Secondary Logo

Journal Logo

An Environmental Intervention Aimed at Increasing Physical Activity Levels in Low-Income Women

Speck, Barbara J. PhD; Hines-Martin, Vicki PhD; Stetson, Barbara A. PhD; Looney, Stephen W. PhD

The Journal of Cardiovascular Nursing: July-August 2007 - Volume 22 - Issue 4 - p 263-271
doi: 10.1097/01.JCN.0000278957.98124.8a

Background: Regular physical activity is a health promotion and disease prevention behavior. Of all demographic groups, low-income women report the lowest levels of physical activity.

Research Objective: The purpose of this study was to test an intervention aimed at reducing community environmental barriers to physical activity in low-income women.

Methods: The research design was mixed methodology: (1) quantitative (quasi-experimental, pretest-posttest, cohort design in which no treatment partitioning was possible) and (2) qualitative (focus groups). The setting was a church-sponsored community center centrally located in a low-income urban neighborhood. The comparison group was recruited first followed by the intervention group to control for setting. The sample consisted of 104 women (comparison group, n = 53; intervention group, n = 51) between the ages of 18 and 63 years who were residents of neighborhoods served by the community center.

Results: No between-group differences were found for physical activity behavior. Significant between-group differences in cholesterol (P = .007) and perception of physical activity (P = .033) were observed. Significant intervention group increases from pretest to posttest were found related to advanced registered nurse practitioner support, friend support, and more positive physical activity environment at the community center. Qualitative data supported and enriched the quantitative data.

Conclusions: Physical activity levels were not significantly different between the groups. In a sample of low-income women who have multiple barriers, improving attitudes, expanding their knowledge of community resources, and providing physical activity opportunities in their neighborhoods are important intermediate steps toward initiation and maintenance of regular physical activity.

Barbara J. Speck, PhD Associate Professor, School of Nursing, University of Louisville, Louisville, Ky.

Vicki Hines-Martin, PhD Associate Professor, School of Nursing, University of Louisville, Louisville, Ky.

Barbara A. Stetson, PhD Assistant Professor, Department of Psychological and Brain Sciences, University of Louisville, Louisville, Ky.

Stephen W. Looney, PhD Professor, Department of Biostatistics, Medical College of Georgia, Augusta, Ga.

This study was funded by National Institute of Diabetes & Digestive & Kidney Diseases #RO1 DK63523. Principal Investigator: Barbara J. Speck.

Corresponding author Barbara J. Speck, PhD, School of Nursing, University of Louisville, Louisville, KY 40292 (e-mail:

Several decades of research support the significance of regular physical activity in promoting health and preventing/reducing disease risk factors for cardiovascular and other chronic diseases.1-3 According to Healthy People 2010, only 15% of the US population engages in regular physical activity. This rate varies by sex (female, 13%) and ethnicity (African American, 10%).4 Minority and low-income populations have the highest rates of cardiovascular disease and also report the lowest rates of leisure physical activity.5 Of the hundreds of studies published on physical activity, few have focused on physical activity in minority and low-income women. Descriptive studies and studies with low-income postpartum women have been reported in the literature.5,6 There are few intervention studies, although one study reported increased physical activity after a psychosocial intervention.7

The purpose of the present intervention study was to decrease environmental barriers to promote physical activity in low-income women. It was hypothesized that subjects in the intervention group would have greater improvement in level of physical activity and disease risk factors, greater improvement in total perceived benefits and barriers, increased benefits, decreased barriers, and greater increase in social support and self-efficacy than subjects in the comparison group. It was also hypothesized that the intervention would result in increased participation in physical activity opportunities from months 1 to 6. In addition, 2 research questions were posed: "What are the perceptions of physical activity in a subsample of low-income urban women?" and "What are the perceptual changes in a subsample of low-income women after a nurse-facilitated physical activity intervention?"

