kristoff@GENBANK.BIO.NET (Dave Kristofferson) (09/23/90)
DEPARTMENT OF HEALTH AND HUMAN SERVICES Public Health Service Alcohol, Drug Abuse, and Mental Health Administration National Institute on Alcohol Abuse and Alcoholism Office for Substance Abuse Prevention Request for Applications RFA: AA-91-01 COMMUNITY-BASED RESEARCH ON THE PREVENTION OF ALCOHOL-RELATED PROBLEMS Application Receipt Date: February 12, 1991 P.T. 34; K.W. 0404003, 0745027, 0403004, 0404000 (Catalog of Federal Domestic Assistance No. 13.273) INTRODUCTION The National Institute on Alcohol Abuse and Alcoholism (NIAAA) and the Office for Substance Abuse Prevention (OSAP) are sharing sponsorship of research in the area of community-based intervention trials for the prevention of alcohol-related problems. NIAAA promotes prevention research on a broad range of populations, including the high-risk youth groups and young adults of special interest to OSAP. NIAAA is especially experienced in the design, review, and monitoring of outcome-oriented research, while OSAP has focused on the implementation of community intervention programs and service demonstration projects. These convergences of interest and expertise make it mutually advantageous for NIAAA and OSAP to combine resources to encourage and facilitate community-based research of an appropriate scale. This joint Request for Applications (RFA) describes the area of community-based prevention research, indicates directions of needed research, and discusses some design considerations. The procedures for submission and review of grant applications and the terms and conditions for grant support are given below. Applications received in response to the RFA will be assigned for review and funding consideration in accordance with established Public Health Service (PHS) guidelines. The statutory authorities for these grant awards are sections 301 and 510 of the Public Health Service Act (42 USC 241 and 290bb) and section 508(b)(10) of the Public Health Service Act (42 USC 290aa6(b)(10). This is a collaborative effort, jointly funded by NIAAA and OSAP. BACKGROUND A. The Social and Economic Costs of Alcohol-Related Problems Alcohol use and abuse play a major role in numerous medical and social problems in the United States. Alcohol problems do not result only from alcoholism. The adverse consequences of alcohol use arise as a result of single episodes of drinking, episodic drinking, persistent alcohol abuse, and alcohol dependence. These consequences can affect not only the drinker but the drinker's family, friends, and associates, as well as others with whom the drinker may come in contact. Moreover, the cost to society of the adverse outcomes of alcohol use is high. In addition to alcohol-involved motor vehicle crashes, alcohol use and abuse have been linked to other types of accidental injuries and fatalities, including drownings, falls, and burns. Individuals with alcohol-related problems require more general health care, may be less productive at their jobs than individuals who do not abuse alcohol, and are also over-represented among criminals and persons who commit suicide. The negative impact that these and other alcohol related problems have on the quality of life of society is paralleled by considerable economic cost. B. The Public Health Model The growing body of epidemiological and etiological studies on alcohol use and abuse has led researchers to see that alcohol problems arise through a complex interaction of individual, interpersonal, and social factors. This perspective has been expressed in the public health model, a conception of the causation of alcohol-related problems that specifies three major elements acting together either to produce or attenuate specific problems. These elements are: o the agent -- alcohol beverages or ethanol itself; o the individual (host) -- traits that affect a person's susceptibility or vulnerability to the effects of alcoholic beverages; and o the environment -- physical, interpersonal, or social milieu surrounding the use of alcohol that either regulates the individual's exposure to the agent or mediates the risk that the agent poses to the individual.[1] The strength of the public health perspective is that it directs attention to interactions between the individual and the environment that result in a specific problem. It facilitates the development of promising interventions by identifying different elements in the etiology of alcohol problems to which the interventions may be directed. It also helps one visualize points that may be only peripherally related to the etiology of a specific problem but that nevertheless can be effective opportunities for interventions. "Whether bars and taverns cause someone to become intoxicated, for example, is not as important as whether they can be designed or modified to prevent intoxication."[2] The public health perspective thus provides guidance for the development of multifaceted, integrated prevention strategies that more closely match the complexity of alcohol problems and the heterogeneity of drinkers and drinking situations. C. Community-based Prevention Programs Research in varied fields of health and illness has indicated that comprehensive prevention programs informed by the public health model can achieve significant results when they are implemented so as to involve total communities. In the area of cardiovascular disease (CVD), systematic intervention programs targeted at reducing cigarette consumption, controlling high blood pressure, and modifying unhealthy eating habits have resulted in community-wide health risk reductions that can be linked to a lowered incidence of disease and early death. In general, community-based CVD prevention programs are characterized by the following features: o the target populations usually consist of a few geographically or administratively defined communities; o the chief prevention strategy is to change behavior through information and attitude change; o the chief interventions are directed at entire populations rather than selected groups or individuals; o the health interventions are incorporated into the existing health services system and economic structure of the community; and o the interventions are based on community involvement and participation and take advantage of existing social networks.[3] Brief descriptions of some of the more influential studies appear in Appendix A. Several important strategic conclusions have been drawn from these community studies. These are: o it is possible to change the health habits of entire communities and to mobilize community resources to achieve these changes; o it is possible to transfer responsibility for program maintenance to local organizations; and o community-based intervention programs offer the possibility of cost-effective replication of programs in other communities.[4] It should be noted, however, that the results of these community-based studies must be interpreted with caution because of various weaknesses in research design and practice. It has generally not been feasible to select the communities studied randomly or to assign them to interventions randomly. In addition, it is always difficult to control the effects of extraneous factors occurring during the course of field research, and it is especially problematic with long-term field studies such as these. However, the analogous research findings of the various studies have supported the notion that this type of community approach may be generalizable, not only to cardiovascular disease, but, with appropriate modifications, to the prevention of alcohol-related problems as well.