To expand the body of knowledge and provide further insight into the complex area of homelessness and health, health practices of sheltered homeless women were investigated using a crosssectional, descriptive, and non-experimental design using Pender’s Health Promotion Model as the theoretical framework. The sample (w = 137) was weU educated, mostly unemployed, primarily single, and homeless due to relationship problems/conflict per self-report.
Homeless women were noted to practice health-promoting behaviors in all areas but scored the lowest on physical activity and nutrition. Significantfindingsreflected women’s personal strengths and resources in the areas of spiritual grovrth and interpersonal relations. Keywords: health disparities, health promotion, hometessness, quantitative research, women’s health acute and chronic physical disorders and mental health/emotional issues.
‘”‘” Few studies address the positive health practices of this vulnerable population. The ptirpose of this study was to describe sociodemographic and personal characteristics, health practices, and health-promoting behaviors in a population of sheltered homeless women in a specific Midwest geographical region to increase awareness, understanding, and provide further insight into the complex area of homelessness and health. The theoretical framework for this study was based upon Pender’s revised Health Promotion Model.
” The Health Promotion Model (HPM) provides a framework to examine influences on participation in healthpromoting behaviors and provides direction From the Department of Nursing, School of Health for effective interventions. The HPM illusSciences, University of Saint Francis, Fort Wayne, trates that each person is a multidimensional Indiana. holistic individual who continually interacts The author thanks her dissertation committee for support of this research project and the dissertation pro- with both interpersonal and physical envicess: L.
Kathleen Sekula, PhD, APRN, Chair, Duquesne ronments and emphasizes the active role University, Pittsburgh, Pa; Rick Zoucha, DNSc, APRN, Duquesne University; and Jacquelyn Feller, PhD, Uni- of the individual in the achievement of an versity of Saint Francis, Fort Wayne, Ind. This study improved healthy state. The 3 major conwas supported by grants from the St. Joseph Commustructs of the HPM (individual characterisnity Health Foundation and the Midwest Alliance for tics and experiences, behavior-specific cogHealth Education. Fort Wayne, Ind.
nition and affect, and behavioral outcomes) Correspondence to: Meg Wilson, PhD, RN, Department were used to select specific study variables of Nursing, School of Health Sciences. University of Saint Francis, 2701 Spring Street, Fort Wayne, IN 46808 as conceptualized within the model. Health(e-mail: [email protected] edu). promoting behaviors, the outcome of the 51 OMELESSNESS IS A critical concern for all communities as the nation faces continued uncertainty in unstable economic markets and worldwide events. Throughout history, the magnitude of homelessness has fluctuated in response to current economic, political, and social environments.
Society has observed homelessness change and evolve from a primarily male-oriented population to a more heterogeneous group. Today, the demographic scope of homelessness includes a rapidly growing segment of young single women, alone or accompanied by their children. ”^ Health problems of the homeless have been well documented and include H 52 FAMILY ; COMMUNITY HEALTH/JANUARY-MARCH 2005 clude routine preventive and health promotion care is more difficult to obtain for homeless women than for men,”*’^-^’ exacerbating already serious health problems.
‘^ While access to care may be available at some level, several studies have documented high rates of emergency room utilization that traditionally have focused on secondary and tertiary level care and have not included primary level interventions of health promotion and disease Many descriptive studies have documented sociodemographic characteristics of the homeless along with precursors, consequences, and physical and mental health problems; few have examined healthpromoting behaviors and lifestyles.
Great variation exists within the published literature in regard to conceptual definitions of health-promoting practices, especially in homeless populations. While many defme these practices as participation in preventative measures, such as Pap tests, HIV tests, and tuberculosis screening,”*-^^ health promotion is more accurately defined as a dynamic process in which behaviors are intended to expand the positive health potential by increasing well-being of the individual and actualizing human potential through participation in a lifestyle that supports a holistic perspective of health.
” Alley and colleagues^^ investigated 59 low-income and homeless women who sought care at a nurse-managed clinic using the Health Promoting Lifestyle Profile II (HPLP II), noting low participation in all areas of healthpromoting behaviors (exercise, nutrition, health-responsibility, interpersonal relations, stress management, and spiritual growth). Level of education was significantly related to the HPLP II, but housing (stable versus unstable) did not influence the practice of health-promoting behaviors.
