Promoting early breast cancer screening: strategies with rural African American women.

Abstract: Despite the documentation of high death rates from breast cancer among African American women, there is limited information published on elective interventions to increase early breast cancer screening among this population. Research examining the efficacy of health promotion appeals about breast cancer, content, and channels of delivery to this population has also been very limited in public health research. The Morehouse School of Medicine is conducting the Breast Health Belief Systems Study. This study is designed to test the saliency of integrating belief systems of the study population into the design of an intervention to increase early breast cancer screening practices. It targets rural African American women living in rural South Georgia. Women indigenous to study sites were trained as lay health workers to deliver the intervention. Conducted in three phases, the study is currently in the phase three, the intervention. While it is too early to draw conclusions, this article describes the developmental strategies of the study and presents some insight into common issues considered important in the implementation of a target-specific, breast cancer prevention and screening program.

**********

Death rates are far too high from diseases that are preventable, or if contracted, have a high probability of being cured if detected during early stages of development. Breast cancer is one of those diseases (NCHS, 1996; Horton, 1992). It is the most common cancer in American women: one in every 10 women will develop breast cancer sometime in her lifetime (Miller et al., 1996). While research findings and professional organizations support regular breast examination and mammography as effective screening methods for early stage breast cancer (CDC, 1994; Miller et al., 1996), surveys indicated that 44 percent of white women, 51 percent of Hispanic, 52 percent African American, and 54 percent of Asian American reported not having a mammogram according to American Cancer Society guidelines for their age group. The highest death rate from breast cancer was reported among African American women (National Women's Health Network, 1996). This higher mortality rate is thought to be due largely to late stage diagnosis. While reasons are suggested for this phenomenon, e.g., lack of knowledge, failure of health provider to initiate screening procedures, costs, culturally inappropriate approaches (Farley & Flanery, 1989; Caplin, Wells, & Haynes, 1992), empirical research is limited.

Despite the documentation of the higher morbidity and mortality of breast cancer among African American women, there is limited information published on effective interventions to increase breast cancer screening among this population. Black women can be members of heterogeneous cultural groups, many with mixed ancestry including individuals with Caribbean, Indian, and European lineage. Marked differences in acculturation exist and contribute to the diversity of their health. Therefore generalizations that create health profiles for these women are dangerous because exceptions to the rules are numerous (Nyamathi et al., 1993). However, relatively little research has been devoted to identifying effective strategies for increasing breast cancer screening rates among African American women of varying cultural and intra-cultural backgrounds. Social scientists and health communication researchers have held that if health promotion campaigns are to influence the audience as intended, they must be culturally, demographically, and geographically appropriate (Williams, 1996; Atkins & Freimuth, 1989; McLeroy et al., 1988). Research defining the variable of cultural sensitivity is also very limited regarding interventions that target diverse populations.

Research examining the efficacy of health promotion appeals, content, and channels of delivery regarding breast cancer has also been very limited in public health research. A number of factors frustrate the formulation of effective campaigns to promote healthy behaviors and practices. First, health promotion campaigns usually exhort people to change deeply rooted beliefs and behaviors that have been continually reinforced over a lifetime, and perhaps throughout preceding generations. Secondly, health promotion information and recommendations are traditionally based upon epidemiologic findings that do not often include a broad, in-depth assessment of culturally driven behaviors that are especially prevalent in multi-ethnic and multi-cultural societies. Further, virtually all health promotion message approaches are based on cognitive/behavioral theories of communication that assume that all people view and interpret situations similarly (Fishbein & Ajzen, 1981; Kreps, 1988; Greenberg et al., 1988). However, health and disease may mean different things to different people. McLeroy and colleagues have suggested extension of the traditional cognitive framework to include interpersonal processes, institutional factors, community factors, and public policy issues as well.

Across the United States, health promotion interventions are using strategies involving community health worker or lay advisor (helping healers) programs to reach traditionally underserved populations (Parker et al., 1998; Blumenthal et al., 1989). The difference between these programs can be explained on a continuum of formal to informal helping (Eng, Parker & Harlan, 1997). The formal end of the continuum consists of the paraprofessional/outreach worker who is a paid employee of some community-based agency or organization. The informal end of the continuum is that community person who always seems to be there volunteering to help individuals and community efforts. Both of these strategies involve identifying individuals, within existing social networks of the targeted audience, who other network members trust and regularly turn for social support. While several studies report the effectiveness of the informal model of this type worker to deliver breast health messages, particularly in rural areas, few reports evaluate the impact of utilizing the formal model for such activities. In an effort to expand the knowledge base on health promotion and breast cancer and related delivery systems, the Morehouse School of Medicine is conducting this Breast Health Belief Systems Study (BHBS).