[5] There are several community-level research projects focused on alcohol-related problems. Most of these have been directly stimulated by community research in the area of heart disease. Brief descriptions of some of the most often cited studies appear in Appendix B. These studies reflect the general state of community level prevention research in the alcohol field. Like the CVD studies discussed above, this research has been limited by problems in research design and practice, and the efforts have shown only equivocal success. Few community-based alcohol studies have adopted environmentally oriented intervention strategies or multifaceted, integrated strategies. Following the CVD community-based studies, most of the alcohol studies have employed mass media information campaigns as their chief strategy, sometimes combined with a secondary, supporting strategy. Such efforts have mainly emphasized the role of education and training to modify individual behavior. Extensive reviews of the alcohol and drug-related literature have concluded that media informational and educational programs by themselves have been largely ineffective in preventing substance use or abuse.[6] Although some programs have succeeded in increasing alcohol or drug knowledge, few have influenced attitudes and even fewer have influenced behavior. However, given the current state of the art, there are good reasons to believe that many of the technical and strategic limitations can be overcome and that methodologically sound, community-based prevention research in alcohol use and abuse can be designed and carried out. The over-all experience with comprehensive, community-based prevention research, in the alcohol field as well as the heart field, justifies a significant investment of energy and resources in such studies in alcohol prevention research. RESEARCH OBJECTIVES OF THE NIAAA/OSAP INITIATIVE The primary objective of this RFA is to encourage long-term, controlled experimentation to test community-based, multifaceted, integrated programs for the prevention of alcohol-related problems. The focus of this announcement is alcohol, but a broader perspective may facilitate this emphasis. In some circumstances it may be expedient to include interventions that bear on other drug abuse as well (e.g., the use of tobacco). A. Characteristics of the Research There are two basic characteristics of community-based research: 1. Communities are the primary units of analysis; and 2. Interventions are targeted at entire sectors of the community, e.g., pregnant women, automobile drivers, and groups or organizations such as schools or worksites. Rates of effect would thus be calculated on a community- based population denominator. In addition to these characteristics, the research programs envisioned here are expected to have the following properties: 1. The chief prevention strategies should be based on environmental factors, such as: o normative factors (e.g., standards of behavior, general attitudes and beliefs regarding alcohol, mass media effects); o legal elements (e.g., alcohol beverage control (ABC) laws, laws regarding driving under the influence of alcohol, minimum purchase age laws, zoning); and o economic factors (e.g., pricing, factors that affect the cost of consumption). The array of interventions may also include strategies that are not considered environmental (e.g., educational programs) even if they have not by themselves been shown to be effective. Such interventions may prove to be more effective when they are combined with environmental approaches, and they might enhance the effectiveness of the other interventions. 2. The targeted outcomes of the interventions should be changes in the behaviors that contribute to the existence of alcohol-related problems or changes in the incidence or prevalence of the problems themselves. Examples of the former include the consumption of alcoholic beverages by pregnant women, binge drinking, driving while intoxicated or under the influence of alcohol, and experimental drinking by people under the age of 21. Examples of target problems are alcohol-related traffic crashes, alcohol-related violence, morbidity and mortality from alcohol-caused cirrhosis, alcohol-related birth defects, some cancers, and alcohol dependency. 3. The target groups must include youth and/or young adults, but need not by any means be limited to these populations. B. Research Alternatives Three distinctive community-based research alternatives are described below. Each of these focuses on an aspect of community-based intervention programs that is believed to bear directly on their effectiveness. Option 1 refers to programs investigating the effect of comprehensive programs, i.e., multifaceted, integrated intervention programs. Option 2 refers to programs assessing the effects of integrated intervention programs. Option 3 refers to research investigating the influence of community participation and involvement on the effects of intervention programs. Support will be available for research projects pursuing one or more of these options. Applications which do not address one of the three options will be considered non-responsive to the RFA and will be returned to the applicant. Applications proposing to investigate more than one option will probably require extra resources. Such applications will be considered if the potential scientific value of the research justifies the cost. The three research alternatives are as follows: 1. Testing effective comprehensive programs. A comprehensive intervention program is one that implements multiple interventions aimed at several outcomes. The objective of this option is to implement a "critical mass" of interventions, a combination of interventions that has the greatest anticipated effect possible with the resources available. The focus here is on the total effect with no concern at this time with determining the effects of the separate interventions. The application should provide a rationale for the selection of interventions and present a reasoned estimation of the magnitude of effects expected. Research using this approach should include comparisons between communities in which the interventions were implemented and communities in which they were not implemented. There should be at least two communities in each condition. 2. Assessing the interactive effects of an integrated set of interventions. An integrated program is one that implements a combination of interventions targeted at the same outcome. For example, a set of interventions targeted at reducing alcohol-related traffic crashes might include a responsible server training program, strict enforcement of driving while intoxicated and/or per se laws, and a mass media informational campaign telling the public about the increased enforcement. It seems reasonable to expect that the total impact of an integrated set of interventions will at least equal the sum of the effects of those interventions implemented separately. It is possible, however, that the combination of interventions might produce a total effect less than the sum of the individual effects. This may be because the different interventions are in some degree redundant. Or it may be that with communities, as is often the case with individuals, there is a diminishing return or satiation mechanism at work. There is also the possibility that an integrated set of interventions will produce synergistic effects, i.e., joint effects that are greater than the sum of the effects of the individual interventions. The existence of such effects is predicted by systems perspectives of the etiology of alcohol problems.