The conceptual model informing this study was an adaptation of the Health Promotion Model8 and included the following variables: physical activity-related situational and interpersonal influences; perceived barriers, benefits, and self-efficacy; immediate competing demands; and demographic and personal characteristics. The primary outcome variable was level of physical activity. Intermediate outcome variables included personal characteristics of risk factors (physiologic), psychosocial variables, and perception of physical activity.

Back to Top | Article Outline


The research design consisted of both quantitative and qualitative methodologies. The quantitative component was a cohort design in which no treatment partitioning was possible.9 The comparison group was recruited first, followed by the intervention group, to control for setting. The qualitative method used in this study was focus group methodology to identify perceived benefits and barriers and environmental, situational, and interpersonal influences on involvement in physical activity for the 2 groups. Focus group methodology was selected as an appropriate triangulation approach to enhance validity of the study findings.10-12 By comparing the different perspectives articulated in focus groups, perceptions and motivating factors could be examined with a degree of complexity not available with other study methods. Because the study aimed to modify behavior that depended on complex information and varying attitudes, knowledge, and past experiences, focus groups provided a tool uniquely suited to evaluation of perception and perceptual change.12,13

A church-sponsored community center (CC) that served 3 urban low-income neighborhoods was the study site. Recruitment, data collection, and the intervention took place at the CC and in the surrounding neighborhoods. The CC had a gymnasium, exercise room, weight room, and an established nurse practitioner clinic.

The convenience sample consisted of 104 women, with 53 in the comparison group and 51 in the intervention group. Inclusion criteria were as follows: (1) physically able to engage in moderate-level physical activity, (2) 18 to 64 years of age, and (3) current or past participant in 1 of the 3 food distribution programs at the CC. The 2 groups (n = 104) did not differ significantly on any pretest variables. Mean age was 39.6 years (±12.8), and average number of children per subject was 1 (1.1 ± 1.25; range, 0-4). The sample was predominantly African American and single with an income below $20,000 (see Table 1).



Back to Top | Article Outline

Variables and Their Measurement

Situational Influences measured were Options and Perceptions of the Environmental Milieu. Options were measured with 2 methods. Perceived activity options were measured via focus groups. Actual options were measured by attendance at any physical activity opportunity or event during the intervention. Because no preexisting standardized instrument was located to measure environmental milieu, focus group data were used for these assessments. In addition, a 3-item scale tapping perception of physical activity environment (availability, accessibility, atmosphere) at the CC was constructed for this study and administered to intervention group subjects at pretest and posttest. The Cronbach alpha for this scale was P = .95.

Interpersonal Influences were Family/Friends Support and Health Care Provider Support. A 6-item questionnaire measured friend and family support for exercise. Questions asked the number of times per week friends or family members encouraged, offered, and actually participated in physical activity with the subject. This format is similar to that used by Sallis et al.14 In a study of working women,15 the friend and family scale included 4 of the 6 questions (Cronbach alpha was .89 for the friend questions and .62 for the family questions). The Cronbach alpha for the present study was .78 for the total scale, .67 for the family scale, and .70 for the friend scale. A second questionnaire was administered to the intervention group using the same questions but addressing friends participating in the intervention group. Cronbach alpha for this scale was .78. The Influence of Advanced Registered Nurse Practitioner (ARNP), a healthcare provider support instrument was a 6-item questionnaire developed for this study to measure ARNP influence. The items asked about ARNP influence on subject's level of physical activity. Some items were adapted from the friends and family scale and other items were related to educative and supportive functions of the ARNP. Five additional items were added before administration based on focus group data from the comparison group. The items were primarily nurse encouragement and support. Cronbach alpha for the 11-item scale was. 98.

Perceived Barriers and Benefits were measured by the reliable and valid 43-item Exercise Benefits/Barriers Scale,16 a measure of perception of exercise with a higher score representing more positive perception of exercise.16 For this study, Cronbach alpha was .87 for the total scale, .94 for the benefits scale, and .85 for the barriers scale.

