 1, 176] = 216.53, p < .0001) accompanied by age as a significant predictor (F [1, 176] = 7.43, p =.007) (Table 5).

Partial correlations for exercise health-promoting behavior showed that exercise self-efficacy offered a statistically significant contribution to the variance of exercise health-promoting behaviors (F [ 1, 176] = 173.58, p < .0001). Education contributed a significant but smaller portion of the variance (F [1, 176] = 4.07, p = .0452). This information is represented in Table 6.

Finally, partial correlations were examined for nutrition health-promoting behaviors. Again, self-efficacy for healthy eating offered the greatest unique contribution to the variance in health-promoting behaviors for healthy nutrition (F [1, 176] = 133.23, p <.0001). Age also offered a smaller but significant contribution (F [1, 176] = 6.04,p = .0150) (Table 7).

LIMITATIONS

Racial homogeneity of the sample as well as higher income and educational attainment could impact the generalizability of the findings. Participants in this study were more likely to hold bachelor's degrees and report higher household income than the general population (U.S. Department of Commerce, 2001). To determine whether similar results would be found in other groups of men, an attempt should be made to conduct this type of inquiry with more diverse populations. There may have also been a self-selection bias in favor of those who have more interest in healthrelated topics, or participants may have overreported their actual participation in health-promoting behaviors.

DISCUSSION

In this study, self-efficacy clearly overshadowed other constructs as the most important predictor of overall, nutrition, and exercise health-promoting behaviors. This finding is consistent in studies of health-promoting behaviors with other populations (Clark, 1996; Gillis, 1993; Kelly, Zyzanski, & Alemango, 1991). These relationships do not establish causality; it is not possible to determine whether self-efficacy promotes greater participation in health-promoting behaviors or whether successful accomplishment of health-promoting behaviors promotes greater levels of self-efficacy. The relationship between self-efficacy and the target behavior, in this case health-promoting behavior, is likely reciprocal in nature.

However, this finding raises the question of whether it is the HPM or self-efficacy alone accounting for the variation in men's behavior. The notion of self-efficacy encompasses the confidence one has in his own capacity to successfully execute a behavior with desirable outcomes (Bandura 1997). Self-efficacy is a construct in social cognitive theory and has been added to other health-behavior theories including the HPM (Bandura 1997). Recently self-efficacy is being referred to in the literature as a theory (Bandura 1997; Holloway, & Watson, 2002). Perhaps future studies of health-promoting behaviors in middle-aged and older men should focus exclusively on self-efficacy.

Age did not influence men's self-efficacy for health-promoting behaviors. This is encouraging, since older adults are sometimes thought to be passive, dependent, and incapable (Adams-Price, Henley, & Hale, 1998; Barusch, 1997; Bland, 1997). In the study, advancing age was positively associated with healthier eating behaviors, stress management, and health responsibility but negatively associated with perceived benefits of HPB. These findings may represent a developmental/maturation effect or may be influenced by beliefs and behaviors common to older age cohorts.

Compared to the general population, respondents on the higher end of the age spectrum had the tendency to rate their health more favorably (Eberhardt et al., 2001). Self-rated health was positively related to most specific health-promoting behaviors, as well as self-efficacy toward these behaviors. The positive association of health, health-promoting behaviors, and self-efficacy has been a fairly consistent finding in other studies (Gillis, 1993; Grembowski et al., 1993; Loeb, 2004; Lucas, Orshan, & Cook, 2000; Misener, Phillips, & McGraw, 2000; Tran, Wright, & Chatters, 1991). Relationships were also demonstrated for self-rated health and self-efficacy, benefits, and barriers, all lending support to the HPM. Individuals who describe their health as poor may lack confidence and see many barriers thwarting their ability to engage in health-promoting behaviors.

Lower levels of income and educational attainment were associated with fewer health-promoting behaviors. These relationships support but are certainly not unique to the HPM. The Institute of Medicine (2003) reports this complex relationship as a "striking and consistent association" (p. 60). The social environment impacts access to resources for health-promoting behaviors and shapes normative behaviors toward healthy (or unhealthy behaviors), further explaining the relationship between low income, less education, lower self-efficacy scores, and fewer health-promoting behaviors.