[7,8,9] The rationale is that each program component reinforces and strengthens other program components. There are some very encouraging results employing computer simulation to validate a dynamic systems model of factors determining drinking and driving and alcohol consumption.[10] At present, there appears to be no experimental research providing answers to any of these questions in the field of alcohol prevention research. 3. The effects of community involvement. This approach invites research designs that employ community involvement as an independent variable, either in a simple present/absent design or in a more elaborate design implementing varieties of community involvement. Community involvement in the planning and delivery of program elements is regarded as an essential facilitating component of community-based research.[11,12,13,14] It has been an integral part of a number of studies, including the influential Stanford Heart Disease Studies, and COMMIT, a new, very large program aimed at reducing the amount of smoking by the smoking public in general and by heavy smokers in particular. (The Stanford Heart Disease Studies and COMMIT are described in Appendix A.) However, community involvement is not a systematically varied experimental factor in any of these studies. Despite the great importance assigned to it in the literature, there appears to be no research that treats community involvement as an experimentally varied factor. The critical task of research in this option will be to explicate the role of community involvement as a causal factor in the experimental design. METHODOLOGICAL ISSUES The research programs envisioned in this RFA will investigate prevention strategies applied to entire communities as the basic units of analysis. In published research reports, school districts, small cities, towns, and regions have been considered communities for the purposes of intervention research. What appears to be common to the operational conception of community is: a definite geographic and public identity, and a set of administrative, political, and associational organizations that together span the geographic entity. Using communities as units of analysis poses difficulties and challenges in the design and execution of research programs that satisfy strong criteria of scientific merit. Some of the more critical of these problems are discussed below. A. Design Issues Research directed at assessing the impact of a prevention strategy is inherently causal research. Community-based intervention trials are field experiments testing particular intervention strategies and, as such, have to satisfy criteria relating to internal validity. The classical paradigm for causal research is the randomized controlled experiment. A randomized controlled experiment may yield unbiased estimates of program effects even in the absence of knowledge about the relationships between exogenous variables, interventions, and mediating and outcome variables. Such experiments can be thought of as "black box" research.[15] However, black box experimentation does not yield an understanding of prevention programs or inform their development. A well- articulated theoretical framework can assist the researcher develop and test hypotheses about mediating mechanisms and develop or select appropriate interventions, program evaluation procedures, and intermediate and longer term outcome measures.[16] Randomized experiments testing theoretically relevant hypotheses about intervention effects not only yield estimates of those effects but have a bearing on the tenability of the theory as well. Few reported community studies have used random assignment of communities to intervention, and those that do have included too few communities to achieve much protection against alternative explanations by the procedure. (The project COMMIT, described in Appendix A, is one significant exception.) The cost in material and human resources of community-based trials is no doubt a significant contributor to this deficiency. Nevertheless, randomization can be a powerful procedure, and applicants are urged to consider its application in community-based research. Where it is not feasible, other procedures such as quasi-experimental designs and time-series analyses may provide valid information.[17] Randomization by itself does not ensure internal validity.[18] Strong efforts have to be made to ensure that the compared communities do not have different histories (except for intervention) during the course of the experiment. This requires comprehensive, close, and repeated observation of all communities, experimental and control, to document their continuing comparability and to assist in the impact assessment of unusual or dramatic events.[19] B. Sampling Issues It is rarely feasible, on financial and other practical grounds, to obtain random samples of communities large enough for statistical generalization. Researchers must usually provide grounds other than statistical ones for generalizing the results of a community study. Efforts should be made to select communities that are similar on at least such cogent characteristics as size, cultural tradition, demographic features, economic base, and regulation of alcoholic beverage distribution and sales. For a variety of reasons, samples of individuals within a community might be used to obtain information in any of the evaluation processes to be discussed below. These samples should be drawn by probability methods. Using convenience samples or allowing subjects to volunteer (select themselves) for study vitiates statistical generalization. Similarly, sample attrition in longitudinal studies makes it difficult to draw accurate conclusions on statistical grounds. The researcher must avoid these problems or find ways to compensate for them. C. Evaluation The research plan must include comprehensive evaluation components that are conceptually and procedurally integrated with the over-all research program. The three areas of evaluation -- formative, procedural, and outcome -- provide information relevant to the interpretation of the research findings. Evaluation bears directly upon the verification of the research hypotheses connecting implementations to outcomes. 1. Formative evaluation assesses the degree to which the instruments and procedures employed in the project have the properties necessary for the interventions of which they are a part. For example, a responsible alcoholic beverage intervention requires a training program that instructs servers how to detect when a patron has reached a specified limit of alcohol consumption. The training program must be developed and tested before it is employed in the research program. 2. Procedural evaluation refers to the periodic monitoring of the implementation of interventions during the course of the experiments to assure adherence to protocol and to document what actually was being done. Using the same responsible server example, it is necessary to ascertain whether the servers utilize their training effectively and dependably on the job. This phase of evaluation is essential to assessing the appropriate execution of the intervention. 