Perceived Self-efficacy was measured by a 5-item self-efficacy for exercise scale. The items asked how confident subjects were that they could exercise in 5 situations using a 7-point Likert scale for responses (Cronbach alpha for the scale was .82 [n = 917]). Test-retest reliability of the scale was .90 during a 2-week period (n = 20).17 For the present study, Cronbach alpha was .82.

The Immediate Competing Demand assessed in this study was lack of childcare during physical activity opportunities. Childcare was provided at the CC for each session. The number of children attending childcare at each physical activity session was collected, and rates were calculated.

Demographic variables included age, ethnicity/race, marital status, number and ages of children in the home, education, and household income. All variables were collected by self-report at baseline.

Back to Top | Article Outline

Primary Outcome, Health Promoting Behavior: Physical Activity

The Yamax SW-701 Pedometer was used to measure level of physical activity for 6 consecutive days at 2 points for both groups in the study (4 and 23 weeks). Comparison studies of 5 electronic pedometers determined that the Yamax was the most accurate in recorded number of steps taken (distance), was the most consistent between units, and was the most accurate at moderate activity levels.18,19

Level of Physical Activity was measured by the 7-day Physical Activity Recall (PAR).20 Subjects identify how many hours they slept and engaged in moderate, hard, and very hard activities during the past 7 days (separately for weekdays and weekends); light activities are calculated from these data. The PAR is frequently used to measure physical activity and has good reliability and validity.21 Although the PAR was designed for administration by interview,20 the original scale included instructions for a self-administered version22 and has been used in previous research studies.23,24 The self-administered PAR was used in the present study.

Back to Top | Article Outline

Personal Characteristics (Risk Factors)

Height was measured with a statiometer to the nearest 0.1 cm. Weight was measured to the nearest 0.1 kg with a balance beam scale. Body mass index was calculated by dividing weight by height squared (kg/m2). Waist circumference was measured with a Gulick II tape. Measurements were taken at the narrowest part of the body.25 Blood pressure was taken with a mercury sphygmomanometer; both regular and large cuff sizes were available for accurate measurements. The procedure followed the American Heart Association guidelines,26 with the subject seated for 5 minutes and 3 readings taken 2 minutes apart. The mean values of the systolic and of the diastolic readings from the second and third blood pressure readings were used in the analyses. Total cholesterol, triglycerides, high-density lipoprotein (HDL), and glucose were measured by a fingerstick sample with the Cholestech LDX analyzer (Cholestech Corporation, Hayward, Calif). A portable analyzer was onsite and the procedure was performed by the ARNP or research assistant trained in correct procedure for use of the analyzer.

Back to Top | Article Outline

Focus Group Interview Guidelines

Focus groups were facilitated through the use of a semistructured interview guide based on past research related to physical activity and minority women and theoretical constructs from the Health Promotion Model.8 Areas for inquiry focused on environmental, situational, and interpersonal influences on involvement in physical activity, perceived benefits of physical activity, and barriers to physical activity. The interview guide consisted of 7 open-ended questions to stimulate useful trains of thought in the participants that were not anticipated.12,27 The same interview guide was used for all focus groups.

Back to Top | Article Outline


The basic intervention was developed by the investigators and tailored based on focus group data. The ARNP role focused on facilitating, educating, supporting, and promoting physical activity. Approximately 6 physical activity opportunities were offered each week that included 4 exercise opportunities at the CC (exercise room, weight room, low-level aerobic-type physical activity classes and hip-hop) and 2 neighborhood walks (1 with a community resident and 1 with the ARNP). Childcare was provided for CC activities, and women were encouraged to bring children on community walks. The ARNP spent 4 hours each week conducting telephone prompts to encourage participation, providing sessions on general health and physical activity, and leading a weekly community walk. Two special activities were a trip to a super store to walk, shop, and read labels and participation at a CC Health Fair. Three newsletters were mailed to subjects during the 6-month intervention that included physical activity information with an emphasis on incorporating physical activities in women's daily lives.