In this study, perceived benefits offered no unique contribution to any of the three health-promoting behavior regression models. Studies have shown perceived benefits to be strongly associated with health-promoting behaviors (Kelly, Zyzanski, & Alemango, 1991; Loeb, 2004; Wang, 1999). The relative homogeneity of benefit scores might explain this finding. Another possibility is the conventional gap between knowing the right thing to do and actually doing the right thing. For example, many people with unhealthy diets and sedentary lifestyles can readily state the benefits of healthy nutrition and physical activity.

CONCLUSIONS AND IMPLICATIONS FOR FUTURE RESEARCH

The Health Promotion Model provided the framework for this study of health-promoting behaviors among middle-aged and older men. Multiple regression models were created using constructs in the HPM to examine variation in overall, exercise, and nutrition health-promoting behaviors. These three separate models explained appreciable amounts of variance in health-promoting behaviors: 66% for overall, 65% for exercise, and 52% for nutrition health-promoting behaviors. These findings lend support to the capacity of the HPM as an explanatory model in this population. The next logical step might be to develop health-promotion interventions for men around constructs of the HPM, particularly self-efficacy, to test the effectiveness of theory-based interventions.

An impetus for this research was a concern about health disparities between men and women, particularly differences in mortality. Some would characterize men as vulnerable in this regard. Behaviors are important determinants of health but not the only determinants. Other determinants include genetics, environment, access to healthcare, and social determinants such as income and education (Institute of Medicine, 2003; USDHHS, 2000). Results of this study lent support to the notion that social determinants are related to health, since men with lower levels of education and income tended to rate their health less favorably and participated to a lesser degree in health-promoting behaviors. Addressing health-promoting behaviors or even illness-avoidance behaviors alone will likely be insufficient to address disparities in men's health. This study may be one of the first to explore health-promoting behaviors in men age 45 and older. The findings suggest the need for further research, development, and intervention testing.

REFERENCES

Adams-Price, C.E., Henley, T.B., & Hale, M. (1998). Phenomenology and the meaning of aging for young and old adults. International Journal of Aging and Human Development, 47(4), 263-277.

Babbie, E. (1997). Survey research methods (2nd ed.). Belmont, CA: Wadsworth Publishing Company.

Bagwell, M.M., & Bush, H. (2000). Improving health promotion for blue-collar workers. Journal of Nursing Care Quality, 14(4), 65-71.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W.H. Freeman and Company.

Barusch, A.S. (1997). Self-concepts of low-income women: Not old or poor, but fortunate and blessed. International Journal of Aging and Human and Development, 44(4), 269-282.

Becker, H., Stuifbergen, A.K., & Sands, D. (1991). Development of a scale to measure barriers to health promotion activities among persons with disabilities. American Journal of Health Promotion, 5, 449-454.

Becker, H., Stuifbergen, A.K., Oh, H.S., & Hall, S. (1993). Self-rated abilities for health practices: A health self-efficacy measure. Health Values, 17(5), 42-50.

Bland, J.H. (1997). Live long, die fast. Minneapolis: Fairview Press.

Bungum, T.J., Orsak, K. C., & Chng, C.L. (1997). Factors affecting exercise adherence at a worksite wellness program. American Journal of Health Behavior, 21, 60-66.

Clark, D.O. (1996). Age, socioeconomic status, and exercise self-efficacy. The Gerontologist, 36, 157-164.

Courtney, W.H. (2000). Constructions of masculinity and their influence on men's well-being: A theory of gender and health. Social Science and Medicine, 50, 1385-1401.

deChesnay, M. (2005). Caring for the vulnerable. Sudbury, MA: Jones & Bartlett.

Denton, D., & Walters, V. (1999). Gender differences in structural and behavioral determinants of health: An analysis of the social production of health. Social Science and Medicine, 48, 1221-1235.

Desmond, A.W., Conrad, K.M., Montgomery, A., & Simon, K.A. (1993). Factors associated with male workers engagement in physical activity: White collar vs. blue collar workers. AAOHN, 41, 73-83.

Doak, C.C., Doak, L.G., & Root, J.H. (1995). Teaching patients with low literacy levels. New York: Lippincott.

Duffy, M.E., Rossow, R., & Hernandez, M. (1996). Correlates of health-promotion activities in Mexican American women. Nursing Research, 45, 18-24.

Eberhardt, M.S., Ingram, D.D., Makuc, D.M., Pamuk, E., Freid, V.M., Harper, S.B., et al. (2001). National Center for Health Statistics United States 2001 with urban and rural health chartbook. Hyattsville, MD: U.S. Government Printing Office.

