Each of these concepts is complicated by the great heterogeneity of rural areas. Sparseness of population, for example, varies widely along a continuum, including small, concentrated villages surrounded by open country as well as frontier areas with great distances between few people. Likewise, the concept of distance is complicated, involving not only number of miles, but also travel time. Mountainous terrain, poor-quality roads, and/or lack of public transportation may greatly lengthen travel time across relatively few miles. Some rural communities may seem more urban because of greater population density, although they remain remote and isolated. 

A consensus about what is "rural" is sorely needed. In the absence of such agreement, the panel, in this report, uses "rural" as a generic term referring to the end of the residential continuum that includes towns and open country with small or widely dispersed populations remote from large, metropolitan cities. Suburbs or small cities close to large urban areas are not included. 

Health Status of Rural America 

Health Care Needs

Many of the health problems facing rural Americans are similar to those affecting urban populations, whereas others are unique. Research suggests that, in the aggregate, rural populations in the United States experience greater morbidity (Yawn, Bushy, & Yawn, 1994) and higher crude rates of mortality from all causes (Miller, Stokes, & Clifford, 1987; Schneider & Greenberg, 1992) than urban or suburban populations. Specific data show that: 

Compared with urban populations, rural populations also have higher rates of adolescent pregnancy (Bayard-de-Volo, 1982), smoking and "heavy drinking" among adults (Bainton, 1981; Thompson & Weeks, 1979), suicide (Cordes & Wright, 1985), and deaths and serious injury due to automobile accidents and unintentional injuries (Banahan & McCaffrey, 1993; Kearney, Stallones, Swartz, Barker, & Johnson, 1990). 

Affecting these health care needs are a number of differences in the composition of rural and urban populations relating to the distribution of poverty, education, race, age, and marital status. In comparison with urban environments: 

Although the health of rural persons thus differs from that of urban residents, these differences are not universal across all indicators of health or all population groups and do not always place rural residents at a disadvantage. That is, when adjusted for differences in population composition (i.e., distributions by age, gender, and race), overall rates of morbidity and mortality are no higher among rural populations than among urban or suburban populations (Miller, Stokes, & Clifford, 1987). 

However, rural populations remain especially vulnerable to certain diseases and health conditions. Even after adjusting for age, gender, and racial distribution between urban and rural dwellers, rural populations have higher rates of chronic illness and infant mortality and significantly greater (40 percent) rates of death from accidents (OTA, 1990). Areas identified as medically underserved counties, many of which are rural, also have higher rates of communicable diseases and lower rates of immunization (Hawkins, Rosenbaum, Zuvekas, Leong, & Young, 1993). And, even though rates for all other causes of death may not be higher in rural areas than in urban areas, factors unique to rural communities (e.g., greater poverty, substandard living conditions, hazardous working conditions, and scarcity of primary health care services) heighten the magnitude of the health problems they face and become major contributors to mortality. 

It also should be noted that there are significant variations within and between rural communities. These variations reflect sociocultural differences, economic activity, income levels, and ethnic and racial identity factors that influence a population's health and health care needs. 

For example, regions vary substantially in sociocultural patterns (Lyson & Falk, 1993; Markusen, 1987). Rural communities in the Mississippi Delta have distinctly different cultural entities and contexts than communities in the Great Plains. Primary economic employment sectors also vary (Bender et al., 1985), with some communities based on agriculture, others based on manufacturing, and still others on tourist or retirement industries. Income levels within a community may be broad or concentrated at high or low levels. 

With respect to race and ethnicity, some rural communities and regions are overwhelmingly white (European American), whereas others are predominantly black, Native American, or Hispanic. Within each of these community populations, there may also be a variety of racial, ethnic, or cultural subgroups (e.g., persons of Norwegian or Croatian origin in white communities). The combination of these differences yields a unique set of health care problems for a community, as well as opportunities for and obstacles to community-based health care services. 

In sum, rural people are not healthier than their urban counterparts, despite popular images. Indeed, there is some evidence that, for certain diseases, rural populations as a whole are worse off than urban populations and that certain rural populations, such as low-income African Americans in the rural South, have unusually high morbidity and mortality rates compared with urban residents. Also, there is evidence for some comparisons that, even after adjusting for compositional differences between rural and urban populations, there is still a negative effect associated with life in a rural area. 

Health Care Services

Although rural populations have health care needs that are equal to or greater than their urban counterparts, in general they consume fewer health care services and, thus, resources (National Center for Health Statistics, 1989; OTA, 1990). Multiple factors may account for this finding (e.g., fewer services available in rural areas, greater poverty, lower average incomes, less likelihood that rural residents will have health insurance, less Federal funding for health services, cultural barriers to seeking health care). 

A critical factor in the health status of rural America is that small towns and rural communities suffer from a lack of health care providers (Hicks, 1990) and have access to a smaller number and narrower range of community-based health care services than urban populations (Clarke & Miller, 1990; Coward, Bull, Kukulka, & Galliher, 1994; Gesler & Ricketts, 1992; Mick & Morlock, 1990). Given their limited numbers and the paucity of economic incentives to practice in rural areas, health care providers in these environments most likely will focus on persons in most need of care (e.g., the sickest), rather than emphasizing health promotion and disease prevention. Rural populations are thus less able to cope with their illnesses because of the dearth of services even though, in some communities, they may have the same incidence and prevalence of disease as an urban population. 

As with the health problems faced by rural communities, however, not all communities are the same. They do not share health care delivery services equally (Cordes, 1989) and the nature and organization of these services reflect a community's economic and population characteristics. 

Populations and Communities At Risk

Certain rural communities appear to be particularly "at risk" of having high levels of need or low levels of services. For example, rural residents of persistently low-income counties face unusually harsh circumstances (e.g., poverty, lack of resources, chronic needs) when attempting to cope with their health care needs. Despite substantial economic growth and prosperity in most of America since the 1950s, some rural communities remain poor and are considered to be "pockets of poverty" (Weinberg, 1987). These nonmetropolitan, persistently low-income counties have a high need for health care services (Voss & Fuguitt, 1991), but the availability of health care resources for these communities is especially low (Coward, Duncan, & Netzer, 1993). 

Similarly, the health care needs of frontier counties are receiving increased attention. Because of these counties' large size and widely dispersed populations, they are viewed as extremely difficult and costly settings in which to deliver health care services (Hewitt, 1992) and/or call for innovative approaches to providing services. 

Certain rural population groups also are particularly at risk for poor health and the underutilization of health care. As with urban minority populations, this risk for some groups results from high levels of need combined with low levels of access to services (Beaulieu, 1988; Rowland & Lyons, 1989). For other groups, greater risk results from the higher incidence of conditions in rural America, such as poverty, lower levels of education, unemployment, underinsurance, extreme distance and/or difficulty of travel, and lack of transportation services. Despite the dominance of urban images of poverty, research indicates that a higher proportion of rural populations have incomes below the poverty threshold (Hoppe, 1993). 

For some populations, greater risk results from high levels of impairment, such as among frail elders (Coward, Bull, et al., 1994). The reduced availability of specialized services and the long waiting periods to receive these services may also place a particular group at risk. In general, certain population groups face higher probability of a poor health outcome because of physical illness combined with their rural residence. 

In some rural areas, the persons in these at-risk subgroups may be few in number and widely separated. In other areas, most of the community at risk may belong to a particular ethnic or cultural group experiencing poverty, inadequate housing, or poor health, such as African Americans in some southern communities, Hispanics in the Southwest, and Alaskan Natives. 

Community-Based Health Care

Community-based services are especially needed in rural areas. While public health nurses have long focused on the health of communities, the number of nursing positions in public health departments has decreased in many areas, severely compromising the functioning of the public health sector. In addition, traditional health care sites such as doctor's offices, clinics, and hospitals are frequently no longer available because of the out-migration of health care providers. For economic reasons, hospitals have closed and/or physicians have closed practices. For example, in Arizona, Colorado, Nevada, and New Mexico, most counties have either no hospital beds or fewer than 4.2 per 1,000 population (OTA, 1990). 

Other difficulties arise from an emphasis on categorical funding of health care services. Clients and families accessing these services may have to make multiple visits to health care facilities to address different problems, increasing the possibility that some health care needs may not be met. 

Still, the demand for health care services remains, and innovative approaches to these services are needed in rural areas. Opportunities exist for forming partnerships among providers, research teams, and community members to develop and test new strategies of health care delivery. Services that are community based are more likely to be more acceptable and longer lasting because community members are integrally involved in the design, organization, and provision of services. 

With community involvement, services are more likely to meet local needs and tie in with the local culture. Also, community members are more likely to participate in fundraising, identification of donors and sponsors for specific activities, and development of volunteers, thus offsetting some of the financial costs of the services provided. Community health centers linked to larger health systems (e.g., county or regional health departments, managed care organizations, medical centers) offer a means by which community members can retain local decision-making power and yet have financial stability through economies of scale. 

Community-based care includes a broad range of services provided outside of institutional settings to community-dwelling individuals across the lifespan, enabling them to maintain health and independent functioning. The continuum or array of community-based services combine informal care (provided by family, neighbors, friends, clergy, volunteer services, etc.) with formal care to address a variety of health and social needs (Krout, 1994). 

As noted in chapter 1, the aim of community-based intervention strategies is to provide primary health care services. Interventions are accomplished through broad-based involvement of community members and are available to all community members. Although some interventions may be targeted to specific populations or groups at risk, in rural communities the small size of the populations particularly argues against selecting population subsets to receive highly specialized services. 