Back to Top | Article Outline


Comparison group subjects were recruited first and independently of the intervention group. Rolling enrollment was used for both groups; as women were enrolled and pretested, their 26-week study participation began. Enrollment required approximately 20 weeks for each group to reach the desired sample size. Recruitment of study participants for both groups was through 3 established food programs offered by the CC and in the community. In addition, the recruitment team canvassed the 3 neighborhoods distributing information at homes and at subsidized housing offices and attended multiple neighborhood events. Small incentive gifts with the study logo and telephone number were provided. Interested women were given an appointment time for the pretest measurements. All subjects signed an informed consent form before data collection. As part of the routine physical examination by the ARNP, subjects were screened for ability to engage in at least moderate-intensity physical activity.

Back to Top | Article Outline

Data Collection

Subjects were fasting for baseline data collection (pretest). Anthropometric measures were taken in a private room; blood pressure, cholesterol, and self-report questionnaire were completed in the nurse's office or a private room in the cafeteria. Pedometer data were collected in two 6-day periods during weeks 4 and 23 of the 26 weeks. Subjects received written and verbal instructions on use of pedometers. All data collected at baseline (except demographics) were collected again at 26 weeks (posttest) following the same procedure as baseline. In addition, for the intervention group only, 3 additional questionnaires (ARNP and friend support and CC environment) were administered at pretest and posttest.

Back to Top | Article Outline

Procedure for Focus Groups

Focus groups were conducted at 3 points during the course of the study. Because of the broad age range for eligibility (18-65 years), it was hypothesized that perceptions related to physical activity might vary by age. Holding separate sessions with homogeneous but contrasting groups was undertaken to produce information in greater depth than might have been the case with heterogeneous groups.10,27 Focus group membership throughout the study used age break characteristics, with groups broken down by ages 18 to 39 years and 40 to 64 years.

For the comparison sample, 2 focus groups were conducted at the end of the control period. For the intervention group, focus groups were conducted before the initiation of the intervention and at the end of the intervention period. Informed consent and permission to audiotape each session were obtained before focus group assignment. One coinvestigator functioned as the lead focus group facilitator. The facilitator was an African American nurse with extensive training and experience in qualitative methods and expertise in mental health and communication with underserved populations. Facilitators documented focus group discussions through note taking and the use of 2 tape recorders. Each session was allowed to proceed until discussion on the identified topic was exhausted. Audiotape discussions were transcribed verbatim by a professional transcriptionist. Transcripts were verified for accuracy before coding. Data from the focus groups functioned as the foundation for the development of the intervention. Findings regarding the perceptions of participants before and after the intervention were also obtained from these data.

Back to Top | Article Outline

Statistical Analysis

Because the change scores for the outcome variables were of primary interest, separate paired t tests were performed for each outcome within the intervention and comparison groups. The mean change scores were then compared between the intervention and comparison groups using the unequal variance t test.

The numerator of the physical activity event attendance rate was based on the total attendance at all events that were offered within a given month. The denominator of the attendance rate (ie, the total possible attendance) was calculated as the product of the number of study subjects who were eligible to attend the events times the number of possible events, accumulated over the 4 weeks in a month. A similar method was used to calculate the childcare utilization rate.

Back to Top | Article Outline


The main outcome variable, physical activity level, was measured by pedometer (mean number of steps) and PAR score (mean metabolic equivalent task [MET] score [1 MET = energy expenditure of 1 kcal · kg−1 · h−1]).28 There was no significant group difference for change from time 1 to time 2 in mean number of steps per day (P = .206; see Table 2). There was a significant decrease in number of steps per day in the comparison group from pretest to posttest (P = .014). There was a nonsignificant decrease in the number of steps per day in the intervention group from pretest to posttest (P = .306). There was no significant difference between groups in mean MET change score (P = .250; see Table 3). The intervention group showed a nonsignificant increase in mean MET score from pretest to posttest (P = .099).