PROJECT BACKGROUND AND CONCEPTUAL FRAMEWORK

The BHBS is one of several community-based projects in the overall health promotion and disease prevention program at the Morehouse School of Medicine (MSM) in Atlanta, Georgia. Administratively, the study is housed in the Department of Community Health and Preventive Medicine, which historically has functioned as the bridge between the school and the community. Our Breast Health Belief Systems Study was designed to address the saliency of beliefs about breast cancer for constructing effective health promotion messages for African American women. It assumes that the first step for designing meaningful communication about breast cancer preventive and help-seeking behavior is to explore a group's belief system regarding the disease. Beliefs are defined as the basic units of thought that establish a relationship between at least two entities and are culturally driven (Kreps & Thornton, 1984). For the purpose of this study, culture is defined as a set of interlocking cognitive schemata that construct and give meaning to what people do in their everyday lives. These schemata are transformed within specific social environments, and are constrained by the economic and political context of a specific group. Kreps and Thornton also offer organizing principles that can affect meaningful health communication: world view; socialization; and interactions with social organizations such as such as schools, churches, law enforcement, mass media, and the health care system. Empirical as well as historical evidence indicates that religion is a significant factor in the lives of African Americans, particularly women. Further, religion plays a central role in African American women's efforts to cope with a wide array of problems including illnesses (National Survey of Black Americans; Levin & Taylor, 1998; Neighbors et al., 1983; Chatters & Taylor, 1989; Taylor et al., 1999; McAdoo, 1995) and they maintain a deeply rooted faith in God and in the fundamental values of particular faith traditions (Mattis, 2001). Jahn (1961) explains that from an Afrocentric perspective, it is not possible to separate theology from medicine or vice versa without violating the entire worldview of the group. Belief structures, composed of values and attitudes, are considered cultural knowledge (Quinn & Hollard, 1987; Ashante, 1988). Understanding and integrating cultural sensitivity into the design of a program personalizes the message, thereby increasing the likelihood that the target audience will understand and accept the recommendations. Since beliefs about breast cancer are a product of both personal experiences and information obtained through interactions with others (either interpersonally or through media), not all members of a group will have identical sets of cultural knowledge. Delineating variations in beliefs is an important design element to more effective intervention strategies. For example, Matthews et al. (1994). found that African American women in rural North Carolina draw on multiple sources of knowledge in order to come to terms with their breast cancer.

Building on these literature findings and our own research experiences, an initial conceptual framework was developed, which postulates that a breast health approach that proceeds from and responds to specific belief systems among low-socioeconomic, African American women, will motivate increased compliance to recommended cancer screening schedules, and positive shifts in knowledge and attitudes about the disease.

METHOD/APPROACH

The expected outcome of this study is the development of a model community-based intervention designed to educate and motivate rural African American women to adopt good breast health screening practices. The study consists of three Phases: One -- An ethnographic survey of women who have or have had breast cancer to determine intra-cultural variations in the variety, scope, and depth of beliefs and practices (before and after diagnosis/treatment) regarding breast cancer; Two -- Use of the analysis of data gathered from Phase One to develop an interpersonal, target-specific breast health promotion program for women with no known breast cancer and who have not had a mammogram in the twelve months (or 24 for those 49 and younger) preceding participation in the study; and, finally, Phase Three -- implementing and measuring the effectiveness of the breast health intervention and its delivery strategies. Disciplines involved in this study include faculty and students from gerontology, health communications, anthropology, sociology, public health, home economics, theology, and nursing. Collaborating institutions and agencies are Georgia State University, Fort Valley State University, Albany State University, Savannah State University and the Older Americans Council of Middle Georgia. Phases One and Two have been completed. Phase Three, is nearing completion. The following is a description of the conceptualization, development, implementation and findings to date from this study.

Our Breast Health Belief Systems Study involves researchers, community members, and practitioners in a joint process aimed at meeting both research and intervention objections. This process bares similarities to the Participatory Action Research approach (PAR) (Israel et al., 1994) and the Braithwaite-Lythcott (1989) model of community organization for health promotion. Both allow the insiders view of life to be considered in health education program planning. Adapting elements from these approaches, this project involves the use of two research components: a single case study design in Phase One (ethnographic survey) which allows for an empirical investigation of a contemporary phenomenon (under utilization of breast cancer screening methods by African American women) within its real-life context and from the perspective of the effected group; and a quasi-experimental design in Phase Three (the intervention) utilizing qualitative and quantitative methods.

Establishing the Research Team: Senior or graduate students from Historically Black Colleges and Universities near our study sites were trained and utilized as research assistants. These research assistants conducted the Phase One in-home interview of breast cancer survivors. Major criteria for selection of the research assistants were that they live in or near one of the study communities, have a history of regular involvement in local community-based events, and a demonstrated interest in community welfare. A faculty preceptor from each school assumed the responsibility for selecting students and the on-site supervision of the student research assistants. These criteria provided a number of advantages: cultural and geographic familiarity, inter-institutional collaboration, involvement of health-related students in community-based health promotion (Hatch & Lovelace, 1980), and cost effectiveness in terms of controlling travel expenses in connection with the ethnographic analysis. Both the informal and formal models of lay health workers, as described previously in this article, were part of our research team. Informal lay health workers, trained on how to recruit study participants, assisted in Phases One and Two with pilot-testing the instruments and recruitment of study participants. Formal lay health workers assisted in Phase Three with study participant recruitment and delivery of the educational intervention.