Using the HPLP, Brady and Nies^^ found that older African American women living below the poverty level engaged in fewer health-promoting behaviors than those living above the poverty level, particularly in the area of exercise. Similar levels of health-promoting behaviors HPM, were examined in this population and relationships among study variables explored. Homelessness is conceptualized currently through the perspective of two levels of influences: individual and structural. Individual level issues and problems are represented by personal characteristics that contribute to vulnerability and the risk of homelessness.
These consist of psychosocial issues such as adverse early childhood experiences, mental/emotional illness and health, substance abuse, domestic violence, and sociodemographic factors including gender, age, level of education, and ethnicity. ^^’^ Structural issues occur at a societal level, contribute to the risk of homelessness, and include conditions of poverty, unemployment, lack of affordable housing, gender-related problems, insufficient income, inadequate social services and health care, and an increase of femaleheaded famiUes.
^-””^ Health disparities are prevalent in the homeless and are well documented. Significantly higher mortality rates have been noted in the homeless when compared to the general US population,’^ and the homeless often delay seeking medical treatment until their symptoms become intolerable or severe, waiting up to 3 months before seeking treatment. ^ Homeless women have special needs in relation to their health; they have poorer health when compared with the general population, experience an elevated prevalence of acute and chronic diseases, and report a decreased health status.
^-^’^ Higher rates (4 to 8 times) of asthma, anemia, and ulcer disease were found in homeless and low-income mothers in the Worcester Family Research Study’^ when compared to a general population sample of women. Homeless women were significantly more likely to report their health status as poor or fair,'”‘^ as compared to the general population. ‘^ However, some homeless women have been found to actively seek treatment for long-term chronic conditions (hypertension, positive TB skin test) reflecting understanding and a value of health.
^° Access to affordable, high quality, and comprehensive health care and programs that in- Health-Promoting Behaviors of Sheltered Homeless Women While many define these practices as participation in preventative measures, sucb as Pap tests, HIV tests, and tuberculosis screening, health promotion is more accurately defined as a dynamic process in which behaviors are intended to expand the positive health potential by increasing well-being of the individual and actualizing human potential through participation in a lifestyle that supports a holistic perspective of health.
maintained with a designated staff person at each shelter; data collection sessions were scheduled following specified time frames as established by each shelter and ranged from two times per week to once every two weeks in response to individual shelter census. All data were collected by this investigator; a trained female research assistant (master’sprepared nurse) was utilized when 5 or more women were scheduled for data collection at one time. All of those meeting inclusion criteria were invited to participate.
After a verbal and written explanation of the study purpose, confidentiality of responses, and withdrawal rights, research instruments were distributed and self-reported. All potential participants were told that the decision to participate or not participate would in no way jeopardize their present status in the shelter and that they had the right to not finish the questionnaires once started without retribution. All women approached agreed to participate, signed a written consent form, and completed research questionnaires.
In appreciation for their participation in this study, $5 cash was given to each participant after completion of the research questionnaires. If accompanied by children, a small gift (ageappropriate book/toy) was provided for each child residing with his/her mother at the shelter. Sociodemographic and personal characteristics were measured by the Personal History Form, developed by this researcher. Data collected were organized into 3 categories: demographics, health, and homeless history.
Readability was assessed at less than a 5th grade reading level using the Flesch-Kincaid Grade Level measurement computed by Microsoft Word software. Health-promoting behaviors were measured with the Health-Promoting Lifestyle Profile II (HPLP II). The HPLP II is used to identify patterns of health promotion lifestyles and health-promoting behaviors conceptualized as a multidimensional pattern of selfinitiated actions and perceptions that serve to maintain or enhance the level of wellness. ” 53 were noted in low-income high and low-risk pregnant women.