Although there are notable exceptions such as migrant health centers, rural health care delivery most often is focused on an entire community, targeted to all residents across the lifespan and across ethnic, racial, cultural, and geographic differences. Multiple strategies have been applied and multiple variables have been examined in relation to selected rural health problems. However, few community-based interventions in rural areas are described in the research literature. And, because the health problems of rural populations are studied less often than those of urban populations, the research base is scarce and replicated studies are few. 

One particular problem is that the urban or rural nature of a population is not distinguished consistently across studies. Further, no single intervention strategy or model has been definitively tested across a range of geographic locations nor have strategies been assessed across all levels of intervention (primary, secondary, and tertiary) and clients (individuals, families, and communities). 

Focus: Strategies of Care

In this chapter, the panel focuses on the continuum of services provided in primary health care primary, secondary, and tertiary prevention. Its discussion centers on six challenges confronting community-based health care strategies: availa- bility, accessibility, acceptability, affordability, appropriateness, and adequacy. The panel recognizes that these challenges relate to both the provision and receipt of health care, reflecting the perspectives of both providers and recipients of care. Attention also is given to outcomes, the result of effective care delivery. 

In the chapter, the panel addresses parts of the continuum separately; that is, primary and secondary prevention are considered separately from tertiary prevention. This separation reflects different program goals, types of interventions, outcomes, and funding. Whereas primary and secondary prevention is aimed at preventing the onset of disease and its sequelae among individuals and populations, tertiary prevention is focused on supportive and restorative care for individuals with chronic illnesses and disabilities. In rural areas, many programs and resources have been targeted to long-term, tertiary care (e.g., Medicare, Medicaid), but relatively little attention has been paid to primary and secondary prevention. 

The panel does not aim to be exhaustive in its review of the state of the science on community-based health care interventions for rural populations. Rather, the panel focuses on nursing or multidisciplinary interventions and describes examples of research completed or under way which highlight the state of the science of research in this area. Not inclusive of all projects using community-based interventions, these examples illustrate key aspects of assessment protocols and community-based strategies. Based on this review of the state of the science, the panel delineates research needs and opportunities for the future and recommends a specific agenda for research on community-based health care strategies for rural populations in the United States. 

State of the Science

Delivery System Challenges

Researchers studying health care delivery in rural communities have addressed variables most commonly associated with effective outcomes. However, little research exists on outcomes of rural health care or rural nursing interventions. Utilization of services is most frequently mentioned, but is identified from the perspective of care providers. 

Many rural health care delivery challenges are perceived by consumers or recipients of care; other challenges are perceived by health care providers or policy makers as either barriers or facilitators to care delivery. These challenges interact and affect a variety of outcomes in addition to utilization. Research on the identification, naming, and de- scription of these outcomes is grossly insufficient. 

As noted in chapter 1, the panel focuses on six challenges: the availability, accessibility, affordability, appropriateness, adequacy, and acceptability of health care. In the following description of these major challenges, Krout (1994) includes awareness, which the panel incorporates in its definition of accessibility. As noted by Krout (1994, pp. 15-16): 

Overall, rural community-based services must overcome challenges posed by availability, accessibility, awareness, acceptability, affordability, appropriateness, and adequacy (Krout, 1986; Williams, Ebrite, & Redford, 1991). ... It is generally believed that many services are not as likely to be found in rural areas (availability) and that the larger distances between the services that do exist and the people who need them result in higher provision costs in terms of time and money (accessibility). Services that are available may be incomplete or of poor quality (adequacy) to meet the level and diversity of need, and research has shown that rural older persons have relatively low awareness of the services that do exist in their areas. Available services may also be too costly (affordability) for older rural persons. Additional problems can arise in that the services available may not be the ones that are needed (appropriateness), something Wallace (1991) has called the "no-care zone." Finally, observers such as Coward (1979) and Rowles (1991) have noted that the content and presentation of the service must fit in with the distinctive attitudes and values found among different rural populations (acceptability). 

The dimensions of each variable or challenge may be different in rural settings than in urban settings and reflect the unique character of each community. Commonalities in the expression of these challenges may be ascertained across commu- nities and settings. Research findings pertaining to the dimensions and expression of these challenges in rural communities are highlighted below. 

Availability

Availability -- the presence of services has two important components: objective data documenting the presence or lack of services, and rural residents' subjective perceptions of the availability of services. As a predictor of utilization, the perception of available services is equal to, and perhaps more important than, the actual presence of services. 

Objective Data. The lack of health care services in rural areas is well documented (Conway-Welch, 1991). There is a shortage of primary care sites in rural areas, as well as a maldistribution of primary care providers in rural areas (OTA, 1990). 

In 1986, 1,949 areas in the United States were designated by the Federal Government as primary care shortage areas because they had too few primary health care providers for the size of the population (OTA, 1990). Two thirds of these designated areas were rural, and almost half of the Americans living in shortage areas were rural residents. 

Parker, Quinn, Viehl, McKinley, Polich, Hartwell et al. (1992) and the OTA (1990) note that it is costly for health care providers to deliver health care and social services to rural areas because of population disbursement and low population densities, great distances, financial constraints, lack of public transportation, and difficulty in recruiting and retaining providers. 

The inability to recruit and retain health care specialists is well described (Hicks, 1990; OTA, 1990). Many physicians who could provide primary care do not locate in rural areas. Substantial evidence, however, supports the appropriateness of family nurse practitioners as providers of primary care to rural populations (Lawler & Valand, 1988). Research shows that these practitioners provide quality care, are cost effective and widely accepted, and yield satisfied clients (Conway-Welch, 1989; Fagin, 1990; Feldman, Ventura, & Crosby, 1987; Lawler & Valand, 1988). 

The OTA found that nurse practitioners "improve geographical access to primary care and are especially valuable in improving access in rural areas and in health programs for the poor, minorities, and people without health insurance" (OTA, 1990, p. 30). Observed and reported outcomes show that nurse practitioners in underserved rural regions have a positive impact on care. For example, Lawler & Valand (1988) note that use of nurse practitioners results in an increase in the number of clients seen, access to care, available primary care, client education, and attention to adjunct health problems. 

Utilization of diagnostic, treatment, and long-term health care services by rural residents may be limited owing to a combination of factors: ill-equipped and poorly staffed facilities, hours when facilities are open, distances between resi- dents' homes and facilities, and lack of public transportation. However, the impact of these factors for individuals of different ages, incomes, and education is not well described in the research literature. 

Subjective Data. The effect of availability on utilization may be further complicated by residents' perceptions. Rural residents may not consider services in nearby communities accessible because they perceive that nothing is available in their town (L. Phillips, personal communication, July 9, 1994). Also, many individuals may recall available services being closed in their communities and may be despondent about the loss of services that they consider to be basic health care. Geographic distance and the context of residents' perceptions regarding service availability are important subjective variables. 

Accessibility

Accessibility -- the ability of persons to use services, including factors such as distance, effort, cost, and awareness of services is a key issue for rural residents. Accessibility includes a facility's hours of operation and location. 

For example, although a rural clinic might be located within 10 miles of a worker, the person may view the clinic as inaccessible because of its hours of operation (9 a.m. to 3 p.m.). Also, in their description of rural family care activities, Bender, Weinert, Faulkner, and Quimby (1991) indicate that distance traveled is an important issue in rural families' utilization of services. Weinert and Long (1993) and Given, Stover, and White (1994) further report that families may choose nontreatment options for cancer, such as not following up with recommended chemotherapy, when specialized care is not accessible and/or available in their own community, foregoing the specialty care needed. 

With respect to the care received, Howe, Katterhagen, Yates, and Lehnherr (1992) indicate that health professionals may alter treatment for rural women who travel to urban areas for care so that less travel and followup are needed. They further suggest that rural women often have more extensive surgery to reduce the need for adjuvant therapy and subsequent trips to distant cancer centers. 

Acceptability

Acceptability -- the degree to which individuals perceive that they can use the services provided to them is another challenge to rural health care delivery and is viewed from the perspective of individuals, families, and communities. Some families in rural areas make health care choices based on perceptions of competence of care, previous family experiences, language, and cultural or attitudinal sensitivity of health care providers (Magilvy, Congdon, & Martinez, 1994). To some degree, acceptability is a perception held by an individual or family; however, the limited research on acceptability of care impedes understanding of individual, cultural, ethnic, gender, residence, socioeconomic status, age, and other variables influencing health care decisions and utilization of services. The issue of acceptability is very significant, however, because it can influence outcomes such as utilization patterns, satisfaction with care, and cost effectiveness of rural health services. 

In small rural communities, most people know "just about everyone." Difficulties in qualifying for services, perceived inadequacies of services and/or providers, and other incidents or perceptions may be shared rapidly among the population. These shared values may create a situation in which services are not used which, in turn, decreases the likelihood that services will remain and/or quality will be maintained a downward spiral of events that can lead to complete loss of services. 

Acceptability includes perceptions of congruence between the services offered and individual or family values and beliefs about health care. Culture, rurality, gender, socioeconomic status, and age are a few of the factors influencing the acceptability of services for all rural residents. Acceptability is a particular issue when addressing health care for minority segments of the rural population, such as migrant farmworkers; Hispanic, African American, and Asian persons; frail older adults; and poor and medically indigent persons (Friedman, 1990). 

Culture can profoundly influence the acceptability of services. For example, some groups highly value respect which, if not incorporated into the care provided, may lead them to deem the care unacceptable and either decrease or cease their use of the services offered. Providers must be sensitive to such nuances and differences among and within specific groups. That is, all individuals may wish to be respected, but the relative importance of respect or how respect is demonstrated may vary across groups (Coward, 1979; Rowles, 1991). 