Unique to this research team is the Cooperative Extension Service. Though better known for its agricultural focus, the extension service also includes community development programs including health education. One of our HBCU partners, Fort Valley State University Extension Program, operates in 13 rural counties in Georgia and is a longstanding trusted institution among the black community. The Fort Valley Program employs paraprofessionals called Cooperative Extension Program Assistants (CEPA) who are usually indigenous to the community in which they work. Supervised by home economist, their responsibilities include in-home economic and health education activities. Cooperative Extension Program Assistants functioned as formal lay health workers in Phases Two and Three of this study.

History and Context of Rural Georgia: The BHBS study takes place in middle and southern Georgia in thirteen counties in Phase One and three (3) counties in Phase Three. Most of these counties generally rank lower than their urban counterparts on conventionally used, measurable indices of quality of life, i.e., poverty, low population density, large geographic expanses, lack of human service-related resources and negative capital flow to urban areas. Findings from a previous MSM breast cancer study (Blumenthal et al., 1989) in Georgia suggest that far too many African Americans do not participate in early detection for breast cancer, even in areas where tests are low or no costs. This study found that only 30 percent of inner city and 20 percent of rural women were likely to receive screening, according to guidelines published by the American Cancer Society. Thirty-six percent of those surveyed did not feel the urgency to get mammogram or pap smears on a regular basis.

Our research objectives dictate that these counties fall within at least one of three categories: physically isolated area of extreme poverty, a rural area that provides access to a metropolitan center, and a poor coastal area that feature a wide range of intra-cultural variations including one with potential Gullah (practice of traditional West African folkways) influences. In Phase One the thirteen counties are reported in geographic groupings or public health districts: Albany for the southwest, Macon for middle Georgia, and Savannah for the southeastern and coastal regions. The African American population in these counties is 40 percent or greater. All three of these areas contain counties, which fall into at least one of our site categories. The Savannah district is the site where more small pockets of residents who adhere to some practices of West African folkways. Because of demographic characteristics, it was assumed that these three geographical sites would provide variation in the samples for exploring the study's assumptions.

PHASE ONE: ETHNOGRAPHIC SURVEY

Seventy-five women with a current or past diagnosis of breast cancer were targeted for the ethnographic survey. The rationale for interviewing women with a history of breast cancer was to make a pre and post diagnosis comparison of beliefs, knowledge, and practices, then work backwards to use these findings for the development of our intervention study. Three assumptions guided the selection of research sites and the collection of ethnographic data. They were: women from isolated rural communities are less likely to seek preventive biomedical care; women from rural communities with access to complex urban systems are more likely to seek preventive biomedical care; and cultural beliefs about preventive care for breast cancer and for health problems in general are stronger persuasive agents than access to, or affordability of, medical care services.

The methods used in our study were in-home, face-to-face interviewing using an interview schedule; free listing to delineate cultural domains and pile sorting to delineate shared cultural categories for breast cancer. These techniques also allowed us to understand how breast cancer beliefs are organized in the mind of respondents and to provide a baseline for discovering variation in cultural knowledge in the three research sites.

Data collected were analyzed via focused coding methods that utilize the constant comparison approach to develop relevant themes (Patton, 1990; Straus & Corbin, 1990). The interview schedule questions were designed to elicit descriptive information for determining the Explanatory Models (EMs) and decision-making processes for breast health disorders. Collected in narrative form, EMs link beliefs and behavior within the context of people's experience the words of the respondent. They provide a person having breast cancer with a causal explanation for the disease. During the in-home face-to-face interview, the respondents were asked about their pre and post diagnosis beliefs about breast cancer and related problems. Each interview was audiotaped, transcribed and compared to the written narrative to verify that the participants' perspective was accurately captured. Each narrative contained the following variables: causes of breast cancer; symptoms of breast cancer; alternative treatment, definitions of breast cancer; expected outcomes, and attitudes toward breast cancer providers.

From these data an explanatory model was constructed for each woman in the sample. They were subsequently aggregated for each research site. Decision modeling focused on discovering and testing individual's criteria for making treatment choices. Shared standards for decision making about health behavior were used to construct a model that predicted the different treatment options that people can be expected to choose. People's knowledge structures and conceptions of illnesses are inextricably related to their illness responses, and consequently, to their decision making processes about treatment alternatives. The specific data that was systematically collected to construct decision models include the following variables: constraints of decisions about breast cancer treatment choice, intuitive rules for making decisions, rationale for making choices, and sequence of choices of treatment.