Significant between-group differences were noted for mean change score for cholesterol (P = .007) and HDL (p = 0.018) with a significant pretest to posttest increase in the comparison group cholesterol (P<.001) and HDL (P < .001) and a nonsignificant increase in the intervention group cholesterol (P = .392) and HDL (P = .623). There were no significant between-group differences in mean change scores for blood pressure (either systolic [P = .308]or diastolic [P =.974]), waist measurement (P = .823), or body mass index (P = .770; see Table 4).



Significant between-group differences were found for change in perception of exercise. Subjects in the intervention group had an increased positive perception of exercise compared with the comparison group (P =.033). There were no significant between-group differences in change from time 1 to time 2 in friend support (P = .525), family support (P = .102), barriers (P = .072), benefits (P = .151), or self-efficacy (P = .205; see Table 4).

Figure 1 contains a sequence plot of the physical activity event attendance rates during the 10-month intervention period (month 1 was removed from the data because it was based on a much smaller denominator than the other months in the study). There seemed to be an increasing trend in attendance at the physical activity events near the end of the study. A similar trend in childcare utilization throughout the study was also identified. There was a moderate correlation between the physical activity event attendance rate and the childcare utilization rate during the months of the intervention (r = 0.618, P = .043).



Additional analyses with only the intervention group indicated that there were significant differences from pretest to posttest in perceived friend support from friends in the intervention group (P = .007), ARNP influence (P < .001), and CC environment (P = .049). Subjects in the intervention group viewed friend support and ARNP influence as more positive at time 2 compared with time 1.

Data collected from focus groups were analyzed for research questions 1 and 2. There were 2 focus group sessions in the comparison group and 4 in the intervention group. A total of 17 women participated in the focus groups. Seven women fell between the ages of 18 and 39 years, and 10 women fell between the ages of 40 and 63 years. Focus group size ranged from 3 to 6 participants per session.

Back to Top | Article Outline

Research Question 1

The common overall themes during the focus groups were the need for complex benefit from physical activity (physical, emotional, and social) and the need for systems of support for involvement in physical activity. Women in both age groups identified the importance of physical activity to health, maintaining functional status, accomplishing activities of daily living, and for its emotional affects. Ability to accomplish physical activity was affected by systems of support composed of compatible activity, people (social network), life concerns, and guidance/information. They also identified that there were interrelated contextual influences that functioned as barriers to physical activity. Barriers to physical activity were identified for life events, emotional states, and resources.

Women who were in the younger age group identified unique physical activity perceptions. They identified activity that required more rigorous involvement and spoke about those activities being undertaken more for the physical response to challenging the body. In response to the need for guidance, younger women were more interested in obtaining information and follow-up about "how" physical activity was to be undertaken safely and effectively. Stress was more readily identified as a barrier to physical activity. Family and work responsibilities increased with age and this change in life concerns functioned as a barrier. Although limited available resources were identified by both groups of women, younger women identified lack of locations (safety, accessibility, and affordability) to undertake high-energy activity as an important barrier.

Women in the older group also identified age-related issues. These women identified walking as the most important physical activity. There was an increased emphasis on maintaining mobility and functional levels for the purpose of employment and independence. Depression, grief, and loneliness were important barriers to physical activity. Participants in this age group viewed physical activities integrated as part of other social activities as key resources that provide support, connection, and mutual affirmation.

Data identified in response to this research question supplemented the study intervention. Changes made to the intervention before implementation included a variety of types (moderate to vigorous) and times for activities and an increased level of support from the ARNP to include social activities, educational sessions, and encouragement to participants.

Back to Top | Article Outline

Research Question 2

In addition to the information identified in previous focus groups, the following additional perspectives were shared. Participants in the intervention identified the development of a trusting and supportive collaboration with the project personnel as the first step in becoming involved. Project activities and outreach were frequently identified. Women across the age groups could identify resources available for physical activity within their neighborhood as a result of the intervention. All could identify new strategies for physical activity-the most popular ones were the pedometer for at-home use, the walking group through the community, and the activity class at the center. Women of both groups identified the connectedness that had developed over the course of the project as a resource and a source of community pride and involvement.