Respect may involve proper use of specific terms. For example, use of an older woman's first name usually occurs later, and with permission, in a relationship with Mexican American women than in a relationship with white, non-Hispanic women. "Se¤ora" precedes the last name in such a relationship (i.e., between provider and client), whereas "do¤a" is reserved for older, well-respected women in some Hispanic groups. 

Respect also involves assuring the privacy of clients. Among deaf speakers, for example, use of a certified interpreter, rather than a family member, to interpret health care situations demonstrates respect for the deaf person and their right to have a private and accurate conversation with a provider. 

Rural residents also share strong concerns about preserving their anonymity when using certain services (e.g., behavioral health, family planning) (L. Phillips, personal communication, July 9, 1994). The acceptability of services may be less than desired when the providers' employees also are community members and sometimes neighbors, when clients can easily be seen entering a care facility, or when clients have to park their cars facing a main road. 

These examples demonstrate that health care providers must listen and respond to their referent group(s) in order to provide acceptable, culturally congruent care. Without this understanding, a provider's errors in interacting with a client will compromise the acceptability of care and, thus, health care outcomes. 

Affordability

Affordability of health care services also affects rural health care utilization. Both tangible and intangible costs are involved. 

Insurance Coverage. With respect to tangible costs, rural residents are less apt to have health insurance than individuals in urban and suburban areas (Coward, Clarke, & Seccombe, 1993; Ries, 1991). This lack of health insurance coverage (private or public) has been shown to have a negative impact on utilization of health services (Butler, 1988) and, as a consequence, insurance (Freeman, Aiken, Blendon, & Corey, 1990; OTA, 1992; Spillman, 1992). The reduced access and availability of health care providers in rural America, combined with the increased likelihood of not having insurance, may place some rural residents at risk of not receiving the health care services needed. 

Intangible Costs. Other costs of care also must be met. These include lost time from work in order to seek and receive care; assistance of friends, neighbors, and/or family members in caring for children while receiving care; and informal support and volunteered time to assure continuity of care. 

The perception of the costs of care varies between provider and client. Whereas a provider focuses on tangible costs of care, a client must weigh both tangible and intangible costs. For example, a provider may consider prenatal care to be relatively inexpensive or have essentially no cost, but a client must consider the work time her partner loses when driving her to and from the clinic and the time her neighbor spends taking care of her other children. In rural areas, because of the distances traveled, a prenatal visit may consume 5 or 6 hours per visit. Support services such as these bear a cost which, too often, are not included in assessments of rural health services. These costs, and the perception of these costs by individuals and families in a community, are important factors in service utilization. 

Appropriateness

Services provided to rural residents may not be appropriate if they are not the services needed or desired and/or not targeted to those who can make best use of them. As Wallace (1991) noted, inappropriate services place rural residents into a "no-care zone." 

Questions to assess the appropriateness of an intervention may include: "Will the interventions be ones needed or wanted by the client?" "Are personal preferences taken into account?" (Krout, 1994). On a larger scale, questions may relate, for example, to the appropriateness of care or assuring access to all clients when resources are limited. 

Indeed, examination of policies related to appropriateness of care may be warranted. Comprehensive, community-based assessment and planning of health care services can be expected to increase the likelihood that appropriateness criteria are met in community-based interventions for specific populations. However, the panel identified no research on the appropriateness of rural health services. 

Adequacy

Adequacy of community-based interventions involves assessment of merit (i.e., the quality and completeness of services to meet the level and diversity of need) (Krout, 1994). Adequacy can be assessed at both program and community levels. Program-level assessment may include examination of a program's ability to enable individuals to remain independent or meet minimal needs and to prevent potential, anticipated needs or determination of whether interventions are designed at the appropriate level of care or prevention. 

Community-level assessment involves determination of whether a service or nursing intervention meets the stated needs of the community, as defined by community members in partnership with the planners and implementers of care. Assessments that only include, for example, statistical data and observation of the environment by nonresident providers may yield an incorrect or insufficient statement of needs (e.g., for transportation, family support groups) and thereby jeopardize any strategies that are implemented. Community members' participation in assessments may be obtained through focus groups, task forces, town meetings, and community assessment teams that include health professionals and lay persons. 

The issue of adequacy in community-based interventions has received little research attention, particularly with respect to rural populations, and research is needed to explore this challenge in rural areas. It also should be noted that the adequacy of services can only be determined if they are used. An important aspect of adequacy, therefore, is whether the services are known, used, and coordinated. 

Outcomes

Utilization of health care services and providers by rural consumers has been shown to be influenced by consumers' perception of the availability, accessibility, acceptability, affordability, adequacy, and appropriateness of these services (Aday & Andersen, 1984; Krout, 1994). While these issues are shared with urban dwellers, they appear to influence utilization patterns among rural residents to a greater extent (Hicks, 1992). Nevertheless, utilization is not the outcome that should be measured. More meaningful measures of the outcome of health care services are improved health status, quality of life, health promotion and disease prevention, and satisfaction with health care. An examination of outcomes must encom- pass a number of factors, only one of which is utilization. 

Thus, even though low-income individuals have poorer health status than high-income individuals, they may not access much of the health care system because they lack the personal resources to do so (Rowland & Lyons, 1989). Given that a high proportion of the populations in small towns and rural communities have incomes below the poverty threshold (Hoppe, 1993), individuals in rural communities may have especially low utilization rates. 

Summary

Utilization of health care services by rural populations is limited in some areas for some services. Factors affecting utilization include availability, accessibility, acceptability, affordability, appropriateness, and adequacy. As Hicks (1992, p. 30) summarizes: 

The problems faced by rural populations in accessing health care services include lack of financial resources, inadequate number of providers, excessive distances to providers and corresponding transportation problems, and substandard living and hazardous working environments. The rural population, in general, experiences more of these problems and barriers than their metropolitan counterparts. 

Such problems, which often become barriers, may become risk factors for poor health; as such, they must be modified in order to improve health. 

Knowledge of effective health care interventions to ameliorate and/or remove these challenges of rural health care delivery is incomplete, and few studies examine the effectiveness of different nursing interventions that address barriers to utilization across sites. Although health care needs in rural areas are heterogeneous, reflecting differences within and between communities, examination of the commonalities and differences in barriers impeding health care may benefit all communities. Additionally, a spectrum of outcomes, including utilization, needs to be examined to determine success. The knowledge gained may enable providers' to tailor services better while enhancing health care at all levels of prevention. 

Primary and Secondary Prevention

Primary prevention involves health promotion and protection from specific diseases or injuries before they occur, whereas secondary prevention focuses on early diagnosis and prompt treatment of presymptomatic or early disease to prevent further disease and disability (Leavell & Clark, 1965). Health promotion and disease prevention efforts are an important aspect of both primary and secondary prevention. The state of the science of primary and secondary prevention research in rural communities is summarized below in two main sections: key areas of research, and primary and secondary prevention strategies. 

Key Areas of Research

With respect to health care, three major areas of need have been highlighted for rural populations: infant mortality, unintentional injuries, and contributors to mortality in the United States. The first two areas are described in the Report of the Joint Task Force, National Association of Community Health Centers (NACHC) and National Rural Health Association (NRHA) (1988, 1989); the third area is described by McGinnis and Foege (1993). These health care needs, and the problems associated with them, have been well documented in the health care literature. Drawing from the guidance of these resources, the panel reviewed these areas as examples of key areas of research, considering the state of the science in each area. Because infant mortality is part of a broader complex of issues, the panel addressed this topic within the larger area of maternal and child health. 

Maternal and Child Health 

This vast area of need includes prenatal care and larger family issues. As already noted, accessing prenatal care is particularly difficult for rural women. Yet, research shows that adequate prenatal care is positively related to healthy birth outcomes (Institute of Medicine, 1985; Newbitt, Connell, Hart, & Rosenblatt, 1990; U.S. Public Health Service, 1990). Also, adequate maternal and child health care during childbearing and early childrearing years are essential parts of effective family functioning. Rural areas have been thought to be places in which strong family ties develop and sustain the family in meeting its economic and social needs. However, understanding of family mechanisms that affect health outcomes is limited, including the role and function of families in prenatal care, parenting, long-term care of family members, family violence, and other family concerns. 

Interventions focused on the family are also few. Three studies are noted. The first is the Prevention of Prematurity Project (POPP), a community-based intervention in the Midwest, which was associated with accurate case-finding of high-risk pregnancies and better outcomes when prematurity did occur (Orr & Reno, 1986). The second is the "De Madres a Madres" program, an urban-based intervention that was effective in increasing the number of Hispanic women informed about the need to begin prenatal care early (Mahon, McFarlane, & Golden, 1991). In this community partnership among volunteer mothers, businesses, and the public, a community health nurse trains volunteer mothers to identify other high-risk mothers and provide community resource information and social support in a culturally relevant manner. This Houston, Texas, project has potential application for rural areas. 

The De Madres a Madres program was subsequently applied in the Chicago area where it was expanded to include recruiting, training, and supervising Hispanic lay health workers while assessing families, developing care plans, and evaluating family progress (case management). This expanded approach proved effective in improving the health status of Hispanic mothers and children in both low- and high-risk pregnancy situations in this community (Bray & Edwards, 1994). The culturally relevant nursing intervention approach used in this study could be instructive for rural areas. 

The third study, the Wyoming Perinatal Substance Abuse Prevention Program, is a statewide, model demonstration program to prevent substance abuse among pregnant and postpartum women (Wilkerson, 1993). The objectives of this program were to mobilize rural and frontier community involvement in activities to increase awareness of the dangers of substance abuse among perinatal women; to identify and diagnose pregnant women who are abusing substances; and to refer these women to treatment. 