3. Outcome evaluation refers to observations on the targeted outcomes, i.e., on the events and behaviors predicted to be consequences of the various interventions. The observations must reflect an explicit and verifiable causal linkage between the intervention and the targeted outcomes. In the responsible server example, the expectation is that patrons refused service after ingesting a critical amount of alcohol will be less likely to be involved in an automobile crash within an appropriate time period after leaving the establishment. This assumes, at least, that at the time they leave establishments that do and do not have responsible server programs, patrons will have different levels of alcohol impairment. A recent review of server training programs indicates that all programs were evaluated on factors relevant to this assumption, e.g., on the impact of the training program on servers' behavior or on their customers' consumption of alcohol.[20] The accident prevention effectiveness of the intervention, however, requires other observations. Changes in aggregate rates of alcohol-related traffic incidents over the life of the intervention could be used as indicators of this aspect of intervention effectiveness. Perhaps the most cogent evidence would be the rates of alcohol-related traffic incidents of these same groups. D. Measurement issues The conceptualization and procedures of observation of key variables and attributes must satisfy conventional standards of validity and reliability. Direct observations or measurements are generally preferable to indirect ones, but if the latter are employed, the causal linkage between the surrogate measure and the target variable must be explicated and established. The consumption of alcohol is a key variable. It is specified as both a cause and an effect in the etiology of a wide variety of alcohol-related problems. Self reports are very often used as indicators of alcohol consumption, primarily because they have face validity and are convenient to obtain, but there is persistent concern that they substantially underestimate actual consumption.[21] This problem may be less severe in comparing individuals' reports before and after an intervention or in subgroup comparisons, assuming that the bias toward underestimation is reasonably similar across groups. Blood alcohol concentration (BAC) measurements are more reliable and objective measures of consumption than self reports, even though there may be some significant individual response variation and error due to specific methods of test administration.[22] However, they cannot substitute for self-reports in all situations, such as retrospective accounts and large-scale surveys, for example. Aggregate measures of consumption are particularly convenient in community-based research, although they are subject to various biases. Alcohol sales reported for purposes of taxation have been used as indirect estimates of consumption in a number of studies relating price variations to motor vehicle crash rates and cirrhosis mortality rates.[23,24] Indices based on reported sales are subject to a number of biasing factors, but they appear to vary appropriately with other relevant variables. The Fatal Accident Reporting System (FARS) is an often used measure of the involvement of alcohol in traffic crashes, injuries, and fatalities. This approach allocates a proportion of all crashes reported to those in which alcohol is involved, but that proportion is an empirical estimate and requires systematic updating.[25] Community-based research programs will require research teams with diverse experience and expertise. Applicants must demonstrate that the research team possesses the methodological and technical skills requisite to carry out community-based prevention research. This area of research also will require a substantial commitment of time and resources. The research plan must specify the time needed to develop and implement the interventions and the schedule of activities. The research program must be scheduled to operate long enough for the anticipated effects to have an opportunity to occur and to give evidence as to whether these effects are transitory or enduring. ELIGIBILITY Applications may be submitted by public or private nonprofit organizations such as universities, colleges, hospitals, research institutes and organizations, units of State or local governments, and eligible agencies of the Federal Government. Women and minority investigators are encouraged to apply. INCLUSION OF MINORITIES IN STUDY POPULATIONS NIAAA/OSAP requires that applicants give added attention (where feasible and appropriate) to the inclusion of minorities in study populations for research into the etiology of diseases, research in behavioral and social sciences, clinical studies of treatment and treatment outcomes, research on the dynamics of health care and its impact on disease, and appropriate interventions for disease prevention and health promotion. If minorities are not included in a given study, a clear rationale for their exclusion should be provided. INCLUSION OF WOMEN IN STUDY POPULATIONS NIAAA/OSAP requires that applicants consider the inclusion of women in the study populations for all clinical research efforts. Exceptions would be studies of diseases which exclusively affect males or where involvement of pregnant women may expose the fetus to undue risks. Gender differences should be noted and evaluated. If women are not to be included, a clear rationale should be provided for their exclusion. In order to provide more precise information to the treatment community, it is recommended that publications resulting from NIAAA/OSAP-supported research in which the study population was limited to one sex for any reason other than that the disease or condition studied exclusively affects that sex, should state, in the abstract summary, the gender of the population studied, e.g., "male patients," "male volunteers," "female patients," "female volunteers." APPLICATION RECEIPT AND REVIEW SCHEDULE Receipt Initial Advisory Earliest Date Review Council Review Start Date Feb. 12, 1991 May/June 1991 Sept. 1991 Sept. 1991 Applications received after the specified receipt date will be returned to the applicant. APPLICATION PROCESS Applicants should use the grant application form PHS 398 (revised 10/88). The number and title of this RFA, "AA-91-01, Community-Based Research on the Prevention of Alcohol-Related Problems," should be typed in item number 2 on the face page of the PHS 398 application form. When using the PHS 398 application form to respond to an RFA, applicants must affix the RFA label, available in the application kit, to the bottom of the face page. Failure to use this label could result in delayed processing of the application, such that it may not reach the review committee in time for review. Application kits containing the necessary forms and instructions may be obtained from business offices or offices of sponsored research at most universities, colleges, medical schools, and other major research facilities. If such a source is not available, the following office may be contacted for the necessary application material: National Clearing House for Alcohol and Drug Information Post Office Box 2345 Rockville, MD 20852 Telephone: (301) 468-2600 The signed original and six (6) legible copies of the completed application should be sent to: Division of Research Grants, NIH Westwood Building, Room 240 Bethesda, MD 20892** REVIEW PROCEDURES The Division of Research