27 METHODS A descriptive, cross-sectional, and nonexperimental design was used for this study. Nine not-for-profit urban shelters located in the Midwest, providing housing assistance to homeless women were identified through a local community resource manual. Services offered varied slightly among the shelters, but mainly included case management and daily living skills and personal growth programs; none offered direct health care services, referring residents to local indigent health care clinics for physical health care needs.
Sample size was determined by a power analysis based on identified study variables and published literature on the HPLP II questionnaire used in a similar study population. ^^ A sample size of at least 110 participants was estimated to provide a power of 0. 80 needed for estimating correlations in a medium effect size of 0. 40. Inclusion criteria included homeless women who were registered residents of the shelters and could read and understand the English language. Following Institutional Review Board and agency administration approval, data were collected over a 5-month period (May through September).
Weekly telephone contact was 54 FAMILY & COMMUNITY HEALTH/JANUARY-MARCH 2005 The HPLP II consists of a 52-item scale that encompasses a total score and 6 subscaies (8 to 9 items each): health responsibility, physical activity, nutrition, spiritual growth, interpersonal relations, and stress management. Items are scored with a 4-point response format: never = 1; sometimes — 2; often = 3; and routinely — 4, and scores are calculated for the total score and 6 subscale scores by calculation of a mean of the individual’s responses.
Validity and reliability for the HPLP II have been previously established; internal consistency reliability was reported as 0. 94 (total scale) and a range from 0. 79 to 0. 87 (subscales) [personal communication, SN Walker, March 16, 2002]. ^^ In this study, internal consistency reliability of the HPLP II total score was 0. 95 and subscaies were: health responsibility (0. 83), physical activity (0. 81), nutrition (0. 75), spiritual growth (0. 88), interpersonal relations (0. 83), and stress management (0. 80).
The HPLP and HPLP II have been used to determine the health-promoting lifestyles of diverse populations including homeless lower income African American low income pregnant women,^’ women with breast cancer,^^ mothers from the Midwest,^° older adults,^’ employed men and women,’^ women with multiple sclerosis,’^ and young adults. ”’ Duplication of questionnaires was established by comparison of names of participants on consent forms; 6 cases were discarded due to duplication. After careful comparison of the 9 shelters, 5 were selected for inclusion for final data analysis CN — 143), and 4 were not (A’^ = 32).
Shelters not included had similar organizational/administrative structure to those selected but only admitted distinct subgroups (substance abuse and mental illness), focused on provision of long-term transitional housing services, did not provide emergency shelter services, and required a formal referral process from outside sources (eg, other agencies, shelters, court system) as compared to those included in the sample (general population, emergency housing, and self- or help-line referral). Data from the 5 shelters were combined as one group to reach the target sample size as determined by the power analysis.
There were no missing data on any of the remaining questionnaires iN= 137). Data were analyzed for the individual 5 shelters and for the total sample for commonalities and differences. Univariate descriptive statistics (frequencies, percentages, means, standard deviations, and ranges) were used to describe sample characteristics and health-promoting behaviors. One-way analysis of variances (ANOVA) and Chi-square test of independence were utilized to examine differences among the five shelters. Bivariate descriptive statistics consisting of Pearson’s r and Eta correlations were utilized to examine for relation between study variables.
RESULTS Data showed consistency in the characteristics of the sample when examined by individual shelters and the total sample (A’^ = 137). Demographic data are summarized in Table 1. Overall, participants’ ages ranged from 18 to 60 years, with a mean age of 36 years. The majority of w^omen identified themselves as White (53%) or African American (43. 8%). Nearly half (43. 8%) of the sample was single-never married and 27% were divorced. The study sample was well educated; 78% had a high school degree or advanced education relative to the education of homeless women in previous research. ‘^’^’^^ Most (80.
3%) reported that they were unemployed at the time of data collection. Number of children ranged from 0-7, with a mean of 2. 2 children. Median length of stay was 2 weeks, with one week most frequently reported at time of data collection. Distribution of length of stay in the shelter was highly skewed; range was less than 1 week to 57 weeks. However, 80% of the sample had been living at the shelter for 4 weeks or less at the time of data collection. Relationship problems/conflict was identified as the primary factor contributing to the current homeless state (46%), Health-Promoting Behaviors of Sheltered Homeless Women Table 1.