The exercise room at the CC was identified as an unknown resource, although it had been in existence before the beginning of the study. The women who had participated in the intervention identified the community setting as "having something for the women" and not just their children. Older women who had been a part of the comparison group expressed interest in having a role as the "pioneers" in the development of the intervention and expressed interest in the ongoing progress of the project. The benefit most commonly identified by the younger women was in relation to babysitting and the scheduling of the activities to meet their family needs. Older women identified the opportunity for sharing as the greatest benefit. Qualitative data indicated that formulation of any intervention targeting physical activity must consider the priorities, concerns, and developmental tasks of those for whom the program is designed in order to be culturally and socially acceptable. All participants in the intervention group expressed increased awareness of facilitators to physical activity and increased interest in specific activities.

Back to Top | Article Outline


The minimal environmental intervention for this study did not directly result in increased physical activity levels. Numerous changes were noted for physical activity-related perceptions, which were intermediate outcome variables that may moderate physical activity behavior. The increase in positive perception of physical activity in the intervention group is encouraging. Perceptual changes may represent forward movement in readiness for change for engaging in physical activity. Stage of change is one of the constructs of the Transtheoretical Model, which posits that behavior change moves from precontemplation to maintenance, although changes may not be linear.29 Stage of change was not measured in this study, but this finding and results of the focus groups supported a movement from precontemplation to contemplation and action. Increased positive perceptions may represent preparation, a stage that readies one for action as intervention participants identified solutions to barriers. In addition, focus group data supported the characteristics of the following stages: precontemplation stage-lack of information (about resources available in the community), contemplation-understanding the need but not ready or able to change (understanding the benefits of physical activity while struggling with barriers), preparation-understanding but lack readiness (barriers outweigh benefits), and action-behavior change that supports commitment to change (participation in activities at CC).

In the intervention group, there were statistically significant increases in friend support, ARNP influence, and CC environment from pretest to posttest. The social support aspect of ARNP influence and friends support could have contributed to the increased positive perception of physical activity in the intervention group. An additional component of the ARNP influence was to provide healthcare information that may have contributed to changes in perceptions. The more positive perception of the CC environment could also be attributed to ARNP influence as well as participation in physical activity events at the CC. Before the study, women in the community were not only unaware of facilities, few programs targeted women.

The mean cholesterol change scores were significantly different between groups at posttest. Cholesterol and HDL increased in both groups but to a greater degree in the comparison group. Whether this was related to the intervention or other variables is unknown. Data on variables that could impact cholesterol levels, such as current medications, smoking status, diet, and alcohol consumption, were not collected. Physical activity did not increase in either group and was not considered a factor. Lack of information on variables that could affect cholesterol level limits interpretation of cholesterol changes.

Use of pedometers for data collection was problematic in this study. Subjects were interested in the pedometer but had multiple problems wearing them for 6 consecutive days. Problems included losing the pedometer, dropping it with the pedometer resetting to zero, forgetting to wear it, and children playing with the pedometer resulting in loss of data. Second, the timing of data collection may have been a problem. The "baseline" data collection was during the fourth week of participation. This was not a true baseline and could have contributed to the higher step numbers at the first reading. Mean steps per day at both data collection points and groups were around 5,000 representing sedentary behavior (<5,000).30 Future studies utilizing pedometer-derived data should carefully time data collection and provide prompts and backup methods to optimize data acquisition, storage, and validity.

The second physical activity outcome measure was MET scores obtained from the 7-day PAR. In this study, the instrument was self-administered. Of the 104 pretest questionnaires, 25 (24%) were unusable (7 were deleted due to total hours greater than 24/d). Subjects may have had difficulty understanding the instructions and definitions of activity levels. As further evidence of problems with self-administration of the instrument in this sample, mean daily MET scores of 42.9 to 49.2 in this study were comparable to very active lifestyle of >40.22 In a study with low-income Women, Infants and Children (WIC) Program participants, mean MET scores averaged around 34 for women regularly engaged in physical activity.31 In a study of working women currently engaging in regular physical activity, the mean daily MET scores were 38.9 at pretest and 40.1 at posttest.24 Use of the self-administered PAR was a limitation of the study. Interview-based PAR or other self-report activity measures may have yielded more valid, reliable findings.