Emphasis was given in this program to developing volunteer, community lead professionals who could serve as change agents. The strategies adopted (e.g., a decentralized "train-the-trainers" model, community action planning) were applied successfully to empower community members, disseminate prevention information, increase treatment referrals, and ensure sustainability after the end of the funding period. The features of this flexible, culturally sensitive, and cost-efficient program could be generalized and applied in other areas and for other health problems. 

Besides these specific examples, the panel notes that the public health sector in the United States has long attended to the needs of women and children, viz., the Federal Government's Special Supplemental Food Program for Women, Infants, and Children (WIC) and the many state-wide initiatives taken to help pregnant and postpartum women who do not otherwise qualify for Medicaid. The programs may offer special opportunities for community-based research and intervention. 

Infant Mortality. A key index to a nation's health is infant mortality, which, despite public health efforts, continues to be significantly higher than desired. For this reason, the panel chose to focus on this one aspect of maternal and child health. In the United States, rates of infant mortality are slightly higher in rural areas than in urban areas. In 1988, these rates, for blacks, were 19.1 per 100,000 in metropolitan areas and 19.7 per 100,000 in nonmetropolitan areas, and, for whites, 9.6 and 10.0, respectively (NACHC & NRHA, 1988, 1989). 

Infant mortality is a significant problem for both urban and rural areas in the United States, particularly among certain population groups. In rural communities, this problem is exacerbated by higher rates of poverty and the scarcity of primary health care services (i.e., the differences in rates of infant mortality between rural and urban areas cannot be explained by race alone). 

A number of investigators (Buehler, McCarthy, Holloway, & Sikes, 1986; Centers for Disease Control, 1983; Eisner, Pratt, Hexter, Chabot, & Sayal, 1978; Goldenberg, Humphrey, Hale, Boyd, & Wayne, 1983; Hein & Lathrop, 1986; Siegel, Gillings, Campbell, & Guild, 1985) have examined the reasons for changes in infant mortality rates in rural areas. Since infant mortality is closely linked with low birthweight, most of these researchers found improvement in outcomes for low-birthweight (LBW) infants, rather than reduced incidence of LBW. 

In one study conducted in rural Georgia, investigators attributed 94 percent of the decline in infant mortality to improved survival within birthweight categories and 6 percent to higher birthweights (Centers for Disease Control, 1983); this positive change was not accompanied by an increase in admissions to intensive care units for newborns. However, much of the improvement seen in infant mortality rates in other states (i.e., Alabama and Iowa) (Goldenberg et al., 1983; Hein & Lathrop, 1986) is attributed to increased access to neonatal intensive care services. These authors further suggest that the desired direction for the future is increased specialized infant care services. 

Other studies document that improved nutrition, early prenatal care, and cessation of smoking result in reduced incidence of LBW which, in turn, lowers the infant mortality rate and reduces costs associated with neonatal intensive care (Kleinman & Madans, 1985; Kleinman, Pierre, Madans, Land, & Schramm, 1988; McGinnis & Foege, 1993; U.S. Public Health Service, 1988). Tobacco use alone is estimated to cause 10 percent of infant deaths and 20 to 30 percent of LBW cases (Kleinman & Madans, 1985; McGinnis & Foege, 1993; U.S. Environmental Protection Agency, 1992). 

Strategies. An effective strategy for improving birth outcomes and reducing the incidence of LBW is primary prevention through health promotion/disease prevention activities and improving access to primary care. Prevention of LBW through improved nutrition and early prenatal care is consistent with the findings of the Joint Task Force of the NACHC and NRHA, which indicate that a reduction in infant mortality requires relief from poverty and improved access to care (NACHC & NRHA, 1988, 1989). As noted previously, access to care is a particular problem in rural areas, especially medically underserved counties, which, in both rural and urban areas, bear a disproportionate share of infant deaths and LBW infants (Hawkins et al., 1993). Rooks and Winikoff (1990, p. 38) summarize the situation well: 

The United States does better than any other country at saving the lives of low birthweight babies; regionalization of perinatal care contributed to this achievement. However, regionalization did little or nothing to reduce the causes of low birthweight, and therefore its actual incidence. Because of a high proportion of low-weight babies, the United States' international ranking with regard to infant mortality worsened during the 1970s and 1980s.... The underlying causes of preterm labor and intrauterine growth retardation (which are the immediate causes of low birthweight) must be addressed before labor begins. 

Public health nurses have traditionally addressed prenatal care in collaboration with physicians. Currently, care is targeted to clients of categorical programs, such as WIC initiatives. 

Unintentional Injuries 

Unintentional injuries are a widespread problem among rural as well as urban populations. In the rural environment, these injuries arise from agricultural activities, exposure to toxic agents such as insecticides and herbicides, motor vehicle accidents, and firearm use. The last three are specifically mentioned by McGinnis and Foege (1993) as major factors contributing to deaths in the United States. 

Unintentional injury is a major health problem in rural areas for all age groups. Type of employment (e.g., farming), lifestyle, and cultural norms are potential underlying factors in the high rate of injuries among rural residents. Although the nature and extent of injuries have been well described in the literature, few health promotion activities targeted at reducing the incidence of injury have been tested at the community level. 

Agricultural Injuries. No complete and accurate data base exists for agricultural injuries and illnesses such as noise-induced hearing loss, anhydrous burns, chemical and insecticide poisoning, farmers' lung (hypersensitivity pneumonitis), and carbon monoxide poisoning. Most research in this area has been epidemiological, providing estimates of the extent of occupational illnesses and injuries and identifying primary causes. This baseline information is essential for designing appropriate interventions. 

In 1988, about 2.6 percent of the U.S. labor force worked in farm occupations (U.S. Department of Agriculture, 1989). Farmworkers are classified in three categories: resident farmworkers who reside on farms; nonresident farmworkers who live in rural areas, but not on the farm where they work; and migrant farmworkers who travel to farms for seasonal harvesting. Children are often included as workers in all three groups. 

Although farmworkers are a small proportion of the labor force, their rate of injury is higher than in any other industry (National Safety Council [NSC], 1993). However, less than 0.002 percent of Federal resources for occupational health and safety are expended for these workers (Wakefield, 1990). 

In 1992, workers 14 years of age or older sustained an estimated 140,000 agricultural work injuries, and there were 1,200 agricultural work-related deaths in workers of all age (NSC, 1993). Approximately one half of these injuries and deaths involved farm residents; the other half involved nonresident farmers working on farms or in other agricultural industries (e.g., fishing and agricultural services and forestry, excluding logging) (NSC). 

Children sustain a significant portion of agricultural injuries and fatalities. Each year, an estimated 300 children and adolescents die from farm injuries and 23,500 incur nonfatal injuries (Rivara, 1985). These figures represent minimum estimates since the U.S. Department of Labor does not collect data on children less than 14 years of age. Indeed, all estimates of farm or farm-related illness and injury are likely to reflect underreporting for all age groups. 

Farm machinery is the most common cause of injury and death for children and adolescents living or working on farms, and tractors account for one half of these deaths. Injuries and fatalities from farm machinery result from rollovers (many tractors are not equipped with rollover bars); harvesting equipment, which causes crushing or amputation injuries; and power-take-off equipment with shafts that operate at 500 to 1,000 revolutions per minute, which can twist a worker around a shaft, causing suffocation, scalping, and avulsion injuries. Machinery operating in enclosed spaces also may cause carbon monoxide poisoning (Wright, 1993). 

Farm machinery also operates at a high noise level, which can cause noise-induced hearing loss (Wright, 1993). It has been demonstrated that farmworkers suffer hearing losses characteristic of noise-induced hearing loss and, when compared with other workers, are affected at an earlier age and more severely (Thelin, Joseph, Davis, Baker, & Hosokawa, 1983). Results from a study of Wisconsin farmworkers suggest that approximately 2.5 percent of men had a hearing loss affecting their communication by age 30, and 50 percent had a hearing loss affecting communication by age 50 (Karlovich, Wiley, Tweed, & Jensen, 1988). A study of high school students showed an increased prevalence of hearing loss among those involved with farmwork compared with those not involved with farmwork, suggesting that the hearing loss observed in adult farmers may begin in childhood (Broste, Hansen, Strand, & Stueland, 1989). 

Hearing loss associated with use of farm machinery can be prevented by reducing noise levels through engineering changes in the equipment and by workers' use of hearing-protection equipment (Lusk, Ronis, & Kerr, in press; Lusk, Ronis, Kerr, & Atwood, 1994). Manufacturers of farm machinery incorporate many safety features in new equipment they sell, but much of the machinery in use is older and the workers alter the machines to circumvent safety features perceived as interfering with efficiency. Further, farm families are typically opposed to legislation to regulate their work practices or equipment. 

A major intervention effort to combat agricultural injuries is under way. Beginning in 1990, the National Institute for Occupational Safety and Health (NIOSH) expanded its activities to develop a comprehensive, research-based intervention program to reduce injury and disease among agricultural workers and their families (Myers, 1992). According to Myers (1992, p. 548), this program is designed to: 

1. Distribute injury and illness prevention messages to farmers 

2. Assign nurses to rural areas to talk about prevention in farming communities and to assess the incidence of injury and illness among farmers 

3. Provide cancer screening and assess cancer rates in farming communities 

4. Evaluate farms for safety hazards and determine the incidence of illness among farm family members 

5. Award academic grants to establish new Agricultural Health and Safety Centers and for applied research in intervention methods. 