Personal history form: demographic data for total sample (N = 137) Variable Race/ethnicity White African American Latina Asian Native American Marital status Single (never married) Married 1 Separated Divorced Widow Education College degree Some college Trade/vocational High school degree Some high school 8th grade or less Employment Eull time Total African American White Part time Total African American White Not employed Total African American White Number of children Age 55 Total, n (%) 73 (53. 3) 60 (43. 8) 2(1. 5) 2(1. 5) 0 (0. 0) 60 (43. 8) 8(13. 1) 20 (14. 6) 37 (27. 0) 2(1. 5) 11(8. 0) 43(31. 4) 17(12.
4) 36 (26. 3) 25 (18. 2) 5 (3. 6) 14(10. 2) 7(11. 7) 7 (9. 6) 13 (9. 5) 10(16. 7) 3(4. 1) 110(80. 3) 43(71. 7) 63 (86. 3) Mean 2. 2 SD 1. 60 Mean 36 years SD 11. 01 ber (44. 5%) cited previous homelessness at some time during their lives. Twenty-one percent identified a childhood history of foster care. Table 2 summarizes personal health information from the Personal History Form. Overall, the majority (70. 8%) identified a high level of perceived health status. Health care was reported as accessible w^ith a small number (11. 7%) citing the usual location of care as the emergency room. Barriers to health care Table 2.
Personal health data: health status, healthcare provider, barriers to health care physical diseases (A^ = 137) Variable n % and additional circumstances were identified as eviction/lack of money to pay rent (35. 8%), loss of job (30. 7%), violence (24. 1%), drugs/alcohol (23. 4%), and emotional/mental illness (22. 6%). Half of the women (49. 5%) reported that they had been residing with friends or family prior to this current state of homelessness, and nearly the same num- Health status Excellent 10 Very Good 23 Good Fair 30 Poor 10 Location of health care provider Doctor’s office 49 Public clinic 51 Emergency room 16
Nowhere 21 Barriers to health care Money 87 Transportation 44 Unsure where to go 23 Nervous/afraid 18 Ghildcare 5 Lack of trust of doctors 15 Lack of trust of nurses 3 Language 2 Nothing 31 Physical diseases Asthma 37 Chronic bronchitis 35 Hypertension 28 Heart disease 12 Ulcer 21 Gancer 7 Arthritis 23 Diabetes 13 STD 23 7. 3 16. 8 46. 7 21. 9 7. 3 35. 8 37. 2 11. 7 15. 3 63. 5 32. 1 16. 8 13. 1 3. 6 10. 9 2. 2 1. 5 22. 6 27. 0 25. 5 20. 4 8. 8 15. 3 5. 1 16. 8 9. 5 16. 8 56 FAMILY & COMMUNITY HEALTH^ANUARY-MARCH 2005 lated to small sample size of individual shelters.
Significant correlations were found between age and health status and health index: although a moderate relationship, older homeless w^omen were more likely to have a greater number of physical diseases (r = 0. 29, p < 0. 01) and identify their health status as worse (r — 0. 19; p < 0. 05). Another significant correlation was noted between health index (number of physical diseases) and the subscale of health responsibility (r = 0. 18; p < 0. 05): again, although weak, those reporting more physical diseases were more likely to practice more health behaviors related to health responsibility.
Table 5 displays a correlation matrix of Pearson correlations for selected variables of age, number of children, health status, and health index with the HPLP II total score and 6 subscaies. Health status was significantly negatively correlated with HPLP II total score, and the subscaies of nutrition, spiritual growth, and stress management. Homeless women were significantly more likely to practice health-promoting behaviors related to the total lifestyle profile (r = —0. 22, p < 0. 01), nutrition (r = -0. 21, p < 0. 01), spiritual growth (r = -0. 22, p < 0. 01), and stress management (r = —0. 25, p < 0.