Grants, NIH, serves as the central point for receipt of applications under this RFA. Applications received will be assigned to the Initial Review Group (IRG) in accordance with established Public Health Service Referral Guidelines. The IRG, consisting primarily of non-Federal scientific and technical experts, will review applications for scientific and technical merit. Notification of the review recommendations will be sent to the applicant after the initial review. Applications will receive a second-level review by the National Advisory Council on Alcohol Abuse and Alcoholism and the Advisory Committee on Substance Abuse Prevention, where reviews may be based on policy considerations as well as scientific merit considerations. Only applications recommended for approval by these advisory bodies may be considered for funding. Applications submitted in response to this announcement are not subject to the intergovernmental review requirements of Executive Order 12372, as implemented through the Department of Health and Human Services regulations at 45 CFR Part 100, and are not subject to Health Systems Agency review. REVIEW CRITERIA Criteria to be used in the scientific and technical review of applications will include the following: o potential contribution of the proposed research to enhancing the scientific basis of community-based alcohol abuse prevention programs; o significance of the proposed research for the objectives, options, and characteristics of research identified above; o inclusion of youth and/or young adults as one of the targeted populations; o evidence that the investigators are familiar with the state-of-the-art and existing knowledge gaps in their proposed area of study; o degree of scientific rigor in the design and implementation of the study; o responsiveness to the design issues discussed above; o adequacy of the methodology proposed to collect and analyze data; o qualifications and research experience of the principal investigator and other key research personnel; o availability of adequate facilities, other resources, and collaborative arrangements necessary for the research, including evidence of community cooperation; o appropriateness of budget estimates for the proposed research activities; and o adequacy of provisions for the protection of human subjects. AWARD CRITERIA Applications recommended for approval by the appropriate advisory bodies will be considered for funding on the basis of the overall scientific and technical merit of the proposal as determined by peer review, NIAAA/OSAP program needs and balance, and the availability of funds. TERMS AND CONDITIONS OF SUPPORT Grant funds may be used for expenses clearly related and necessary to carry out research projects, including both direct costs, which can be specifically identified with the project, and allowable indirect costs of the institution. Research grant support may not be used to establish, add a component to, or operate a prevention, rehabilitation, or treatment service program. Support for research-related prevention, rehabilitation, or treatment services and programs may be requested only for costs required by the research. These costs must be justified in terms of research objectives, methods, and designs that promise to yield generalizable knowledge and/or make a significant contribution to theoretical concepts. In order to expand a research project to more communities or to increase the number or complexity of the interventions to be tested, applicants may wish to seek additional funding from public or private agencies. If the research budget is to be supplemented by funds from sources other than NIAAA and OSAP, the application must provide documentation of the availability of these funds. Grants must be administered in accordance with the PHS Grants Policy Statement (Rev. January 1, 1987), which should be available from your office of sponsored research. Federal regulations at Title 42 CFR Part 52, "Grants for Research Projects," and Title 45 CFR Parts 74 and 92, generic requirements concerning the administration of grants, are applicable to these awards. PERIOD OF SUPPORT Applicants may request up to 5 years of support (renewable for subsequent periods). Annual awards will be made subject to availability of funds and progress achieved. A competing supplemental application may be submitted during an approved period of support to expand the scope or protocol of a project during the approved period. A competing continuation (i.e., renewal) application may be submitted before the end of an approved period of support to continue a project. AVAILABILITY OF FUNDS Applications submitted in response to this announcement will compete for approximately $2,000,000 in new grant money that is expected to be made available for this purpose in fiscal year 1991. It is anticipated that one to three projects will be supported. FINAL REPORT REQUIREMENTS Grantees are expected to submit an original and two copies of the final report of their project to NIAAA within 90 days of the project's termination. The final report should contain at least the following: o a literature review; o a clear statement of purpose and methodology; o the findings of the project; o an interpretation and discussion of those findings, including a clear exposition of their relevance to the prevention of alcohol-related problems and their implications for further research; o a description of dissemination achieved or planned; o a summary or abstract of the report. CONSULTATION AND FURTHER INFORMATION Potential applicants are encouraged to seek preapplication consultation. For information on preparing an application under this announcement, please contact: S. Frank Camilleri, Ph.D. Prevention Research Branch Division of Clinical and Prevention Research National Institute on Alcohol Abuse and Alcoholism 5600 Fishers Lane, Room 13C-23 Rockville, MD 20857 Telephone: (301) 443-1677 APPENDICES: EXAMPLES OF COMMUNITY-BASED PREVENTION RESEARCH These brief summaries of research are included to provide applicants with an introduction to the more influential research literature. Appendix A: Examples from Heart Disease and Cancer Research. 1. The North Karelia Project.[26] One of the earliest and most influential community-based intervention programs was a research program for health promotion in North Karelia, Finland, a rural county with a population of 180,000 in eastern Finland. The aims of the program were to improve detection and control of hypertension, reduce smoking, and promote diets lower in saturated fat and higher in vegetables and low-fat products. Most of the program was implemented between 1972 and 1977. Of special relevance here, the health problem was defined and dealt with as a community problem rather than one affecting only high-risk individuals, and the project worked together with county and local health officials to reorganize the way hypertension was detected and treated. The project included an information campaign that involved the cooperation and mediation of opinion leaders from both formal and informal groups: for example, members of a local housewives' association were encouraged and instructed in how to prepare healthier types of meals. Outcome evaluation procedures included baseline and follow-up (five years later) cross-sectional sample surveys of North Karelia inhabitants and a matched reference community. Smoking, for example, was measured by a standard set of questions validated at the follow-up survey by corroborative physical measures (analysis of the serum thiocyanate levels) of a random half of the subjects. Significant changes in risk estimates for the targeted elements were observed in North Karelia when compared to the reference area.