This new agricultural initiative includes funding for an Occupational Health Nurse in Agricultural Communities (OHNAC) program. Through the OHNAC program, 31 nurses in rural hospitals, clinics, and health departments in 10 states are engaged in surveillance of agriculture work-related illnesses and injuries. Surveillance is generally defined as "ongoing systematic collection, analysis, interpretation, and dissemination of relevant health data to all who need to know" (Connon, Freund, & Ehlers, 1993). With the assistance of the state health departments, these nurses also are issuing Hazard Alerts for dissemination to farmers and developing targeted interventions (e.g., increasing the acceptability of safety devices, incorporating safety practices into daily activities) to prevent future incidents. Although this program is promising, no assessment of its effectiveness has been published to date. 

Similarly, no large empirical studies have been conducted to determine the effectiveness of interventions to prevent agricultural illnesses and injuries. The role of agricultural health nurses has been described (Randolph & Migliozzi, 1993), and community-wide approaches to promoting farm health and safety have been suggested (Lexau, Kingsbury, Lenz, Nelson, & Voehl, 1993). No findings have been published yet from the research grants awarded in 1990 by NIOSH for applied research on intervention methods to enhance agriculture safety and health. 

Migrant farmworkers are among the most disadvantaged, medically indigent persons and have the poorest health of any group in the United States (Dever, 1991; Goldsmith, 1989). This vulnerable population experiences problems such as poverty, underemployment, occupational hazards, acute and chronic health problems, isolation, and alienation (Dever, 1991; Jezewski, 1990; Littlefield & Stout, 1987; Rust, 1990; Smith & Gentry, 1987). Frequent moves, inadequate housing, and poor access to medical care also characterize this population. 

Limited research exists concerning migrants' health status, problems, and practices; use of traditional and alternative health care services; and cultural influences on these variables. According to the National Advisory Council on Migrant Health (1993, p. 6): 

Anecdotal information has highlighted various aspects of the hardships of migrant health and lifestyle. However, the lack of national research and hard data on migrant and seasonal farmworkers has hindered the efforts of clini- cians, administrators, policy makers, and researchers to effectively make changes and establish priorities for migrant health. 

Specific health problems among adult migrants that warrant increased attention include occupational or environmental diseases and health problems (pesticide exposure, back problems, lack of sanitation, occupational dermatoses, acute injuries, and joint trauma); infectious diseases (tuberculosis, upper respiratory infections, acute conjunctivitis and pharyngitis, streptococcal and viral infections, parasites, and rampant diarrhea); chronic illnesses (hypertension, diabetes); and other health issues (e.g., lack of health screening and primary health care, dental and nutritional problems) (Dever, 1991; Rust, 1990; Smith & Gentry, 1987). 

Local farmworkers who do not reside on farms experience many of the same problems as migrant workers. These farmworkers are one of the lowest paid and least protected workers in the United States. Most of these workers are local rural residents with low levels of education, low wages, and often only seasonal work with no benefits (Beale, 1990). 

Exposures. Rural workers are commonly exposed to high noise, toxic agents, weather extremes, and infectious materials, all of which cause harmful effects. For example, many farmers use anhydrous ammonia fertilizer which, on contact, can cause irritation, burns, and asphyxiation. Insecticides and herbicides, used to increase crop production, can cause poisoning leading to coma and death and are associated with an increased incidence of cancer. Farmer's lung, a chronic debilitating condition, is caused by fungal spores in moldy grains. Chemical pneumonitis and pulmonary edema, caused by nitrogen oxides in silos, may result in illness and death. Methane gases from manure holding tanks in livestock buildings can cause asphyxiation. 

Because of their outside work, farmers may be affected by extreme heat or cold, suffering heat exhaustion or frostbite. Extensively exposed to the sun, farmers also have higher rates of skin cancer and melanoma than the general population. In addition, thousands of farmworkers are affected each year by occupational infections (Klingman, Peate, & Cordes, 1991) acquired from working with soil, animals, and wastes, resulting in illness and disability. 

All of these exposures can be prevented through proper work practices and use of personal protective equipment. The NIOSH program, described above, includes interventions related to farm exposures (Myers, 1992). As noted, nurses are actively involved in case finding and developing targeted individual and community education interventions (e.g., through Hazard Alerts, site visits, and booths at county fairs). No assessment of the program's effectiveness has been published to date. 

Motor Vehicle Accidents. Motor vehicle accidents are a major cause of injury and death in urban and rural areas. In rural areas, accidents may be complicated by several important factors: the length of time before an accident is discovered, the time needed to transport an accident victim to a health care facility, the expertise of providers in treating major trauma, and the availability of supportive equipment. 

In the United States, about 47,000 deaths in 1990 were due to motor vehicle injuries to passengers and pedestrians, and nearly 40 percent of all deaths among persons 15 to 24 years old were caused by motor vehicles (McGinnis & Foege, 1993). Fatalities due to motor vehicle accidents for both passengers and pedestrians are higher in rural areas than in urban areas. The U.S. death rate of occupants of motor vehicles was 18.7 per 100,000 population, varying from a low of 2.5 per 100,000 in Manhattan to a high of 1,456 per 100,000 in Loving County, Texas. All of the 15 counties with the highest death rates had sparse populations (two or fewer persons per square mile) (Baker, Whitfield, & O'Neill, 1987). The reasons suggested for the higher fatality rate in rural areas relate to higher speeds of travel and variations in road characteristics, lack of seat belt use, types of vehicles, and limited availability of emergency care (Baker et al.). 

Protective equipment can significantly reduce deaths and injuries. Lap and shoulder belts reduce risk of death by 45 to 65 percent and risk of serious injuries by 40 to 55 percent; child passenger restraints reduce deaths by 50 to 90 percent; and helmets reduce motorcycle fatalities by 30 percent and serious head injuries by 75 percent (Campbell, 1987; Smith & Falk, 1987). 

Firearm Injury. In 1990, use of firearms across the United States resulted in more than 36,000 deaths and about 16,000 homicides, 19,000 suicides, and 1,400 unintentional injuries (McGinnis & Foege, 1993). Firearms, which cause both unintentional and intentional injuries, are dangerous not only in urban areas, but also in rural communities, which have very high morbidity and mortality rates resulting from firearm injuries. 

Baker (1992) reviewed urban/rural death rates for unintentional injuries, comparing central cities, metropolitan areas with a population of more than 1 million, metropolitan areas with a population less than 1 million, nonmetropolitan areas, and remote rural areas. According to 1986 data, the death rate from unintentional shootings in nonmetropolitan areas was about 2.5 times the central cities' rates, and the death rate in remote rural areas was 4 times the rate in central cities. The suicide rate was 1.5 times higher in rural areas than in central cities, while the homicide rate was more than 2 times higher in central cities than in rural settings (Baker, 1992). While not all suicides and homicides are the result of the use of firearms, a high proportion are. 

The literature on firearm injuries is largely descriptive. No published studies on the prevention of these injuries in rural settings through community-based interventions were identified by the panel. 

Other Contributors To Mortality 

Rural and urban residents both participate in behavior patterns that contribute to more than one half of the deaths in the United States each year. In quantifying these major nongenetic factors contributing to death in the United States, McGinnis and Foege (1993) show that about one-half of all deaths can be attributed to nine factors, all affected by behavior. For both rural and urban residents, these nine contributors to mortality, in descending order, are: tobacco use, diet and activity patterns, alcohol use, microbial agents, toxic agents, firearm use, sexual behavior, motor vehicle accidents, and illicit drug use. 

McGinnis and Foege (1993) acknowledge that socioeconomic status and access to care also are significant contributors to mortality, but they were unable to quantify these factors. Lack of access to care was addressed in the Carter Centers' project, Closing the Gap; this study's estimates indicate that lack of access to screening and preventive interventions accounts for 7 percent of premature deaths and 15 percent of potential years of life lost before age 65 (Amler & Eddins, 1987). 

All of the contributors to mortality are amenable to change through community-based health care interventions. Some interventions, such as those incorporating nursing's expertise in promoting behavioral change, have been designed and tested in rural settings, but systematic assessment of outcomes and comparison of results across settings are sparse. Specific data for rural residents and comparisons between rural and urban areas also are lacking and difficult to obtain. General data on the nine contributors identified by McGinnis and Foege (1993) are summarized briefly below for the entire U.S. population. 

All of these contributors to mortality can be changed, primarily by increasing and enhancing preventive behaviors. Disease prevention and promotion of behavior change have long been a dominant focus of nursing practice and are being appropriately continued and enhanced through empirical research. Examples of the various models developed and applied to combat these health problems are described in the next section. 

Primary and Secondary Prevention Strategies 

As noted in chapter 1, community-based strategies of health care incorporate several key components. Some strategies have a specialized focus, such as prenatal care, whereas others adopt a broader approach, such as lowering cholesterol levels throughout the community. Needs assessment, health priorities, interventions, and evaluation plans are developed locally in partnership with community members and incorporating the needs of the population served. An active effort is made to engage all persons in the target group through formal and informal communication networks naturally occurring in the community (e.g., community organizations, ethnic groups). 

These networks may include lay outreach workers from the community who have proven to be especially helpful in ensuring successful interventions in a variety of settings (Mahon et al., 1991; Meister, Warrick, de Zapien, & Wood, 1992). Working as volunteers or paid employees of public agencies, lay workers are trained to work in partnership with professional health care personnel. Involvement of the local population in health care planning and decision making (e.g., serving on community health councils, developing health care plans in collaboration with providers) also is essential and requires time to elicit changes in beliefs and behavior related to local control, influence, authority, and responsibility for health care. 

In one effort, Phillips and Ferketich (personal communication, June 13, 1995) worked with rural residents to organize a legislative hearing to obtain public testimony on the need for county and state funding of the residents' health care. The researchers helped community members and lay outreach workers learn how to organize hearings, and residents practiced giving testimony. A translator was obtained for residents who did not speak English. Through this process, the residents gained confidence and control over "their" legislative hearing and increased their participation in resolving their health care problems. 