01) when they associated their health status as higher (good or very good). As expected, significant positive relationships (r = 0. 66 to 0. 86, p < 0. 01) were demonstrated between the HPLP II total score and all 6 subscaies. The strongest relationships among the subscaies were noted between spiritual growth and interpersonal relations (r = 0. 78, p < 0. 01), and spiritual growth and stress management (r — 0. 72, p < 0. 01). This suggests that those who participated in spiritual growth behaviors also participated in behaviors of interpersonal relations and stress management. Several findings were noted involving health responsibility.
Significant positive relationships were noted between health responsibility and nutrition (r = 0. 59,p < 0. 01), spiritual growth (r = 0. 62; p < 0. 01), interpersonal relations (r = 0. 69, p < 0. 01), and stress management (r = 0. 59, p < 0. 01). Homeless women who were diverse with money (63. 5%) and transportation (32. 1%) tnost frequently identified. Asthma (27%), chronic bronchitis (25. 5%), and hypertension (20. 4%) were the most commonly self-reported diseases identified by the women. A large number of participants (67. 9%) reported one or more physical diseases; 35% had two or more conditions.
Tobacco use was widespread (68.6%), with most using cigarettes at a rate of one or more packs per day (47. 5%). Supporting the reported access to health care, 84. 7% had received medical care, and 63. 5% had a Pap test during the past tw^o years. Dental and vision care were identified as the greatest unmet needs; nearly 50% had not received these health care services in over two years. Descriptive analyses of the HPLP II (total score, 6 subscaies, and individual subscale items) are presented in Tables 3 and 4. Variability was noted within all subscaies as evidenced by ranges and standard deviations for each; physical activity and nutrition were the lowest.
To assess if differences existed between the shelters and the HPLP II total and subscaies scores, a one-way analysis of variance and Chi-square test of independence was performed; no statistical differences were noted in the HPLP II Total and subscaies. A valid Chi-square could not be calculated due to improper loading of cells reTable 3. Means, standard deviations, and ranges for Health-Promoting Lifestyle Profile II (HPLP II) and its subscaies (N = 137) Variable HPLP II total Health responsibility Physical activity Nutrition Spiritual growth Interpersonal relations Stress management Mean 2. 44 2. 38 1. 97 2. 27 2. 86 2. 67 2.
41 SD Range 1. 55-3. 60 1. 22-3. 89 1. 00-3. 63 1. 11-3. 56 1. 22-4. 00 1. 56-4. 00 1. 38-3. 88 0. 46 0. 60 0. 56 0. 52 0. 63 0. 56 0. 57 Health-Promoting Behaviors of Sheltered Homeless Women Table 4. Means and standard deviations of HPLP subscale behaviors (N = 137) Health behavior Health responsibility subscale Read or ^vatch TV programs about improving health. Question health professionals in order to understand their instructions. Get a second opinion when I question my health care provider’s advice. Discuss my health concerns with health professionals. Inspect my body at least monthly for physical changes/danger signs.
Ask for information from health professional about how to take good care of myself. Attend educational programs on personal health care. Seek guidance or counseling when necessary. Physical activity subscale Follow a planned exercise program. Exercise vigorously for 20 or more minutes at least three times a week (such as brisk walking, bicycling, aerobic dancing, using a stair climber). Take part in light to moderate physical activity (such as sustained walking 30-40 minutes 5 or more times a week). Take part in leisure-time (recreational) physical activities (such as swimming, dancing, bicycling).
Do stretching exercises at least 3 times per week. Get exercise during usual daily activities (such as walking during lunch, using stairs instead of elevators, parking car away from destination and walking). Check my pulse rate when exercising. Reach my target heart rate when exercising. Nutrition subscale Limit use of sugars and food containing sugars (sweets). Eat 6-11 servings of bread, cereal, rice, and pasta each day. Eat 2-4 servings of fruit each day. Eat 3-5 servings of vegetables each day. Bat 2-3 servings of milk, yogurt, or cheese each day.
Eat only 2-3 servings from the meat, poultry, fish, dried beans, eggs, and nuts group each day. Read labels to identify nutrients, fats, and sodium content in packaged food. Eat breakfast. Spiritual growth subscale Believe that my life has purpose. Look forward to the future. Feel content and at peace with myself. Work toward long-term goals in my life. Find each day interesting and challenging. Am aware of what is important to me in life. Feel connected with some force greater than myself. Expose myself to new experiences and challenges. Interpersonal relations subscale Praise other people easily for their achievements.