[27] 2. The Stanford Studies.[28] Also in 1972, a team of researchers at Stanford University began a cardiovascular disease (CVD) prevention research program involving several communities in California. Their first study, The Three Community Study, implemented mass media and intensive, face-to-face education programs in one experimental community and a mass media program alone in a second community over a period of three years. A third community served as a control. Both media and media plus face-to-face instruction had significant effects on most outcome variables. The estimated net difference between control and treatment samples in a composite measure of risk for cardiovascular disease was 23-28%. The combined interventions (mass media plus face-to-face) were somewhat more effective than mass media alone in increasing knowledge and reducing smoking, but not in affecting other outcome variables. This study served primarily as a pilot test for the second study, The Five City Project (FCP).[29] The FCP is an ongoing, 13-year study that began in 1978. It involves 350,000 people in five communities and employs multiple methods of education and community organization. The primary goal of the FCP is to reduce the risk of cardiovascular disease. Among its other goals, the ones most significant here, are those of developing community organization methods and of transferring control and maintenance of the program to community organizations. CVD risk for total mortality and for coronary heart disease events was reduced significantly in treatment communities as measured in four time periods from two to 5-1/3 years after education began. 3. The Minnesota Heart Health Program.[30] This is a research and demonstration project of community health education designed to reduce the risk of CVD. A comprehensive educational program modeled after the North Karelia and Stanford studies was implemented in three communities using multiple media channels and withheld in three comparison communities. Surveillance is carried out in annual population surveys with periodic cohort studies. The program is carried out by a local staff and a community advisory board with campaign task groups in each educated community. The program began in 1981 and is designed for a nine-year period. 4. The Pawtucket Heart Health Program.[31] A continuing project first funded in 1980, the project has utilized a community activation approach to major levels of the community (individual, small group, organization, and community) and placed major emphasis upon the involvement of community volunteers in delivery of intervention components. The objective of the study is to prevent atherosclerotic heart disease by modifying behaviors (smoking, cholesterol elevating diets, high blood pressure, obesity, physical inactivity, and stress) that increase the risks of the development of the disease. The overall outcome evaluation is based in part on biennial cross-sectional random sample surveys of subjects in the intervention city and a comparison community. The surveys concentrate on risk factor changes and physiological assessment (blood sample and expired air analyses, and physical fitness). A baseline survey, completed in 1982, has been converted into a cohort with a second measurement being completed in 1987 and a third scheduled for completion in 1990.[32] 5. COMMIT.[33] On September 30, 1986, the National Cancer Institute launched a massive, highly structured research program, the Community Intervention Trial for Smoking Cessation (COMMIT). The project entails working with communities across North America to examine methods of getting smokers to quit, especially those who smoke 25 or more cigarettes per day. Eleven matched pairs of communities -- one community per pair for intervention and the other for comparison -- are involved in the research project. The project consists of three phases (planning and development, community implementation, and analysis and reporting) that will continue through 1994. As a research study, COMMIT requires the implementation of a standardized intervention process. A central feature of this process is the formation of a Community Board at each intervention site to work with the research institution to decide how to carry out the activities specified in the protocol, who should carry them out, and how best to use the available resources. Task forces are to be formed in each community to organize and implement activities in their areas. A "public education" task force is charged with sponsoring media advocacy training, developing communication networks, developing and conducting public events, encouraging policy changes, monitoring smoking regulations, and facilitating prevention efforts. A major goal of this group is to promote social action that leads to a smoke-free community. A "health care providers" task force will coordinate activities that train and encourage health care professionals to take an active role in smoking cessation efforts. A major goal of this group is having all health care facilities adopt and effectively implement policies to become smoke-free. A "worksites and organizations" task force has the responsibility of identifying ways to promote smoking control activities in worksites and other organizations. One of its major goals is to increase quitting rates among workers and organization members who smoke. The "cessation resources and services" task force will organize and coordinate the delivery of smoking cessation services such as cessation materials and classes. A major goal of this group is to assist smokers in identifying available assistance when they attempt to quit smoking. Appendix B: Examples from Alcohol Abuse Research. 1. Project CRASH.[34] A public education mass media campaign, Project CRASH (Countermeasures Related to Alcohol Safety on the Highways), was conducted in Vermont between 1972 and 1974. The program objectives were to increase knowledge and to modify attitudes and behavior so as to reduce the incidence of alcohol-impaired driving. The mass media campaign was conducted in one geographical test area in the state along with a countermeasure that consisted in deploying eight extra state troopers throughout four counties. In another area the campaign was conducted by itself, and a third (control) area received neither media campaign nor the countermeasure. There were no efforts made to involve elements of the community in the program activities. The media campaign was evaluated by a series of roadside surveys conducted in both test areas and in the comparison area. A breath alcohol determination was obtained at the roadside interviews. The program appeared to be modestly successful in increasing knowledge and affecting attitudes. It appeared to be somewhat successful in reducing the incidence of high-risk respondents driving with impairing blood alcohol concentrations (BACs). Changes in the proportion of alcohol-related fatal crashes to total fatal crashes could not be reliably attributed to the campaign because of uncontrollable environmental factors (particularly a gasoline shortage during the second year). 2. The "Winners" Program.