Foster, M.F. (1992). Health promotion and life satisfaction in elderly black adults. Western Journal of Nursing Research, 14, 444-463.

Fries, J.F. (1999). The case for healthy aging and compression of morbidity. In K. Dychtwald (Ed.), Healthy aging: Challenges and solutions (45-61). Gaithersburg, MD: Aspen Publishing Company.

Gillis, A.J. (1993). Determinants of a health-promoting lifestyle: An integrative review. Journal of Advanced Nursing, 18, 345-353.

Grembowski, D. Patrick, D., Diehr, P., Durham, M. Beresfor, S., Kay, E., et al. (1993). Self-efficacy and health behavior among older adults. Journal of Health and Social Behavior, 34, 89-104.

Hazzard, W.R. (2000). Preventive gerontology: Optimizing health and longevity for men and women across the lifespan. The Journal of Gender-Specific Medicine, 3(3), 28-34.

Holloway, A., & Watson, H.E. (2002). Role of self-efficacy and behaviour change. International Journal of Nursing Practice, 8, 106-115.

Institute of Medicine. (2003). The future of the public's health in the 21st century. Washington DC: National Academies Press.

Jones, J. (2001). Around the globe, women outlive men. Population Today. Retrieved November 30, 2001, from http://www.prb.org/PrintTemplate.cfm?HTMLDisplay.cfm&ContentID=367.

Jones, M., & Nies, M.A. (1996). The relationship of perceived benefits of and barriers to reported exercise in older African American women. Public Health Nursing, 13, 151-158.

Kelly. R.B., Zyzanski, S.J. & Alemango, S.A. (1991) Prediction of motivation and behavior change following health promotion: Role of health beliefs, social support, and self-efficacy. Social Science in Medicine, 32(3), 311-320.

Levinson, R.J. (1986). A conception of adult development. American Psychologist, 41, 3-13.

Loeb, S.J. (2003). Older men's health program inventory: An inventory for assessing health practices and beliefs. Geriatric Nursing, 24(3), 278-285.

Loeb, S.J. (2004). Older men's health: Motivation, self-ratings, and behaviors. Nursing Research, 53(3), 198-205.

Lucas, J.A., Orshan, S.A., & Cook, F. (2000). Determinants of health-promoting behavior among women ages 65 and above living in the community. Scholarly Inquiry for Nursing Practice, 14, 77-100.

Lusk, S.L., Ronis, D.L., Kerr, M.J., & Atwood, J.R. (1994). Test of the health promotion model as a causal model of workers using hearing protection. Nursing Research, 43, 151-157.

Misener, T.R., Phillips, K.D., & McGraw, E. (2000). Psychosocial development and health-promoting lifestyle. Journal of Theory Construction and Testing, 4, 14-19.

Murdaugh, C.L., & Hinshaw, A.S. (1986). Theoretical model testing to identify personality variables effecting preventive behaviors. Nursing Research, 35, 19-23.

National Center for Health Statistics. (2004). Health, United States, 2004 with chartbook on trends in health of Americans. U.S. Government Printing Office: Hyattsville, MD.

Nunnally, J.C., & Berstein, I.H. (1994). Psychometric theory (3rd ed.). New York: McGraw-Hill.

Pender, N.J., Murdaugh, C.L., & Parsons, M.A. (2002). Health promotion in nursing practice (4th ed.). Upper Saddle River, NJ: Prentice Hall.

Rea, L.M., & Parker, R.A. (1997). Designing and conducting survey research. San Francisco: Jossey-Bass.

Reed, D.M., Foley, D.J., White, L.R., Heimovitz, Burchfiel, C.M., & Masaki, K. (1998). Predictors of healthy aging in men with high life expectancies. American Journal of Public Health, 88, 1463-1468.

Stuifbergen, A.K., Seraphine, A., & Roberts, G. (2000). An explanatory model of health promotion and quality of life in chronic disabling conditions. Nursing Research, 49, 122-129.

Tapler, D.A. (1996). The relationship between health value, self-efficacy, health locus of control, health benefits, health barriers, and health behavior practices in mothers. Dissertation Abstracts International, 5 7(01), 3132B, (UMI No. 9630152).