Maintain meaningful and fulfilling relationship with others. Spend time with close friends. Find it easy to show concern, love, and warmth to others. Touch and am touched by people I care about. Find ways to meet my needs for intimacy. Get support from a network of caring people. Settle conflicts with others through discussion and compromise. Mean 2. 13 2. 74 2. 15 2. 49 2. 63 2. 40 1. 77 2. 61 1. 78 2. 05 2. 28 2. 02 1. 90 2. 59 1. 53 1. 61 2. 13 1. 97 2. 17 2. 35 2. 43 2. 55 2. 32 2. 36 3. 05 3. 02 2. 49 2. 80 2. 70 3. 02 2. 98 2. 83 2. 97 2. 73 2. 37 3. 10 2. 83 2. 27 2. 64 2. 59 SD 57 0. 74 0.
91 0. 94 0. 94 0. 97 0. 99 0. 86 0. 90 0. 77 0. 98 0. 93 0. 76 0. 89 0. 94 0. 80 0. 76 0. 82 0. 82 0. 85 0. 87 0. 99 0. 87 1. 00 0. 87 0. 84 0. 87 0. 86 0. 93 0. 94 0. 87 0. 98 0. 79 0. 82 0. 84 0. 91 0. 79 0. 87 0. 86 0. 99 0. 81 (continues’) 58 FAMILY & COMMUNITY HEALTH/JANUARY-MARCH 2005 Table 4. Means and standard deviations of HPLP subscale behaviors (A’^^ 137) (Continued) Health behavior Stress management subscale Take some time for relaxation each day. Accept those things in my life that I cannot change. Concentrate on pleasant thoughts at bedtime. Use specific methods to control my stress.
Balance time between work and play. Practice relaxation or meditation for 15-20 minutes daily. Pace myself to prevent tiredness. Mean SD 2. 41 2. 72 0. 83 0. 84 2. 66 2. 38 2. 38 2. 07 2. 17 0. 90 0. 90 0. 85 0. 95 0. 86 took more responsibility for their personal health also practiced more health behaviors related to nutrition, spiritual growth, interpersonal relations, and stress management. Pearson correlations were also utilized to examine relationships between length of time respondents had been at the shelter when data were collected and the HPLP IL Based on a 2-tailed test, no significant relationships were found.
Relationships between nominal variables of race/ethnicity, education, marital status, and employment status and the HPLP II total score and 6 subscaies were examined with Eta correlations. Weak relationships were noted between race and spiritual growth (0. 21), education and stress management (0. 23), and employment and stress management (0. 22). Table 5. Pearson correlation matrix of selected sociodemographic variables and HPLP II total and subscale scores (A^ = 137) 1 2 3 4 5 6 7 8 9 10 11 l. Age 2. Number of children 3. Health status 4. Health index 5. HPLP II total 6. Health responsibility 7.
Physical activity 8. Nutrition 9. Spiritual 1. 00 0. 15 0. 19* 0. 29* 0. 06 0. 05 0. 17 -0. 02 0. 07 1. 00 -0. 02 1. 00 -0. 04 0. 31+ -0. 03 -0. 22+ -0. 08 -0. 14 -0. 01 -0. 15 0. 03 -0. 21* -0. 01 -0. 22* -0. 03 -0. 10 1. 00 0. 09 1. 00 0. 18* 0. 83+ 1. 00 -0. 02 0. 01 0. 16 0. 09 0. 66+ 0. 45+ 1. 00 0. 78+ 0. 59+ 0. 55+ 1. 00 0. 86+ 0. 62+ 0. 40+ 0. 56+ 1. 00 0. 83+ 0. 69+ 0. 29+ 0. 53+ 0. 78+ 1. 00 0. 84+ 0. 57+ 0. 58+ 0. 55+ 0. 72+ 0. 64+ 1. 00 growth 10. Interpersonal -0. 09 relations 11. Stress 0. 11 management -0. 04 -0. 25+ -0. 01 *p ; 0. 05 level (2-tailed); ^p ; 0. 01 level (2-tailed).