A number of community-based strategies in health promotion and disease prevention are being developed and applied today in rural or urban settings. The NINR is contributing to these efforts through support of several interventions that are being tested in, or have applicability to, rural populations. 

Two NINR-supported studies are addressing prenatal care. In one study, aimed at refining, implementing, and testing a community-based nursing practice strategy, investigators are complementing standard prenatal services with community outreach to increase access and retention using culturally sensitive strategies congruent with the style of Hawaiian, Filipino, and Japanese residents on the island of Hawaii. In a second study, an intensive outreach case management strategy is targeted to pregnant Native American and Latino women residing in rural Oregon, for the purpose of improving the birth outcomes of their infants. 

Two additional studies are focusing on cancer among rural residents in the South. In one of these studies, nurse researchers are investigating the effect of four different nursing interventions on colorectal screening behavior among economically disadvantaged African American and white elderly persons. In the second study, investigators are implementing a nursing intervention aimed at primary prevention of colon cancer through increased fiber intake. For this study, African Americans aged 45 to 75 with lower to upper-middle socioeconomic status are being recruited from local churches and the surrounding community. The researchers will assess adherence outcomes and develop a composite adherence marker (Atwood et al., 1994). 

Another project is designed to improve dietary management of hypercholesterolemia, a risk factor for coronary heart disease, using public health nurses in county health departments who serve primarily low-income and minority clients. In this randomized controlled trial, nurse researchers are testing the effectiveness and feasibility of a structured dietary assessment and intervention program the Food for Heart Program. The NINR also continues support for a randomized field trial to test the effects of interventions designed to reduce cardiovascular disease risk factors in children attending 21 rural and urban North Carolina elementary schools. 

Despite these initiatives, however, there is a paucity of research directed specifically toward community-based health care interventions for rural populations. Most intervention programs that are being developed are designed to include close involvement of community members, active outreach efforts, and plans to decrease barriers to service utilization (i.e., availability, accessibility, acceptability, affordability, appropriateness, and adequacy of services). Most programs also include empowerment of community members through assessing health needs, helping to structure health care delivery, and participating in health care and health promotion strategies. 

However, even though models of community-based health care interventions exist, they are infrequently tested. Demonstration projects are showing promise, but only descriptive findings are reported. The full effect of true and complete involvement of the community in designing and implementing these intervention models is not known. It is anticipated that these models will have the structural underpinning to allow for examination of them across communities. Further research is needed to assess this applicability across communities and/or populations and to compare model effects. Thus, while key health problems are well documented, the most effective method(s) of intervening to alter health behaviors to prevent these problems remains ill defined. 

Highlighted below as examples are several assessment models that have been used successfully as a basis for community interventions in primary and secondary prevention; an intervention model, currently being implemented, that incorporates a community-based nursing approach; and several models that are potentially useful for rural populations and are focused on specific health problems. The models described here address three care functions identified in an Institute of Medicine report (1988): assessment, policy formation including intervention, and assurance of the population's health. 

Assessment Models 

Before a community-based intervention is planned and implemented, a thorough community assessment must be conducted to ensure that the intervention is appropriate, acceptable, and adequate to meet the needs of the community or population group. Participation of community members in this assessment is essential and will promote their interest and involvement in the proposed intervention. Several assessment models are described below to convey the state of research in this area. Some of these could be applied in either rural or urban settings. 

The Public Health Nursing Conceptual Model (White, 1982), also known as the White Model, is designed specifically to guide public health nursing practice from assessment through assurance of a good level of health (Institute of Medicine, 1988). The model, applicable to primary, secondary, or tertiary prevention, includes a core that contains scope of practice (groups, families, communities, populations), practice priorities (prevention, promotion, and protection), and practice interventions [educating, engineering (including social engineering), and enforcing (as in communicable disease control)]. Surrounding this core are unifying "public health nursing dynamics" that comprise the nursing process, with valuing included at each stage. Valuing is "the process of assigning or determining the worth or merit of something" (White, p. 529); for example, the public's health may be more or less valued than an individual's desires. 

Four determinants of health are also included in the model: human/biological, environmental, medical/technological/organizational, and social. These determinants offer intervention points for health care professionals (e.g., teaching the dangers of high blood pressure's impact on human/biological determinants using an educa- tional intervention; helping a group decrease the incidence of sexually transmitted and other diseases). 

In addition, the model includes the key components of a community-based health care strategy: population level assessment, community involvement, intervention at appropriate level(s), and evaluation in an ongoing, recycling process. Outcomes of the interventions accrue for individuals, families, groups, communities, and populations. The White model is found in some public health nursing and community health nursing texts and has been effective, for example, in community interventions to reduce exposure to secondary smoke (Stotts, 1991). 

The PRECEDE-PROCEED Model (Green & Kreuter, 1991; Green, Kreuter, Deeds, & Partridge, 1980) follows the nursing process in a linear progression from assessment through evaluation (Swanson & Albrecht, 1993). The acronym PRECEDE stands for the main components of the nursing process: predisposing, reinforcing, and enabling constructs in educational diagnosis and evaluation (Swanson & Albrecht, p. 175). The acronym PROCEED, added later to the model by Green & Kreuter (1991) to accommodate all aspects of health promotion, stands for: policy, regulatory, and organizational constructs in educational and environmental development. This model is applicable to any intervention, including tertiary prevention, that involves an educational component. 

This model also incorporates community involvement at all phases: social diagnosis, including consideration of the quality of life of the community and social implications of both the problem and the potential solutions; epidemiological diagnosis to develop an aggregate data base; behavioral diagnosis of individual and community health behaviors potentially contributing to the particular health problem; diagnosis of community needs and prioritization of these needs, including examination of predisposing, enabling, and reinforcing factors contributing to the problem; identification and prioritization of the most amenable interventions at individual and community levels; intervention; and evaluation. The elements of community-based interventions are clearly apparent in this model in relation to diagnoses, potentials for partnership, "universal coverage" of persons affected, outreach, community involvement, empowerment, and reduction of barriers. 

This model is well constructed to implement the Institute of Medicine's (1988) recommended core public health functions of assessment, policy formation, and assurance. It is designed for interventions that include an education component (Green et al., 1980) and is applicable to many intervention settings (e.g., health promotion and disease prevention activities related to HIV and AIDS, teenage pregnancy, weight control, physical fitness). 

The model has been widely used for more than a decade by community-based social scientists and health services personnel, including nurses. For example, Selby, Riportella-Muller, Sorenson, & Walters (1989) applied the model to public health nursing practice research on disease screening. Reports of tests and uses of the model are most commonly reported in journals such as Health Promotion and Disease Prevention and the American Journal of Public Health, among others. 

The University of Colorado, School of Nursing-Project Community Assessment (formerly called Project GENESIS) is a model particularly applicable to primary health care and an essential precursor to planning community-based health care. The community analysis involves community members, focuses on a population or whole community, and leads to discovery of factors important to primary and secondary prevention. 

This model has been used by advanced practice nurses, students, and faculty in one graduate degree program to assess the health status and needs and recommend interventions in more than 15 Colorado communities (Stoner, Magilvy, & Schultz, 1992). As described in a series of articles (Barton, Smith, Brown, & Supples, 1993; Magilvy, McMahon, Bachman, Roark, & Evenson, 1987; Schultz & Magilvy, 1988), the model combines quantitative and qualitative methods, defines health broadly, and recognizes that community health is influenced by a variety of factors (e.g., spiritual, physical, environmental, recreational, educational, health care system). 

The steps of the model include data gathering and secondary analysis of existing health data (e.g., epidemiological, census) about the community; identification and interviewing of key community members; and ethnographic participant observation and interviewing of persons living and working in the community. The community's strengths and weaknesses, as well as recommendations, are derived from ethnographic analysis and primary and secondary data synthesis. 

This model is useful for assessing the health of many types of communities or of subgroups within a community (Magilvy et al., 1987; Stoner et al., 1992). Although most of the 15 communities surveyed have been rural, the model is equally useful in urban settings. The unique combined methodology and "grass-roots" involvement of community residents have contributed to the success of this model. Communities involved in previous projects have found that the data and reports generated are useful in grant writing and implementation of primary and secondary prevention programs. Application of qualitative research methods to community analysis places this model on the "cutting edge" of assessment. It may also have applicability to tertiary prevention programs addressing chronic illness and long-term care. 

Community-Based Nursing Interventions 

The intervention described below demonstrates the usefulness of community-based nursing strategies for rural populations. Focused on improving the health status of rural Hispanic communities, this example highlights the type of community-based health promotion and disease prevention strategies under way. 

Improved Health for Rural Hispanics. In this study, initiated in 1990, Ferketich, Phillips, and Verran (1990) are testing a comprehensive, multilevel, community health nursing model for rural Hispanics. The investigators are comparing longitudinal data within and between two populations: Mexican Americans and white, non-Hispanics residing in local communities. 

The three components of the model are: (a) personalized preventive care for individuals and families through nursing-based clinics ("Las Clinicas de los Pueblos"); (b) outreach, identification, and followup with community members by teams consisting of a community health nurse and lay care workers ("promotoras"); and (c) community empowerment via teams comprised of a community health nurse and "promotoras." In order to examine the effectiveness of these components singly and in combination with other components, communities are assigned different components of the model. 