[35] The California Prevention Demonstration program was begun in 1978. The interventions included mass media messages designed to increase awareness of the dangers of alcohol and to change attitudes toward alcohol. These interventions, it was hoped, would lead to changes in drinking behavior. The media materials were developed around a positive theme of "Winners Quit While They're Ahead" and targeted to a variety of populations. The program was implemented in two experimental sites in the San Francisco-Oakland East Bay Area. The primary site, approximately one-half of Oakland and all of San Leandro, was subject to moderation-in-drinking messages disseminated through mass media and to community organization and development activities. The secondary site received only the mass media messages. The comparison site, Stockton, received no special prevention activities. Unfortunately, budget cuts in midstream curtailed the evaluation component, and political intervention prohibited types of media messages felt to be effective but deemed unacceptable by the State of California, which sponsored the program. Overall program effects were limited. There was an increase in awareness of alcohol-related messages, but levels of concern about the dangers of alcohol remained generally constant. There was some increase in knowledge about alcohol and its effects among adults, and to a lesser extent among youth, in the primary site. No indication of significant attitude or behavior change was evident.[36] 3. The Rhode Island Alcohol Abuse/Injury Prevention Project.[37] This project was instituted in 1984 as a five- year community prevention trial to develop and test interventions to reduce the incidence of some of the most dangerous drinking-related behaviors. Three Rhode Island cities were selected for the project based on size, incidence of alcohol-related health problems, socio- demographic characteristics, and community resources. The demonstration community was chosen at random from among these. The other two were used as comparisons. The intervention strategy was directed at changing the knowledge, attitudes, and enabling behaviors of gatekeepers in their occupational roles as regulators of community drinking practice. For example, a person who protects his/her spouse from some of the consequences of their alcohol abuse, or an alcoholic beverage server who gives an intoxicated person "just one more drink," is seen as facilitating or enabling the abusive drinking. The three main intervention efforts included community mobilization, responsible alcohol server training, and more intensive and visible law enforcement. Several process indicators suggest that the gatekeepers responded appropriately to the interventions. A majority of servers participated in the training, and a post-test done twelve to eighteen months after the training shows enduring changes in attitudes and behavior. The police carried out a vigorous enforcement campaign. Alcohol-related arrests increased in the intervention community while in the two comparison communities they held level or decreased. The impact of the project on alcohol-related accidents is unclear. In the first year of the project, the rate of alcohol-related accidents increased in the intervention community, while in the comparison communities, it remained unchanged or decreased. It was not possible to disentangle genuine prevention effects from improved reporting and from some extraneous events.[38] 4. The Tri-Community Prevention Project.[39] The project was conducted on three small communities in Ontario, Canada, in the early 1980's. The key objective of the study was to determine whether a change in the proportion of heavy users of alcohol has a measurable impact on the overall distribution of consumption, in particular whether a shift in drinkers from the heavier to the lighter end of the consumption curve was accompanied by a comparable shift in the consumption patterns of lighter drinkers. The interventions were educational/counselling programs aimed primarily at heavy drinkers, but the project included some efforts at stimulating community organizational involvement in the problems of heavy drinking. Although relatively few heavy drinkers participated in the counselling and educational component of the intervention, it appears that on average there was a reduction in alcohol consumption in the course of their participation. However, as measured by self reports, the consumption patterns of one intervention community and the control community did not change appreciably over the period of study, nor were there evident substantial changes in the rates of alcohol-related problems such as drinking-and-driving incidents. 5. The New Zealand study.[40] In 1982, a community-based intervention program was initiated in New Zealand to reinforce moderate drinking patterns. The interventions included a media campaign and the use of community organizers to stimulate public and open discussion of alcohol policy issues. Change was monitored in six cities: two cities with an alcohol-focused community organizer and media campaign, two cities with the media campaign only, and two reference cities that received no intervention. The organizers focused on alcohol availability and pricing policies. The media campaign focused on reducing large quantity drinking by young men. The program effects were confounded with national trends, but the results of a survey suggest some change in community attitudes toward the program's policy objectives. 6. The Vermont study.[41] In the early 1980's, a community education program was designed to train individual drinkers to regulate their own blood alcohol concentration below a level of impairment (.05g/dl). The education program included a series of meetings with community leaders, a system through which drink calculators were distributed and demonstrated to community members through licensed alcoholic beverage outlets, and broadcast of public service television spots in which use of calculators was demonstrated. Program components were evaluated in three matched Vermont communities: one receiving the full community education program, one receiving TV spots only, and one serving as control. Results suggest that the media messages alone had the effect of deterring the consumption of any alcohol before driving, but the effect was stronger when the messages were combined with the community program. 7. The Midwestern Prevention Project (MPP, also referred to as Project STAR).[42] This is a recent and ambitious community drug intervention project that includes the prevention of alcohol problems. MPP is a multilevel program addressing individual, social, and environmental factors thought to influence adolescent drug use. The program was implemented first in 1984 in 50 schools in 15 communities that comprise Kansas City, Kansas, and Kansas City, Missouri. The program components, delivered from September 1984, through January 1986, consisted of a 10- session youth education program on skill training for resistance of drug use, 10 homework sessions involving active interview and role-plays with parents and family members, and mass media coverage. Mass media coverage was offered each year of the program but was not experimentally controlled. The other prevention program components were added sequentially, at the rate of one a year, in an effort to assess the separate and aggregate effects of combinations of interventions. The program has been repeated in Indianapolis, Indiana, in an effort to test its efficacy and replicability. The project has reported significant reduction in the use of tobacco, marijuana, and alcohol in both sites.