Health-Promoting Behaviors of Sheltered Homeless Women DISCUSSION 59 Characteristics of homeless women are affected by the multiple influences of the geographical region. African Americans were highly represented in this sample as compared to local demographics (43. 8% versus 17. 4%) supporting previous national research. ‘535,36 Although a high rate of unemployment existed in this sample, African Americans had a higher percentage of full and part-time employment as compared to Whites. An unexpected finding was that women in this study were well educated when compared with homeless women in previous research.
^”^’^^ The findings of this study suggest that African American women in this area may be critically affected by complex interacting factors other than race/ethnicity and employment and may have fewer critical social support networks that might assist in the prevention of a homeless situation. The presence of a well-educated sample suggests educational opportunities are accessible, but other factors cause significant impact on the lives of women who become homeless in this region such as increased employment issues related to influences of declining and unstable local and national economic markets.
Although this study’s results were congruent with previous research citing high numbers of co-morbid conditions in the homeless,'”‘^^ limited access to health care was not, suggesting that coordinated and accessible health care was available in the geographical region. Issues related to the availability of primary prevention (health promotion and disease prevention) should be further investigated since other research suggests that most visits to health providers in the homeless are related to acute conditions and access to health promotion and preventive services are limited.
‘^ Women who reported co-morbid physical conditions were more likely to have lower self-perceived health status, validating previous research in similar populations of homeless women both in the Midwest and East Coast. ‘^”^^ Physical prob- lems related to respiratory conditions and hypertension were most prevalent and are consistent with the research of CraftRosenberg and colleagues;” however, rates of hypertension and diabetes were found to be higher when compared to homeless clients seeking care at a free clinic.
‘^ Since African Americans represent nearly half of the sample, it is important to direct further attention to special health care concerns identified in this racial/ethnic group such as screening services for hypertension, heart disease, and diabetes. ‘^””’ Dental and vision care were identified as primary, unmet physical needs, validating findings of others, both in rural and urban populations of homeless women. “^’^ This may be explained by a lack of accessibility for dental/vision services in the local community for homeless women and/or that women may view dental and vision problems as important as other physical needs.
A high rate of tobacco use (68. 6%) was noted in this study sample, which is consistent with other studies of homeless women;’^”^’ however, this rate is twice as high as the average percentage of smokers (27. 2%) as reported for the specific geographic region. ‘*^ These findings are noteworthy as results show that respiratory-related problems were the most frequently identified physical conditions suggesting the need for interventional smoking cessation programs/assistance designed for specific needs and lifestyles of the homeless.
Homeless women were found to participate in a variety of health-promoting behaviors indicating both their ability and interest in their personal health and wellness, despite their current housing crisis. Total levels of health-promoting lifestyles are similar to those found in low-income and homeless women,^” but are lower than others from diverse population groups such as those with Parkinson’s disease,'” African American women,’*” commutiity-dwelling adults,”^ and working adults.
“*^ Previous studies of diverse populations using the first version of the HPLP questionnaire also reported higher mean total scores for low-income pregnant 60 FAMILY & COMMUNITY HEALTH/JANUARY-MARCH 2005 In this study, a significant relationship between health index (number of self-reported physical conditions) and health responsibility subscale reflected that women who identified specific physical problems were cognizant of their problems and practiced more health behaviors directed at addressing their health concerns.
Findings of significant relationships between self-reported level of health status, health index (number of physical conditions), and the HPLP II total scores reflect the women’s understanding that positive behaviors can impact their health. Although homeless women in this study were well educated, no association was found between education and the HPLP II total or subscale scores. These results support the work of Jefferson and colleagues”” who reported no relationships between educational level and the HPLP II total score in a sample of African American women of similar age and educational status.
However, in a similar population of homeless and low-income women. Alley and coUeagues^^ found a significant relationship between education level (mean of 11 years of education) and the total score on the HPLP II questionnaire. Pender’s HPM” is of great value to guide health care interventions for sheltered homeless women and should be used to assess current influences and provide a framework for services directed at increasing their healt