Study Participant Recruitment: Because cancer databases for the state of Georgia were incomplete for several of the counties our study targeted and the challenges encountered in gaining access to existing files on demographically similar counties, we use a grass roots effort to develop our own registry of African American breast cancer survivors. Utilizing existing relationships between Morehouse, local medical facilities, health departments, professional associations, and community-based agencies/ organizations, representatives were contacted for assistance with publicizing our study and recruiting participants. Additionally, flyers were disseminated at medical facilities, churches, community centers, senior centers, grocery stores and other common gathering places for African American females. Cancer support groups, home health agencies, and medical facilities, particularly cancer treatment centers, were the most helpful in this regard. While this route proved to be more laborious than accessing existing registries, the effort was worth the time and energy, because it provided more opportunities for direct community involvement and proved to be an excellent learning experience for staff and students.

The targeted sample number was 75 (25 from each site). The following criteria were used for selecting respondents: Low SES ($1.7,000 yearly household income or less); 40 years old and older, and lived in rural areas for past 15 years. Once identified, our lay health workers recruited participants into this study. The lay health workers administered the personal information questionnaire and referred each respondent to the research assistants who obtained the informed consent and conducted the interviewed. Upon completion of the interview, participants received a cash incentive.

Phase One Results: A total of 64 respondents were interviewed. Table 1 summarizes the demographic profiles for each site.

Ethnographic findings only partially supported the assumptions of our ethnographic survey. For the first assumption, the findings indicate that women who live in more isolated sites sought preventive care only slightly less often than those women at the least isolated sites. Women who have more physical access to complex urban systems were no more likely to seek preventive medical care than the women in the other two study sites. The ethnographic findings, however, upheld the third assumption that cultural knowledge about preventive care is strong persuasive agents for seeking medical care services for breast cancer. While costs and transportation were problematic for some respondents, particularly those who did not drive, these factors were not as persuasive as cultural beliefs.

Pre-Diagnosis Beliefs: Ninety,five percent of participants did not think about the disease; 65 percent did not discuss it with anyone; 50 percent believed they had no symptoms. The indications are that few symptoms appeared in this group or symptoms were not recognized, both significant issues to explore in future research. Some did not seek help immediately after finding a lump. Seventy-five percent had regular doctors to whom they made regular visits. While most doctors' offices displayed information about breast cancer and other diseases, participants reported receiving little counseling on the importance of regular screening. Those participants who participated less often in regular breast cancer screening were most often those who were: under 45 years of age, regardless of family history, no regular doctor, had less than a high school education, and those over 65 regardless of whether they had a regular doctor or not.

Post-Diagnosis Beliefs: At the point of diagnosis and subsequent medical treatment, the women's beliefs changed to: use of medical terms learned from their experiences with breast cancer for their symptoms, causes and treatments. However, the distinction between diagnostic procedures and treatment was not clearly understood. Nonetheless, this medicalization process results in an increased use of preventive care and an increased level of communication about the disease. Terms and concepts learned in their diagnostic and treatment experiences were framed within a religious context. More than 90 percent believe that faith helped them survive the disease. Belief that God works through doctors increased trust in the medical system and demonstrated a truly integrative belief system between religion and medicine. This finding is consistent with studies previously cited in this article regarding the role of religion in the lives of African American women. All women believed that God works through the medical providers; medical providers who demonstrated a spiritual component to healing were most trusted and more likely to receive compliance from the patient. These findings verify that the medical setting is a critical context for increasing health promotional messages and that breast-health messages personalized for African American women need to be transmitted to medical personnel.

Site Variations: To date, there appears to be more similarities than significant variations between the sites. The variations in cultural knowledge and behaviors appear greater between the Savannah sites and the other two, Albany and Macon. However, it must be noted that some of this variation may result from the smaller sample size in the Savannah sites. For example, in response to the question "Who did you first talk to about your breast cancer?": Albany Health District: sisters, daughters, doctors, husbands, female friends; Savannah: older female relatives, doctors, husbands; Macon: husbands, doctors, other family. Friends, ministers, and co-workers were the least mentioned first-talk choices across sites. This finding is consistent with Dressler's studies of rural African Americans, which stressed the extended kin network more than friends as protective against depression. Findings from the Save-our-Sisters project in rural North Carolina, show that support from female peers has been a significant predictor of accepting recommendations, following through with scheduling, and keeping appointments with mammography, even after controlling for physical health status, age, education type of insurance and source of care (Eng, Parker & Harlan, 1997). While all of the study participants wanted others to pray in church for their recovery, the Savannah site participants were more likely to prefer private prayer in their homes. Additionally, as seen in Figures 1, participants receive information about breast cancer from a variety of sources. Figure 2, however, illustrates the most trusted sources and here some notable differences between sites are seen. These between site variations suggest possible differences in intervention strategies.