Torabi, M.R. (1991). Factors affecting response rate in mail survey questionnaires. Health Values, 15(5), 57-59.

Tran, T.V., Wright, R., & Chatters, L. (1991). Health, stress, psychological resources and subjective well-being among older blacks. Psychology and Aging, 6, 100-108.

U.S. Department of Commerce. (2001). Bureau of the Census.

Wang, H. (1999). Predictors of health promotion lifestyle among three ethnic groups of elderly rural women in Taiwan. Public Health Nursing, 16(5), 321-328.

Weitzel, M.H. (1989). A test of the health promotion model with blue collar workers. Nursing Research, 38, 99-104.

RITA E. ARRAS

Southern Illinois University, Edwardsville

ROBERTA J. OGLETREE

KATHLEEN J. WELSHIMER

Southern Illinois University, Carbondale

Correspondence concerning this article should be sent to Rita E. Arras, PhD, RN, SIUE School of Nursing, Box 1066, Edwardsville, IL 62026-1066. Electronic Mail: [e-mail].

Table 1
Summary and Comparison of Demographic
Characteristics of Sample

                               Sample
Characteristic                 Number    Sample %       State

Education                                (n = 188)
  High School or Above           180       95.3          83.5
  Bachelor's Degree or Above     101       53.0          27.0

Age Range (a)                            (n = 191)
  45-54                          39        20.3          42.1
  55-64                          40        20.9          26.2
  65-74                          61        32.2          18.1
  75-84                          46        24.1          10.8
  85-90                           5         2.6           2.8

                               Sample Median          State Median
                                  (Range)
Median Household Income (n = 181)

                               $50,000-59,999           $41,994

(a) These state percentages for age represent the percentage
of men age 45 and older rather than all men.

Table 2
Health-Promoting Behavior and
Behavior-Specific Cognition Scores

                             Range      Possible    Mean     SD
Scale / Subscale            of Scores    Scores

HPLPII Total                1.52-3.81     1-4       2.71    0.43
  Nutrition                 1.00-4.00     1-4       2.69    0.56
  Physical Activity         1.00-4.00     1-4       2.23    0.72
  Stress Management         1.50-4.00     1-4       2.70    0.50
  Interpersonal Relations   1.33-4.00     1-4       3.01    0.53
  Spiritual Growth          1.33-4.00     1-4       3.11    0.57
  Health Responsibility     1.00-3.78     1-4       2.45    0.58

Self-Efficacy Total          34-138     28-140    101.62   18.38
  Nutrition                   7-35        7-35     26.19    5.34
  Exercise                    7-35        7-35     22.75    7.50
  Health Responsibility       8-35        7-35     28.00    5.06
  Stress Management          10-35        7-35     24.68    5.23

Total Benefits               33-118     20-129     87.96   14.08
  Exercise                   14-60       10-60     42.54    8.14
  Nutrition                   6-36        6-36     26.94    4.74
  Barriers                   18-72       18-72     27.70    8.51

HPLPII = Health-promoting lifestyle profile; self-efficacy measured
by Self-Rated Abilities for Health Practices, benefits measured by
Benefits Scale, and barriers measured by the Barriers to Health
Promoting Activities for Disabled Persons.

Table 3
Summary of Regression Models for Total, Exercise,
and Nutrition Health-Promoting Behaviors

                                       Sums of     F        p
Source               [R.sup.2]   df    Squares   Value    Value

Total HPB Model         0.6635    12   21.011    26.95   < .0001

Error Total HPB Model            164   10.653
Exercise Model          0.6501    12   58.231    25.31   < .0001
Error Exercise Model             164   31.344

Nutrition Model         0.5242    12   29.524    15.67   < .0001
Error Nutrition Model            164   25.745

HPB = Health-promoting behavior (overall).

Table 4
Intercorrelations between Independent Variables

Age                              1.        2.         3.

Education                     -.167
Income                        -.415 **   .618 **
Self-Rated Health              .034      .218 **    .239 **
Overall Self-Efficacy (SE)    -.063      .274 **    .304 **
Exercise SE                   -.060      .347 **    .321 **
Nutrition SE                  -.031      .261 **    .271 **
Stress Management SE          -.083      .144 *     .225 **
Health Responsibility SE      -.022      .062       .105
Benefits Overall              -.156 *    .363 **    .325 **
Benefits Exercise             -.126      .326 **    .317 **
Benefits Nutrition            -.167 *    .338 **    .297 **
Barriers                       .134     -.099      -.245 **
Interpersonal Relationships    .011      .114       .139

Age                              4.        5.         6.