During the first year, researchers conducted community assessments and gained entre into the communities. The assessments were developed based on information obtained in extensive formal and informal interviews of Mexican Americans and white, non-Hispanics in the communities. A baseline health survey was constructed using the Andersen and Aday framework (Aday & Andersen, 1984). Community meetings were held to negotiate the exact wording of survey items and to gain residents' commitment to the data gathering process. The survey was administered by bilingual interviewers to a random sample in each community; the response rate averaged above 75 percent across the communities. Community members were then involved in planning the interventions, which were initiated in the second year. 

In this study, all "promotoras" and community health nurses were from local areas they served, and all were bilingual except two (one promotora and one community health nurse) who were fluent in Spanish. Community advisory boards provided guidance on the provision of culturally congruent care and on overall activities of the nurses, including hiring. 

About 7,000 patient encounters have occurred in the 39 months since Las Clinicas de los Pueblos opened. Each clinic is open 1 day per week. The total population served by the demonstration program is approximately 8,000. A "one-stop-shopping" approach to health care is used, allowing clients and families to access, during a visit, any of the clinic services available. For example, a mother who comes to the clinic with several children can be treated for her health problem and, at the same time, obtain family planning services as well as immunization for the children. WIC and other providers and Las Clinicas personnel cooperate in providing comprehensive care without duplicating existing programs. 

In this study, outreach teams have been very successful in finding cases, providing health education, and following up with clients. The community component was focused on facilitating the development of a community board to direct and monitor the continuation of services after the end of the investigators' research grant (Ferketich, Phillips, & Verran, 1990). 

A second survey revealed initial changes in the outcome variables measured. For example, indicators of health promotion and disease prevention showed marked improvement after 18 months of the intervention. Positive, statistically significant changes were noted in immunizations and utilization of screening procedures, such as mammograms; pap smears; and blood pressure, cholesterol, and glucose measurements. The percentage of immunizations, for example, among the population needing this intervention increased from 47 percent to 79 percent. 

Long-term results of outreach and empowerment have resulted in the opening of two new clinics at the end of the granting period. Both clinics are managed by nurses; one is located in a renovated building in one community and the second is located in a high school. A community board has been formed to oversee community access to and utilization of the health care services offered. 

Other Potentially Useful Strategies 

A number of other preventive strategies have been developed and applied in different settings. Shown to be effective in the specific sites studied for the populations and problems targeted, these strategies may be more generally applicable in both rural and urban communities. Examples of primary and secondary prevention strategies, in some of which nurses and nurse researchers have played significant roles, are described below. 

Tobacco Use and Other Health-Compromising Behaviors. Nursing interventions to promote smoking cessation and reduce the risk of secondary smoke are exemplars of primary prevention activities. Stotts (1991) used the White Model (White, 1982), which he calls the Salmon Model, reflecting White's change in name, in two cities to guide effective public health nursing intervention to decrease the risk of secondary smoke to the population through the designation of "no-smoking" areas. A parallel process could well occur in a rural setting. 

Another model, the Stages of Readiness for Change Model (Prochaska, DiClemente, & Norcross, 1992), has been used to maximize the effectiveness of interventions to promote smoking cessation as well as other healthy behaviors. For example, Campbell et al. (1994) used the stages in this model to predict more and less successful adherers to a change in dietary behavior among a random sample of adults. 

Pender's Health Promotion Model (Pender, 1987), a modification of the Health Belief Model, has been tested, at least in part, and found to be effective in predicting healthy behaviors, for example, in health promotion (Laffrey & Isenberg, 1983). Also, Lusk et al. (1994) identified key factors in auto plant workers' use of hearing protection to reduce their risk of hearing loss. The Pender model is targeted ostensibly toward individual behaviors, although social support is clearly an element. However, Pender (1987) and others (Waller, 1994) argue that interventions targeted toward high-risk individuals need to accompany macro-level, community-based interventions because these latter interventions alone are not consistently effective over time. 

Elements of another, non-nursing, model, Hirschi's control theory (Hirschi, 1969), have been shown to be related to adolescent cigarette smoking (Foshee & Bauman, 1994; Foshee, Bauman, Kock, Haley, & Downton, 1989; Krohn, Massey, Skinner, & Lauer, 1983), drug use (Burkett & Jensen, 1975; Kandel, 1978), and sexual behavior (Foshee & Bauman, 1992). In addition, Jemmott, Jemmott, Spears, Hewitt, and Cruz-Collins (1992) have successfully applied the Social Cognitive Theory of Planned Change in an intervention among inner-city black adolescent women to increase their self-efficacy and favorable hedonistic expectancies and to promote use of condoms in preventing AIDS. The strategy remains to be tested in a rural setting. 

Unhealthy Diet and Activity Patterns. To address factors contributing to cardiovascular disease and diabetes, the Colorado Action for Healthy People program uses a combination of primary health care techniques to assist communities in developing community-wide intervention programs (Hill, 1994). Through activities such as community health assessment, technical assistance, training, and grantwriting assistance, this program has helped rural and urban Colorado communities tailor community-specific interventions to their needs. For example, "Project New Self," a Denver heart disease prevention program, addresses weight loss through nutrition education, counseling, food samples, and support groups (Hill, 1994); used successfully with urban populations, this program has potential applicability for rural groups as well. 

In another series of studies, Whitehead (1984, 1992) examined culturally sensitive interventions in African American and European American households using better nutritional habits to promote health. He found that the most effective interventions incorporated knowledge of primary foods and preparation styles congruent with cultural practices. 

Agricultural Illness and Injuries. As noted above, nurses are providing surveillance, case findings, and interventions through a NIOSH program under way in 10 states (Connon, Freund, & Ehlers, 1993). Locally based agricultural occupational health nurses are working closely with health departments, physicians, and other providers, as well as agricultural community and organization programs, to provide meaningful, community-based interventions targeted to agricultural illness and injuries. 

Cardiovascular Risk. Religious organizations are a focal point for community interventions with ethnic groups, such as African Americans. For example, church-based programs have been successful in reducing cardiovascular risk, especially among southern black church members. In the Fitness Through Churches program, interventions (blood pressure measurement, education, and aerobic exercise) involving the community were delivered effectively through the social structure of black churches in North Carolina (Hatch & Voorhorst, 1992). 

Nurses were pivotal in another effective community intervention aimed at mobilizing black churches in Maryland to reduce hypertension. Church nurses were recruited and effectively trained to monitor the blood pressure of persons at risk, provide counseling, and make appropriate referrals (Hatch & Monnett, 1993). 

Using the Ecologic Well-Being Model, Ruffing-Rahal (1994) also substantiated positive results in a group-based health promotion nursing intervention among low-income African American women over age 65 (mean age, 77). This intervention was associated with protecting the women from decreasing their health practices and perceiving lower well-being over time. Lastly, an intervention based on a Holistic Model of Client-Centered Nursing Practice, and centered in a church-run neighborhood house, was successful in decreasing blood pressure and blood glucose levels in a small Appalachian community (Porter & Howard, 1986). 

Despite these efforts, multilevel community strategies are not consistently effective. The large cardiovascular risk reduction trial sponsored by the National Heart, Lung, and Blood Institute suggests caution. As part of this trial, the Minnesota Heart Health Program, which was implemented at individual, group, and community levels using a variety of theories, demonstrated only a modest impact in terms of size and duration of risk reduction, which was, for the most part, no greater than secular trends in this highly publicized health promotion effort (Luepker et al., 1994). The Stanford Five-City Project had the same difficulty (Fortmann, Taylor, Flora, & Winkleby, 1993). 

Breast Cancer. To promote early detection of breast cancer, the lay health advisor model (Eng & Hatch, 1991; Eng & Young, 1992), with black interveners, is being used successfully in the Save Our Sisters Study (Tessaro, Eng, & Smith, 1994) to encourage women in rural North Carolina to obtain routine mammograms. Early findings show that individuals diagnosed with breast cancer seek advice on treatment issues from lay health advisors. This result has led to efforts to establish a network of retired professional nurses who advise clients on health care access, treatment interpretation, social concerns, and other issues. 

Colon Cancer. Weinrich (1990) used Orem's Theory (Orem, 1991) to promote self-care by nursing clients, determining predictors of participation in fecal occult blood test (FOBT) screening for colon cancer among older adults in rural South Carolina communities participating in a meal program. The study showed that predictors of self-care (FOBT screening) included functional ability and being a woman. It also showed that more stool samples were returned when the intervention included lay health advisors and materials adapted for elders (Weinrich & Boyd, 1992; Weinrich, Weinrich, Boyd, Atwood, & Cervenka, 1994). 

Summary

Primary and secondary prevention relate mostly to health promotion and disease prevention: preventing disease before it begins, diagnosing and treating it early, and promoting health and well-being to better prepare individuals to fend off disease or injury when exposed. The nation's health statistics for rural populations and U.S. goals outlined in Healthy People 2000 (U.S. Public Health Service, 1990) point to three key areas of intervention (maternal and child health, uninten- tional injuries, and the major contributors to mortality tobacco use, diet and activity patterns, alcohol use, microbial agents, toxic agents, firearm use, sexual behavior, motor vehicle injuries, and illicit drug use) (McGinnis & Foege, 1993). 

To meet challenges in these areas, several community-based models have been developed and applied by nurses for intervening unilaterally or interdisciplinarily within nursing practice. Positive outcomes have been demonstrated, and nursing has contributed to health promotion and disease prevention in various settings such as health departments, rehabilitation facilities, and nursing centers (Barger, 1991). 

These models and strategies, however, have not been well tested, compared, or assessed rigorously in rural communities and across sites and populations. Studies to evaluate such models and strategies need to address a variety of perspectives, including degree of specificity of the model components, sensitivity of outcome measures to the model components studied, timing of data collection points, and duration of implementation. Also, influences of specific economic factors have rarely, if ever, been taken into account, and few cost or cost-effectiveness studies of various community-based models are known to have been reported (e.g., Ho et al., 1991). Laying the groundwork for addressing these cost issues is important. 