[FIGURE 1-2 OMITTED]

Our findings also suggest that the quality of advice received either promotes visiting a doctor or trying other treatments first, including the length of time she delays visiting a physician. Social influences through informal networks are pertinent for breast cancer screening. In summary, tentative decision models for help seeking behavior were constructed from the ethnographic data (Figure 3). Criteria were selected based on the findings of the explanatory modeling and the cultural domains for symptoms, causes, and treatment.

[FIGURE 3 OMITTED]

The decision tree generated from the overall sample suggests the criteria affecting the decision-making process of African American women to go to the doctor and reduce diagnosis delay. This model is viewed as a hypothesis only for seeking medical care. This decision tree for going to a doctor has not been yet tested on a second and third sample and is, therefore, not a predictive model of the help-seeking process of African American women regarding breast cancer. It does, however, illustrate the preliminary criteria involved in going to a doctor for this sample.

PHASE TWO: FORMULATION OF THE INTERPERSONAL MESSAGE AND DIFFUSION STRATEGIES.

Methods: Several emerging communication models, e.g., Witte's (1995) Persuasive Health Message (PHM) framework, have expanded the purely cognitive/behavioral approach to health promotion by integrating cultural, environmental, knowledge, attitude, and belief variables into message development. Utilizing these models and data from population profiles for each of our study sites and common themes that emerged from our ethnographic survey in Phase One, we developed a breast health message that is uniquely personalized to our study population. The content for the health messages was taken from publications of the National Cancer Institute and the American Cancer Society and adapted, as needed, to an eighth-grade reading level. Messages included basic information about breast cancer such as statistics, risk factors for low-income and African American women, the diet/nutrition/stress information, the importance of early detection, screening guidelines and procedures, and information resources. The presentation of these messages is described below in Figure 4, the Intervention.

A diffusion strategy deals with how information is communicated through certain channels over time among members of a social system. Our strategy utilizes elements from the Community-based Diffusion Model (1987) and the Braithwaite-Lythcott Model (1989) for community organization for health promotion. Both models recommend an understanding of the role, patterns of use, and the impact of the local, mainstream health care systems such as private physicians, neighborhood clinics, hospital emergency rooms, and other health care resources. For example, since one of our study sites included counties situated along Georgia's Sea Island coast, we anticipated that some respondents may practice folkways handed down from their west African ancestors and may obtain advice from complimentary health providers such as herbalists and folk healers. "For members of a target community who continue to hold traditional health beliefs, it will be more effective to fit new health information into the old frame of reference rather than to dismiss traditional beliefs as ineffective or superstitious" (Ashante, 1988).

Important to the diffusion strategy is who delivers the message. The messengers for our intervention are lay health workers selected because of their position of trust and support giving within the social networks of the study population. Both the informal and formal models of lay health workers are used in this phase of our study. Our formal model utilizes Cooperative Extension Program Assistants (CEPA) who are paid employees of the Cooperative Extension Service. These individuals have been trained to work directly with county extension personnel in assigned programs of nutrition, agriculture, family services, youth development, and health promotion. A CEPA provides clients with direct assistance through home visits and/or other gatherings. Most have a high school education supplemented by in-service education in the fields of home economics, agriculture, or food service and nutrition. Our informal lay health workers are volunteers in a community center.

Lay health workers were trained by the study's principal investigator (a health educator and clinician), a co-investigator (a health communications specialist) and a registered nurse. The training curriculum included: providing a clear understanding of why the study was being done, discussions on each of the topics in the health message, demonstration on how to perform the breast self examination, preparation of the participant prospect list, selecting study participants, confidentially, data collection, recording keeping, making referrals, following research protocols, and delivery of the intervention. Nine talking points are highlighted as a guide for the information that each lay health worker must cover during the intervention. However, the lay health worker has the liberty to adapt her presentation of the information to the communication type most comfortable to the participant. Both the message and delivery were pilot-tested among demographically similar women living in rural communities not targeted by the proposed research. Role-plays were used to provide opportunities for hands-on learning and evaluation of the trainee's ability to implement the intervention. Technical support and remedial training occurs on an as-needed basis throughout the implementation.