Education
Income
Self-Rated Health
Overall Self-Efficacy (SE)    .401 **
Exercise SE                   .421 **    .801 **
Nutrition SE                  .327 **    .833 **    .562 **
Stress Management SE          .303 **    .767 **    .414 **
Health Responsibility SE      .175 *     .773 **    .404 **
Benefits Overall              .326 **    .418 **    .382 **
Benefits Exercise             .381 **    .430 **    .453 **
Benefits Nutrition            .164 *     .308 **    .216 **
Barriers                     -.319 **   -.261 **   -.154 *
Interpersonal Relationships   .200 **    .548 **    .265 **

Age                             7.         8.         9.

Education
Income
Self-Rated Health
Overall Self-Efficacy (SE)
Exercise SE
Nutrition SE
Stress Management SE          .533 **
Health Responsibility SE      .584 **    .575 **
Benefits Overall              .382 **    .258 **    .288 **
Benefits Exercise             .370 **    .251 **    .284 **
Benefits Nutrition            .315 **    .258 **    .207 **
Barriers                     -.171 *    -.303 **   -.227 **
Interpersonal Relationships   .378 **    .638 **    .538 **

Age                             10.        11.

Education
Income
Self-Rated Health
Overall Self-Efficacy (SE)
Exercise SE
Nutrition SE
Stress Management SE
Health Responsibility SE
Benefits Overall
Benefits Exercise             .746 **
Benefits Nutrition            .744 **    .595 **
Barriers                     -.218 *    -.202 *
Interpersonal Relationships   .241 **    .194 **

Age                             12.        13.

Education
Income
Self-Rated Health
Overall Self-Efficacy (SE)
Exercise SE
Nutrition SE
Stress Management SE
Health Responsibility SE
Benefits Overall
Benefits Exercise
Benefits Nutrition
Barriers                      -.240 **
Interpersonal Relationships    .237 **    -.165 *

* Significance at 5%. ** Significance at 1% or less.

Table 5
Unique Variance Predicted by Independent
Variables: Total Health-Promoting
Behaviors
                         Type III
                         Sums of                         Partial
Variable          df      Square    F Value   p Value   [R.sup.2]

Age             1, 176    0.483      7.43      .0071      0.023
SE Total        1, 176   14.065    216.53    < .0001      0.669
Education       1, 176    0.170      2.62      .1072      0.008
Income          5, 172    0.361      1.11      .3564      0.017
Health          2, 175    0.035      0.27      .7625      0.002
Barrier         1, 176    0.017      0.27      .6051     < .001
Benefit Total   1, 176    0.007      0.11      .7367     < .001

SE Total = Total self-efficacy.

Table 6
Unique Variance Predicted by Independent
Variables: Physical Activity

                      Type III
                      Sums of                         Partial
Variable       df      Square    F Value   p Value   [R.sup.2]

SE PA        1, 176    33.176    173.58    < .0001     .5700
Health       2, 175     1.019      2.67      .0725     .0180
Age          1, 176     0.003      0.01      .9079    <.001
Education    1, 176     0.778      4.07      .0452     .0130
Income       5, 172     0.755      0.79      .5580     .0120
Benefit PA   1, 176     0.027      0.14      .7072    <.001
Barrier      1, 176     0.300      1.57      .2124     .005

SEPA= Self-efficacy physical activity.

Benefit PA = Benefit physical activity.

Table 7
Unique Variance Predicted by Independent
Variables: Nutrition

                      Type III
                      Sums of
Variable      df       Square     F Value

Age         1, 176      .948       6.04
SE Nut      1, 176    20.917     133.25
Health      2, 175      .048       0.15
Education   1, 176      .272       1.73
Income      5, 172      .872       1.11
Barrier     1, 176      .065       0.42
Ben Nut     1, 176      .053       0.34

                       Partial
Variable    p Value   [R.sup.2]

Age          .0150      .032
SE Nut      <.0001      .709
Health       .8591      .002
Education    .1897      .009
Income       .3567      .030
Barrier      .5195      .002
Ben Nut      .5610      .002

SE Nut = Self-efficacy nutrition,
Ben Nut = Benefits nutrition.