Study findings suggested that for low-income women, a multimethod approach to increasing activity is critical. Providing a safe inviting environment for physical activity that is easily accessible is a key component but does not guarantee success. Site-based social support is essential. Providing programs offering even more structure may be more beneficial. Findings suggested that childcare was not used as much by participants as projected based on focus group data. Nonetheless, childcare should be considered an important component of physical activity programs for low-income women because it is a common barrier.

One outcome of this research was that its positive influence on women in the low-income community resulted in local funding to continue the physical activity sessions for at least another year. Long-term tracking of environmental changes in activity behavior and perception may provide useful information regarding the trends observed in the present study. In the face of numerous barriers to activity, low-income, urban women may exhibit slow but continuous progression in activity behavior change. Alternatively, in the absence of intensive project personnel contact, perceptions and intermediate gains may return to baseline. Study findings from focus groups and responses to psychosocial questionnaires highlight the substantial barriers low-income women must overcome to initiate health promoting behaviors. The progression from preparation to action to maintenance of physical activity may evolve more slowly than with other more advantaged populations. Future studies of environmental manipulation to promote physical activity in low-income women would benefit from traditional randomized controlled designs, improved methods of physical activity measurement, and long-term assessment of both perceptions and activity behaviors.

Back to Top | Article Outline


The authors wish to thank Robyn Kuhn, who assisted with the statistical analyses.