Intervention Development/Description: A breast health information and promotion intervention has been developed that is uniquely personalized for our study population. The presentation of this intervention is illustrated below:

Figure 4:The Intervention Description (Phase Three)

1. Prepare prospect list 2. Recruit study participants 3. Assign ID numbers

Group A (Intervention): 4. Contact prospect, explain study and set-up in-home interview; 5. Conduct intervention: obtain signed consent, administer Survey 1; conduct conversation about breast cancer; review brochures allowing participant time to read and ask questions; demonstrate breast self examination; administer Survey 2; give and explain response card and referral material for mammography sites and information hot-lines; give cash incentive; 6. Conduct 30-day follow-up: Has participant obtain mammogram? Yes I" intervention ends; No but has appointment b encourage and inquire about any transportation or other access problems; No and has no appointment [THORN] repeat step 5 including all instruments except the PIQ repeat step 6: Has participant obtained a mammogram? Yes [THORN] intervention ends; No but has appointment [THORN] encourage and inquire about transportation needs Repeat Step 6; No and has no appointment [THORN] inquire about any concerns participants has about mammography and interest in continuing in the study - if no interest, study ends - interested set-up third and final visit, encourage participant to invite family and friends Repeat Step 5 [THORN] Repeat Step 6, the final follow-up. Step 7 - conduct random telephone survey of participants who do not get mammograms to assess intervention process and intentions regarding breast cancer screening.

Group B (the Control): 4. Contact prospects and set-up group session; 5. Conduct session: consisting of administering the PIQ and Survey 1 distribute breast cancer brochures allowing reading time administer Survey 2 distribute referral packages thank participants and give incentive.

Six women indigenous to the population have been trained as lay health workers to recruit study participants and to deliver the intervention. Utilizing this approach has several advantages: it piggybacks on a well-established community-based agency with ongoing, structured relationships with the targeted communities. Most importantly, it leaves the community with members who have more skills and knowledge about breast cancer and a implementing health promotion campaign to promote early screening practices.

PHASE THREE: DELIVERY OF THE INTERVENTION.

STUDY PARTICIPANT DESCRIPTION AND SELECTION.

Intervention sites include one county in each of the study sites, which meets the aforementioned study site selection criteria. The study sample consisted of two groups, A and B. Criteria for selection as a member of the study includes the following: African America female aged 40-65; current resident and native of target community; no history of breast cancer; no history of breast surgery; no history of mammogram in the 12 months preceding the study (women 50 and older) and 24 months (women less than 50). Group A receives the intervention. Group B does not receive the intervention until the end of the study after follow-up and therefore constitute the control group. Given that the overall goal of this intervention is to increase the number of African American women who participate in screening for breast cancer, we hypothesize from a quantitative perspective, that at follow-up: 1) Knowledge of breast cancer risks, prevention, and detection will increase among women in Group A (the intervention) from a baseline by 30 percent and by 10 percent among women in Group B; 2) The percent of women who have had a clinical breast examination within the last year will increase by 40 percent in Group A and by 20 in Group B; 3) The percentage of women who have had a mammogram within the past year will increase by at least 30 percent in Group A and by 15 percent in Group B. Research instruments for this phase of the study include: personal information questionnaire includes demographics and questions about breast cancer knowledge, health behaviors and practices including herbal and other folk remedies; color-coded response to track mammogram obtainment by county and group assignment; list of talking points on the information each participant in Group A must receive about breast cancer. Surveys One (pre-intervention) and Two (post-intervention): In order to detect and quantify the effects of the intervention in terms of the degree of change, short, 5-6 item pre-/post surveys are being orally administered to both groups. Survey questions seek to capturer beliefs about seriousness of the disease, personal risks, prevention capability, efficacy of breast self-exams and mammograms, personal intent towards screening practices. The follow-up survey ask the same questions with the addition of a response question, "Since our previous conversation about breast health and breast cancer, have you obtained or attempted to obtain a clinical breast examination or mammogram?"

SUMMARY/DISCUSSION

A breast health promotion intervention has been developed and personalized for a rural, southern, African American audience. The study is now in Phase Three in which 600 participants, 300 experimental and 300 controls, are being recruited as participants. While it is too early to draw conclusions on the effectiveness of the approaches used in this study, this article has presented some insight into common issues considered important in the development and implementation of a target-specific, breast cancer prevention and screening program.

Table 1. Sample Demographic Profile by Location.

                                           Albany
Demographic Variable              N    Mean        Median

Age                               25   51.8        50
Income                            24   16666.67    15000
Number of children                25   2.92        3
Number of pregnancies                  3.28        3
Number of people in household     25   2.4         3
Years in the commmunity           24   26.88       25.5
Weeks of delay in seeing doctor   25   4.4         1
(after discovering symptoms)
Length of time since diagnosis    25   2.92        2
Number of years married           25   13.91       4

                                       Most
                                       frequent    %

Marital status *                  25   1           52
Employment status **              24   0/2         46/46
Educational status ***            25   3           44
Healthcare coverage ****          24   1           87.5
Healthcare payment ([dagger])     25   1           56

                                       Frequency   % yes

Family members                    25   18          72
with cancer
Family members with               25   8           32
Breast cancer
Have had other cancer             25   0           0
Still menstruating                25   5           20
Taking hormone pills              25   5           20
Had a hysterectomy                25   13          52