Tertiary Prevention: Supportive and Restorative Care 

Tertiary prevention is a necessary component of community-based health care services in rural areas. Aimed at minimizing disease and the disabling effects of acute, chronic, or terminal illness and accidents or disability, tertiary prevention includes supportive and restorative care services. These services are provided increasingly in community-based settings, such as homes, rehabilitation centers, community health clinics, and primary care centers, and are delivered by public and home health nurses, hospice agencies, and multidisciplinary rehabilitation teams. 

Integration of services is a growing trend, with existing health care institutions or agencies working together to meet identified needs. For example, nursing homes and small rural hospitals are providing home care and hospice services in their local or regional geographic areas. Tertiary prevention can offer rural Americans an improved quality of life over a longer lifespan and can influence and decrease the cost of health care in rural areas by limiting hospitalization and the need to move clients to urban care centers. 

In this section, the panel describes the state of research on tertiary prevention in relation to community-based health care strategies for rural populations. Similar to the previous section, two major topics are addressed: key areas of research, and tertiary prevention strategies. 

Overall, research and models of tertiary prevention for rural populations are limited, and little systematic research has been conducted on supportive and restorative care. Much of the literature on supportive and restorative services is not specific to rural communities, with the consequence that knowledge on the uniqueness or differences in tertiary prevention in rural areas is limited. In addition, research on supportive and restorative care is usually multidisciplinary, with nursing research making strong contributions in home care, family caregiving, and hospice care. There are few well-tested models of tertiary prevention interventions. In this section, the panel highlights several models as examples of current research in this area. 

Key Areas of Research

Many rural Americans face a wide range of chronic illnesses and lifelong disabilities. Disabling sequelae of occupational exposures, hazards, accidents, injuries, and postacute illnesses are a major problem in rural America. As Ran- dall (1993) notes, the nation's two most dangerous occupations are farming and mining, both of which are primarily rural occupations. The problems unique to these occupations (e.g., machinery accidents, sun exposure, exposures to chemicals and pesticides or mineral dusts and ores) may require long-term rehabilitative care, cancer treatment, and respiratory assistance. 

Rural populations also have high rates of chronic illness, such as hypertension and cardiovascular diseases, and many rural residents experience terminal or life-threatening illness as well (Bushy, 1993; Coward, Duncan, & Freudenberger, 1994). And, as with other population groups, the incidence of chronic illness increases with age, making elderly persons who reside in rural areas particularly vulnerable. Other groups, such as Native Americans, Alaskans, Native Hawaiians, migrant workers, southern African Americans, and rural homeless face additional health problems related to poverty, loss of their homes and farms, and a migratory lifestyle (Bushy, 1993). Mental illness and stress-related disease are common among rural populations, although the precise incidence and prevalence of these problems have not been established (Bushy, 1993). 

Further complicating this health picture is the fact that rural adults often continue to work despite illness or injury (Bushy, 1992; Lee, 1993) and they are less likely to engage in preventive behaviors, increasing their exposure to risk (Bushy, 1993). Specific risky behaviors include not wearing seat belts; smoking; not having regular blood pressure checkups, pap smears, or other screening tests; and not performing breast self-examinations (Bushy, 1993) all of which can, ultimately, lead to an increased need for supportive and restorative care. This wide array of chronic, terminal, postacute, or mental illnesses or accidents and their disabling sequelae indicate a need for community-based prevention services in rural America. 

Despite these needs, however, the coordination, management, and delivery of supportive and restorative care for rural Americans have fallen victim to demographic, geographic, economic, and human forces. For example, as noted previously, rural residents have fewer available health care services, often reside or work far distances from available health care providers, and usually do not have access to public transportation factors that impede health care in general and, especially, long-term, chronic care. 

The limited number and distribution of health care professionals often result in a lack of continuing, followup care for individuals in need, making them dependent on an informal support network of friends and family who may not be available or able to provide appropriate care. In addition, rural residents, especially farm residents, may have to pay enormous sums for insurance because they are not part of a larger purchasing network in which risk is spread over a large, heterogeneous group. 

Nurse researchers and their colleagues in public health, health administration, sociology, anthropology, and other disciplines have recently become more actively engaged in investigating the health needs and services in tertiary prevention for rural populations. Described below is the state of research on two important aspects of community-based tertiary prevention: chronic illness, and formal and informal support (including family caregivers). Other aspects of tertiary prevention receive little attention in the research literature. 

Chronic Illness 

Most researchers have attended to two age groups when studying chronic illness in rural America: older adults and middlescent, or middle-aged, adults. Research on health problems, disability and functional status, life with chronic illness, organization and delivery of health services, and community-based care strategies specific to rural populations is limited. 

Older Adults. Researchers addressing rural-urban differences in health services often focus on frail or chronically ill older adults because data are available for comparisons. However, rural nursing research on this population is very limited, and insufficient research exists to serve as a knowledge base for rural nursing practice (Weinert & Burman, 1994). However, research from a variety of other disciplines gives insight into some of the problems of frail or chronically ill rural elders. 

Persistent disadvantages have been reported for rural populations compared with urban populations (Himes & Rutrough, 1992; Krout, 1994; Weinert & Burman, in press). For example, rural elders have a more restricted range of available services (Hassinger, Hicks, & Godino, 1993) and there is greater variability in the health services developed for elders (Salmon, Nelson, & Rous, 1993). The fewer number and poor distribution of health care providers result in a lower utilization rate for services (Wallace & Colsher, 1994). Further, rural older adults have to travel over long distances for care and have longer waiting times once they arrive at a health service provider (National Center for Health Statistics, 1993). 

Gaps also are reported in home-based care for rural older adults (Buehler & Lee, 1992; Burman, Steffes, & Weinert, 1994; Congdon & Magilvy, 1995; Magilvy et al., 1994). Evidence shows that access to home health services is less for rural than urban elders due to lower availability, attitudinal and behavioral characteristics, or decreased referral patterns (Redford & Severns, 1994). Rural home care and public health nurses have reported problems with limited physician referral to home care, lack of knowledge by community members of home care resources, and limited eligibility requirements for covering care for older patients as factors inhibiting home care utilization (Magilvy et al., 1994). Excessive documentation and paperwork tied to reimbursement requirements also have been cited by rural home health nurses and physicians as presenting barriers to efficient delivery of home and primary care services for older adults (Congdon & Magilvy, 1995). 

A previous NINR Priority Expert Panel report, entitled Long-Term Care for Older Adults: A Report of the NINR Priority Expert Panel on Long-Term Care (National Institutes of Health, 1994), describes in depth the chronicity and related health care needs of older adults. The present panel supports the discussion and recommendations in this report and notes that most of the issues addressed which pertain to the organization and delivery of long-term care for older adults (e.g., mobility and functional status, personal care, formal and informal home care, nursing home care, family caregiving, long-term care transitions) apply to rural, as well as urban, elders. Interventions recommended in this document also are relevant. Although community-based strategies are not specified, the strategies suggested could be tested in community-based models and compared among rural and urban populations. 

Middlescent Adults. Research findings on the health status and level of chronicity of rural middlescent adults (generally defined as ages 40 to 65, although this range is sometimes broadened) are conflicting. While rural dwellers are reported to experience more long-term illness and more disability than urban dwellers (NACHC & NRHA, 1988, 1989), members of the rural middlescent population are reported to be healthier than their urban counterparts (Eggebeen & Lichter, 1993). 

Differences are noted in health status and perceptions of health among rural adults living on farms and ranches, in small towns, and in different geographic areas; those living in small towns experience the poorest health (Long & Weinert, 1992). Poverty, harsh economic conditions leading to the loss of farm ownership, limited health insurance coverage, and health prob- lems unique to rural areas, as described above, may lead to increased chronicity, disability, and depression or other mental health problems. Community-based health care strategies can help resolve these problems and should be included as an important part of more comprehensive solutions. 

Lee (1993) examined the health perceptions of rural middle-aged (30-50 years) and "new middle-aged" adults (51-69 years) in a sample of 162 adult respondents. Findings indicated that rural health care professionals worked effectively with people in the farming/ranching culture by recognizing the impact of the seasonal work cycle on the timing of care services. Lee sug- gested that health professionals give more emphasis to chronic illness programs, following the seasonal growing patterns of agricultural crops in the area so that middle-aged adults could take advantage of health services. Rural nurses who are part of the community and who are sensitive to work patterns should be involved on a formal and informal basis in presenting programs and information on health, especially to support rural adults when major health crises occur or when consequences of destructive health behaviors result in declining health status (Lee, 1993). 

Transition care presents an even greater problem for rural residents with chronic illness, disability, and continuing care needs. Research is lacking on the effectiveness of current protocols for discharge planning and other transitions in all areas of care. Knowledge about the process and patterns of delivery and receipt of rural home care or family and community support is limited; research on the continuity of care and health care transitions in rural populations also is scant (Magilvy et al., 1994). 

Coordination and management of supportive or restorative care provided by formal health care organizations and by informal family and support networks so critical in rural areas are limited. Little research has been conducted on the role and effectiveness of informal family care and the relationship to use of formal care (Buckwalter, Abraham, Smith, & Smullen, 1993; Given & Given, 1991, 1994; Hall et al., 1995; Weinert & Long, 1993). Few systematic investigations have been undertaken of collaborative efforts to link formal health care services, such as home care, with informal support provided by family, friends, and community. Reports of research on the quality of informal care also are lacking. Some limited, recent research findings are described below. 

In addition, few reports have been published of studies examin