[43] A full report is expected in 1991, following completion of the Indiana phase. REFERENCES 1. Wallack, L., Alcohol advertising reassessed: the public health perspective, in M. Grant, M. Plant, and A. Williams (eds), Economics and Alcohol: Consumption and Controls (London and New York: Gardner Press), 1983, pp. 243-248. 2. Prevention and Treatment of Alcohol Problems: Research Opportunities (Washington, D.C.: National Academy Press), 1989, p.24. 3. Salonen, J.T., T.E. Kottke, D.R. Jacobs, Jr., and P.J. Hannan, Analysis of community-based cardiovascular disease prevention studies -- evaluation issues in the North Karelia Project and the Minnesota Heart Health Program, International J. Epidemiology 15:176-182, 1986. 4. Farquhar, J.W., The community-based model of life style intervention trials, Am. J. Epidemiology 108:103-111, 1978. 5. Prevention and Treatment of Alcohol Problems, p. 128. 6. Moskowitz, J.M., The primary prevention of alcohol problems: a critical review of the research literature, J. Studies on Alcohol 50:54-88, 1989. 7. Room, R., Alcohol control and public health, in L. Breskow, J.E. Fielding, and L.B. Lave (eds), Annual Review of Public Health 5:293-317, 1984. 8. Holder, H.D. and L. Wallack, Contemporary perspectives for preventing alcohol problems: an empirically derived model, J. Pub. Health Policy, Autumn:324-339, 1986. 9. Holder, H.D. and N. Giesbrecht, Concepts and issues for community-based action to prevent alcohol and other drug- related problems. Paper read at the Symposium on Experiences with Community Action Projects for the Prevention of Alcohol and Other Drug Problems, Scarsborough, Ontario, Canada, March 11-16, 1989. 10. Holder, H.D. and J.O. Blose, Reduction of community alcohol problems: computer simulation experiments in three counties, J. Stud. Alcohol 48:124-135, 1987. 11. Farquhar, J., N. Maccoby, and P. Wood, Education and communication strategies, in W. Holland, R. Detels, and G. Knox (eds), Oxford Textbook of Public Health, Vol. 3 (London: Oxford Press), 1985, pp. 207-221. 12. Tuchfeld, B.S. and S.H. Marcus, Social models of prevention in alcoholism, in J. Matarazzo, et al., Behavioral Health: A Handbook of Health Enhancement and Disease Prevention (New York: John Wiley and Sons), 1984. 13. Pentz, M.A., C. Cormack, B. Flay, W.B. Hansen, C.A. Johnson, Balancing program and research integrity in community drug abuse prevention: Project STAR approach, J. School Health 56:389-393, 1986. 14. Kelly, J.G., A Guide to Conducting Prevention Research in the Community: First Steps (New York: The Hayworth Press), 1988. 15. Chen, H. and P.H. Rossi, Evaluating with sense: the theory driven approach, Evaluation Review 7:283-302, 1983. 16. Prevention and Treatment of Alcohol Problems, p.112. 17. Reichardt, C.S., Estimating the effects of community prevention trials. Paper prepared for the National Invitational Conference "Methodological Issues in Community Prevention Trials for Alcohol Problems," University of California, Berkeley, December 4-5, 1989. 18. Cook, T.D. and D.T. Campbell, Quasi-Experimentation: Design and Analysis Issues for Field Settings (Boston: Houghton Mifflin), 1979. 19. Reichardt, C.S. and H.F. Gollob, Ruling out threats to validity, Evaluation Review 13:3-17, 1989. 20. Saltz, R.F., Research needs and opportunities in server intervention programs, Health Education Quarterly 16:429 438, 1989. 21. Midanik, L., The validity of self-reported alcohol consumption and alcohol problems: a literature review, British J. Addiction 77:357-382, 1982. See also Fuller, R.K., K.K. Lee, and E. Gordis, Validity of self-report in alcoholism research: results of a veterans administration cooperative study, Alcoholism: Clinical and Experimental Research 12:201-205, 1988. 22. Hume, D. and E. Fitzgerald, Chemical tests for intoxication: what do the numbers really mean? Analytical Chemistry 57:876A-886A, 1985. See also Jones, A., How breathing techniques can influence the results of breath alcohol analysis, Medical Science and Law 22:275-280, 1982. 23. Saffer, H. and M. Grossman, Beer taxes, the legal drinking age, and youth motor vehicle fatalities, J. Legal Studies 16:351-374, 1987. 24. Cook, P.J. and G. Tauchen, The effect of liquor taxes on heavy drinking, Bell J. Econ. 13:379-390, 1982. 25. Ravenholt, R.T., Addiction mortality in the United States, 1980: tobacco, alcohol, and other substances, Population and Development Review 10:697-742, 1984. 26. Puska, P., J. Salonen, A. Nissinen, and J. Tuomilehto, The North Karelia Project, Preventative Medicine 12:191 195, 1983. 27. McAlister, A. et al., Theory and action for health promotion: illustrations from the North Karelia Project, AJPH 72;1:43-50, 1982. 28. Farquhar, J.W., et al., Community education for cardiovascular health, Lancet 1:1192-1195, 1977. 29. Farquhar, J.W., et al., The Stanford five-city project: design and methods, Amer. J. of Epidemiology 122:323-334, 1985. 30. Blackburn, H., et al., The Minnesota Heart Health Program: a research and demonstration project in cardiovascular disease prevention, in J.D. Matarazzo, et al., op. cit., pp. 1171-1178. 31. Lasater, T., et al., Lay volunteer delivery of a community based cardiovascular risk factor change program: the Pawtucket experiment, in J.D. Matarazzo et al., op. cit., pp. 1166-1170. 32. Lasater, T., Community designs employed in heart disease and cancer prevention projects. Paper prepared for the National Invitational Conference "Methodological Issues in Community Prevention Trials for Alcohol Problems," University of California, Berkeley, California, December 4-5, 1989. 33. COMMIT: Protocol Summary. National Cancer Institute, Bethesda, Maryland, 1988. 34. Worden, J.K., J.A. Waller, and T.J. Riley, The Vermont Public Education Campaign in Alcohol and Highway Safety: A final review and evaluation. CRASH Report I-5 (Montpelier, Vermont), 1975. 35. Wallack, L.M. and D.C. Barrows, Preventing Alcohol Problems in California: Evaluation of the Three-Year "Winners" Program, Social Research Group, School of Public Health, University of California, 1981. 36. Ibid., p.171. 37. Speare, M.C. and S.L.Buka, The Rhode Island alcohol/injury prevention project, in Evaluating Community Prevention Strategies. Papers prepared for a conference in San Diego, California, January 11-13, 1990. 38. Stout, R. L., Prevention experiments in the context of on going community process: opportunities or obstacles for research. Paper prepared for the National Invitational Conference "Methodological Issues in Community-Prevention Trials for Alcohol Problems," University of California, Berkeley, December 4-5, 1989. 39. Giesbrecht, N., and A. Pederson, General populations, alcohol-related problems or high-risk drinkers as foci in community-oriented prevention projects. Paper prepared for the National Invitational Conference " Methodological Issues in Community Prevention Trials for Alcohol Problems," University of California, Berkeley, December 4-5, 1989. 40. Casswell, S. and L. Gilmore, An evaluated community action project on alcohol, J. Studies on Alcohol 50:339-346, 1989. 41. Worden, J.K., et al., Preventing alcohol-impaired driving through community self-regulation training, Amer. J. Public Health 79:287-290, 1989. 42. Pentz, M.A. et al., A multicommunity trial for primary prevention of adolescent drug abuse, JAMA 261:3259-3266, 1989. 43. Pentz, M.A. Unpublished results presented to an audience at NIAAA, Rockville, Maryland, February 1990.