Back to Top | Article Outline


1. Borer KT. Physical activity in the prevention and amelioration of osteoporosis in women. Sports Med. 2005;35:779-830.
2. Hu G, Tuomilehto J, Silventoinen K, Barengo N, Jousilahti P. Joint effects of physical activity, body mass index, waist circumference and waist-to-hip ratio with the risk of cardiovascular disease among middle-aged Finnish men and women. Eur Heart J. 2004;25:2212-2219.
3. Manson JE, Hu FB, Rich-Edwards JW, et al. A prospective study of walking as compared to vigorous exercise in the prevention of coronary heart disease in women. N Engl J Med. 1999;341:650-658.
4. United States Department of Health and Human Services (USDHHS). Healthy People 2010 (Conference Edition, in Two Volumes). Washington, DC; 2000.
5. Eyler AA, Matson-Koffman D, Vest JR, et al. Environmental, policy, and cultural factors related to physical activity in a diverse sample of women: The Women's Cardiovascular Health Network Project-summary and discussion. Women Health. 2002;36:123-134.
6. Walker LO, Greeland-Graves JH, Milani T, et al. Weight and behavioral and psychosocial factors among ethnically diverse, low-income women after childbirth: Methods and context. Women Health. 2004;40:1-17.
7. Albright CL, Pruitt L, Castro C, Gonzales A, Woo S, King AC. Modifying physical activity in a multiethnic sample of low-income women: one-year results from the IMPACT Project. Ann Behav Med. 2005;191-200.
8. Pender NJ, Murdaugh CL, Parsons ML. Health promotion in nursing practice. 4th ed. Upper Saddle River, NJ: Prentice Hall; 2002.
9. Cook TD, Campbell DT. Quasi-Experimentation: Design & Analysis Issues for Field Settings. Boston, Mass: Houghton Mifflin; 1979.
10. Krueger RA, Casey ME. Focus Groups: A Practical Guide for Applied Research. 3rd ed. Thousand Oaks, Calif: Sage; 2000.
11. Lincoln YS, Guba EG. Naturalistic Inquiry. Newbury Park, Calif: Sage; 1985.
12. Morgan D. Planning Focus Groups. Thousand Oaks, Calif: Sage; 1997.
13. Kamberelis G, Dimitriadis G. Focus groups: strategic articulations of pedagogy, politics and inquiry. In: Denizin NK, Guba YS, eds. The Sage Book of Qualitative Research. 3rd ed. Thousand Oaks, Calif: Sage; 2005.
14. Sallis J, Hovell MF, Hoffstetter C, Barrington E. Explanation of vigorous physical activity during two years using social learning variables. Soc Sci Med. 1992;34:25-32.
15. Speck BJ. Maintenance of regular physical activity in working women. In: Funk SG, Tornquist EM, Leeman J, Miles MS, Harrell JS, eds. Key Aspects of Preventing and Managing Chronic Illness. New York: Springer; 2001:95-108.
16. Sechrist K, Walker SN, Pender NJ. Development and psychometric evaluation of the exercise benefits/barriers scale. Res Nurs Health. 1987;10:357-365.
17. Marcus BH, Selby VC, Niaura RS, Rossi JS. Self-efficacy and the stages of exercise behavior change. RQES. 1992;63:60-66.
18. Bassett DR, Ainsworth BE, Leggett SR, et al. Accuracy of five electronic pedometers for measuring distance walked. Med Sci Sports Exerc. 1996;28:1071-1077.
19. Bassett DR, Ainsworth BE, Swartz AM, Strath SJ, O'Brien WL, King GA. Validity of four motion sensors in measuring moderate intensity physical activity. Med Sci Sports Exerc. 2000;32:S471-S480.
20. Sallis JF, Haskell WL, Wood PD, Fortmann SP, Rogers T, Blair SN, Paffenbarger RS. Physical activity assessment methods in 5-city project. Am J Epidemiol. 1985;121:91-106.
21. Sallis JF. Seven day physical activity recall. Med Sci Sports Exerc. 1997;29(Suppl 6):91-103.
22. Blair SN. How to assess exercise habits and physical fitness. In: Matarazzo JD, Weiss SM, Herd JA, Miller NE, eds. Behavioral Health: A Handbook of Health Enhancement and Disease Prevention. New York: Wiley; 1984:427-447.
23. Miller DJ, Freedson PS, Kline GM. Comparison of activity levels using the CaltracR accelerometer and 5 questionnaires. Med Sci Sports Exerc. 1994;26:376-382.
24. Speck BJ, Looney SW. Effects of a minimal intervention to increase physical activity in women: daily activity records. Nurs Res. 2001;50:374-378.
25. Callaway CW, Chumlea WC, Bouchard C, et al. Circumferences. In: Lohman TG, Roche AF, Martorell R, eds. Anthropometric Standardization Reference Manual. Champaign, Ill: Human Kinetics; 1988:39-54.
26. United States Department of Health and Human Services (USDHHS). Fifth Report of the Joint Commission on Detection, Evaluation, and Treatment of High Blood Pressure. Bethesda, Md: National Institutes of Health. USDHHS Pub No. [NIH] 93-1088; 1993.
27. Knodel J. The design and analysis of focus group studies: a practical approach. In: Morgan D, ed. Successful Focus Groups: Advancing the State of the Art. Newbury Park, Calif: Sage; 1993:35-50.
28. Waden J, Tikkanen H, Forsblom C, et al. Leisure time physical activity is associated with poor glycemic control in type 1 diabetic women. Diabetes Care. 2005;28:777-782.
29. Prochaska JO, Redding CA, Evers KE. The Transtheoretical Model and stages of change. In: Glanz K, Rimer BK, Lewis FM, eds. Health Behavior and Health Education: Theory Research and Practice. 3rd ed. San Francisco, Calif: Jossey-Bass; 2002:99-120.
30. Tudor-Locke C, Bassett D. How many step/day are enough? Sports Med. 2004;34:1-8.
31. Fahrenwald SL, Walker SN. Application of the Transtheoretical Model of Behavior Change in the physical activity levels of WIC mothers. Public Health Nurs. 2003;20:307-317.

intervention; low-income women; physical activity

© 2007 Lippincott Williams & Wilkins, Inc.