                                           Macon
Demographic Variable              N    Mean        Median

Age                               24   56.43       55
Income                            24   12142.86    10000
Number of children                24   2.86        3
Number of pregnancies             24   3.0         3
Number of people in household     24   3.14        3
Years in the commmunity           24   29.29       28
Weeks of delay in seeing doctor   23   12.29       4
(after discovering symptoms)
Length of time since diagnosis    24   7.14        7
Number of years married           18   29.29       29

                                       Most
                                       frequent    %

Marital status *                  24   1           54
Employment status **              24   0/2         37.5/42
Educational status ***            24   3           67
Healthcare coverage ****          24   1           92
Healthcare payment ([dagger])     24   1           67

                                       Frequency   % yes

Family members                    24   19          79
with cancer
Family members with               23   9           39
Breast cancer
Have had other cancer             24   3           12.5
Still menstruating                24   2           8
Taking hormone pills              24   0           0
Had a hysterectomy                24   13          54

                                          Savannah
Demographic Variable              N    Mean        Median

Age                               14   55.3        55
Income                            13   11666.67    10000
Number of children                14   1.67        2
Number of pregnancies             13   3.15        3
Number of people in household     14   1.67        1
Years in the commmunity           14   34.4        31.5
Weeks of delay in seeing doctor   14   10.0        8
(after discovering symptoms)
Length of time since diagnosis    14   3.33        3
Number of years married           9    21.33       19

                                       Most
                                       frequent    %

Marital status *                  14   1           43
Employment status **              14   0           29
Educational status ***            14   5           36
Healthcare coverage ****          14   1           57
Healthcare payment ([dagger])     14   1/3         43/50

                                       Frequency   % Yes

Family members                    14   13          93
with cancer
Family members with               14   4           29
Breast cancer
Have had other cancer             14   1           7
Still menstruating                14   3           21
Taking hormone pills              14   0           0
Had a hysterectomy                14   5           36

* Single=0, Married=1, Seperated=2, Divorced=3, Widowed=4

** Not working=0, Part-time=1, Full-time=2, Unemployed=3, Retired=4,
Self-employed=5

*** Less than 6 yrs=1, 7-9 years=2, 10-12 years/vocational school=3,
1-2 years college/community college=4, completed 4 year college=5

**** Other=0, Private doctor=1, Clinic=2, Doctor/clinic=3, Military=4,
Hospital=5

([dagger]) State/other=0, Insurance=1, Personal income=2,
Medicare/medicaid=3

Acknowledgement: the Department of the Army, U.S. Army Medical Research Acquisition Activity, Grant No. sponsored this study: DAMD17-97-1-7312. The content of this article does not necessarily reflect the position or the policy of the government, and no official endorsement should be inferred.

REFERENCES

Ashante, M.K. (1988). Afrocentricity. Trenton, NJ: Africa World Press.

Atkins, C.K. & Freimuth, V. (1989). Formative evaluation research in campaign design. In R.E. Rice and CK Atkins (Eds). Public communication campaigns (2nd Ed.), Newbury, Park, CA: Sage.

Braithwaite, R.L., Murphy, F., Lythcott, N. & Blumenthal, D.S. (1989). Community organization and development for health promotion within an urban Black community: a conceptual model. Health Education. 20(5);56-60.

Braithwaite, R.I. & Lythcott, N. (1989). Community empowerment as a strategy, for black and other minority population. Journal of American Medical Association. 261(2); 28 2-283.

Blumenthal, D.A., Williams, J., Sung, J.F., Coates, R.J., & Liff, J.M. (1989). Lay health workers: can they help increase the rate of participation in breast and cervical screening in inner-city population? American Public Health Association Annual Meeting, Chicago, Ill.

Caplin, L.S., Wells, B.L., & Haynes, S. (1992). Breast cancer screening among older minorities and whites: barriers to early detection. The Journal of Gerontology. 47;101-110.

Chatters, L.M., Taylor, R.J. (1989). Life problems and coping strategies of older Black adults. Social Work. 34(4); 319.

Centers for Disease Control and Prevention. (1994). Results from the National Breast and Cervical Cancer Early detection program, October 1, 1991-September 30, 1993. MOIRE More Mortal Wily Rep. 43(29); 530-534.

Eng, E., Parker, E., & Harlan, C. (1997). Lay health advisor intervention strategies: a continuum from natural helping to paraprofessional helping. Health Education behavior. 24:413-417.

Farley, H. & Flanery, J.T. (1989). Late-stage diagnosis of breast cancer in women of lower status: Public Health Implication. American Journal of Public Health, 79(11), 1508-1512.

Fishbein, M. & Ajzen I. (1981). Acceptance, yielding and impact: cognitive processes in persuasion. In R.E. Petty, T.M. Ostrom, and T.C. Brocks (Eds.), Cognitive responses in persuasion. Hillsdale, N:, Lawrence-Erlbaum.

Greenberg, R.S., Liff, J.M., Clark, W.S., & Sung, J.F.C. (1988). Report of the Georgia State Cancer Registry. 1975-1985. Journal of Medical Association of Georgia 77;895-900.

Hatch, J.W. & Lovelace, K.A. (1980). Involving the southern rural church and students of the health profession in health education. Public Health Reports. 96 (l), 23-25

Horton, J.A. (Ed.). (1992). The women's health data book: A profile of women's health in the United States. Washington, D.C.: Jacobs Institute of Women's Health.

Israel, B.A., Checkoway B., Schulz, A.J., & Zimmerman, M.A. (1994). Health education and community empowerment: conceptualizing and measuring perceptions of Individual, organizational and community control. Health Educational Quarterly 21;149-170.

Jahn, J. (1961). Muntu: An outline of Neo-African Culture. New York :Grove Press, Inc.

Kreps, G.L. (1988). The pervasive role of information in health and healthcare: implications for health communication policy. In J. Anderson (Ed.). Communication yearbook 11, pp. 238-279, Newbury Park, CA: Sage Publications.

Kreps, G.L. & Thornton, B.C. (1984). Health Communication. New York: Longman, Inc.

Levin, J.S. & Taylor, R.J. (1998). Panel analyses of religious involvement and wee-being in African Americans: contemporaneous vs. longitudinal effects, Journal for the Scientific study of religion. 37. (4), 695-709.

McLeroy, K.R., Bibeau, D., Steckler, A. & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education Quarterly. 15;351-377.

Matthews, H.F., Lannin, D.R. & Mitchell, J.P. (1994). Coining to terms with advanced breast cancer: black women's narratives from eastern North Carolina. Soc. Sci. Med. 38 (6), 789-800

Mattis, J. (2000). African American women's definitions of spirituality: a qualitative analysis. Journal of Black Psychology. 26 (1); 101-122.

McAdoo, H.P. (1995). Stress levels, family help patterns, and religiosity in middle and working-class African American single mothers, Journal of Black Psychology. 21 (4); 424-449.

Miller, B.A., Kolonel, L.N., Bernstein, L., Young, J.L., Jr., Swanson, G.M., West, D., Key, C.R., Liff, J.M., Glover, C.S., Alexander, & G.A., et al. (Eds). (1996). Racial/Ethnic patterns of cancer in the United States 1988-1992. NIH publication No. 96-4104. Bethesda, MD: National Cancer Institute.

Nyamathi, A., Bennett, C, Leake, B, Lewis, C., Flaskerud, J. (1993) AIDs-related knowledge, perceptions, and behaviors among impoverished minority, women. American Journal of Public Health, 83 (1);65-71.

National Center for Health Statistics. (1996). Health United States. 1995 Hyattsville, MD: US Public Health Service.

National Women's Health Network. (1996). Perspectives: Breast cancer and African American women. Washington, DC: National Women's Health Network.

Parker, E.A., Achulz, J., Israel, B.A., & Hollis, R. (1998). Detroit's East Side Village Partnership: community-based lay health advisor intervention in an urban area. Health Education and Behavior 25(1), 24-45.

Patton, M.Q (1990). Qualitative evaluation methods. Beverly Hills, CA: Sage.

Quinn, N. & Hollard, D. (1987). Culture and cognition. In Cultural Models in Language Cognition. Cambridge, UK: Cambridge University Press.

Straus, A., Corbin, J. (1990). Basis of qualitative research; grounded theory procedures and techniques. Newbury Park, CA: Sage.

Taylor, R. J., Mattis, J., & Chatters, L. M. (1999). Subjective religiosity among African Americans: a synthesis of findings from five national samples, Journal of Black Psychology. 25 (4); 534-543.

US Department of Health and Human Services. (1987). National Institute of Heart, Lung and Blood. Strategies for diffusing health information to minority populations. September.

Williams, M.P. (1996). Increasing participation in health promotion among older African Americans: developmental processes of a model outreach program. American Journal of Health Behavior, 20(6); 389-399.

Witte, K. (1995). Fishing for success: using the persuasive health message framework to generate effective campaign messages. In E. Maibach and R. Parrott (Eds.), Designing Health Messages. Thousand Oaks, CA:Sage Publications.

Mary P. Williams is an Associate Professor in the Department of Community Health & Preventive Medicine and is Director of the Gerontology Center at the Morehouse School of Medicine. Larry Brown is an Assistant Professor and is the Director of Community Partnerships in the College of Health and Human Studies at Georgia State University. Carole E. Hill is a Professor in the Department of Geography and Anthropology at Georgia State University. Deborah Schwartz is a Graduate Research Assistant at Georgia State University. Address all correspondence to Dr. Williams at Morehouse School of Medicine; Atlanta, GA; Ph: 404.752.1626; Fax: [phone]; Email: [e-mail].