INTRODUCTION The National Cancer Institute (NCI) established the Tobacco Research Implementation Group (TRIG) in 1998 to assist the Institute in formulating a Tobacco Research Implementation Plan (TRIP). The TRIG, consisting of NCI and non-NCI scientists, was established to examine the NCI's extramural research portfolio in tobacco and identify priorities for tobacco-related research for the next 5-7 years. METHODS An initial analysis of the Fiscal Year 1997 NCI research portfolio of tobacco-related research was prepared by the NCI members of the TRIG. This initial analysis organized the FY97 research portfolio into eight research categories, and within these primary categories, into a varying number of subcategories. Following deliberation by the full TRIG, a ninth category was added to the analysis. Research categories are presented in the following section. Research on the treatment of tobacco-related diseases, such as lung cancer, was specifically excluded from this analysis. The first step in the analysis was to identify all intramural and extramural research and training grants and contracts that received funding in FY97 and addressed tobacco-related issues. Searching the NCI's Research Analysis and Evaluation Branch (RAEB) database of grants and contracts identified all NCI extramural research projects. All projects that were coded as tobacco-related and were active in FY97 were identified for the analysis. Training grants and intramural research projects were identified by reviewing internal grant files and tracking databases. Only research projects involving greater than 10 percent relevance to tobacco were included in the analysis. Only the portion of the research budget reflecting the relative percent of the project that focused on tobacco issues was included in financial calculations. Projects that did not have an associated budget, such as those completing research under a no-cost extension, were excluded from the analysis. The RAEB search was compared to a search of the NIH CRISP database to ensure that all projects had been identified. Once the projects to be included in the analysis had been identified, a single reviewer categorized all projects. Each project was assigned to only one category based on the primary hypothesis of study. Area experts then reviewed these category assignments. Discrepancies were resolved across reviewers and appropriate refinements to category definitions were made. The categories used for the analysis represent research in the current portfolio and may not represent all of the important issues that might or should be addressed in the future. The NCI portfolio primarily includes tobacco-related issues as they relate to cancer. Additional research addressing the basic biology of drug dependence and tobacco-related heart disease are addressed by other NIH institutes, such as the National Institute of Drug Abuse (NIDA) and the National Heart Lung and Blood Institute (NHLBI), and these are not included in this analysis. TOBACCO RESEARCH CATEGORIES Each tobacco research project was assigned to one of nine tobacco research categories and associated subcategories. These categories and the criteria for assignment include:   Basic Biobehavioral Research   Research studies that assess basic principles of behavior, including basic laboratory research as well as research directed at individuals, and seeking to develop and test behavioral models of action.   Behavioral/Biological Associations - Studies to assess the basic relationships between biological and behavioral factors and/or their potential relationship to tobacco-use prevention and cessation interventions.   Behavioral Mechanisms - Studies to identify behavioral and social factors that influence tobacco-use behaviors and studies to develop and test specific interventions aimed at changing tobacco-related behaviors.     Prevention Research   Studies to develop and test interventions that are designed to prevent the initiation of tobacco use among individuals.   Treatment of Nicotine Dependence   Research studies that assess the clinical treatment of individuals for nicotine dependence. The specific focus of treatment interventions is the individual in clinical and physician-based settings.   Pharmacological Treatments - Studies to identify and test pharmacological interventions, such as nicotine replacement, specifically for the treatment of physiological addiction to nicotine.   Behavioral Treatments - Studies to develop and test behavioral interventions specifically for the treatment of physiological addiction to nicotine that involve health care providers as well as studies directed at behavioral interventions among health care providers and in medical practice settings.   Self-help Treatments - Studies to develop and test interventions directed at individuals that do not involve face-to-face interaction with a health care provider or tobacco control counselor, such as self-help and telephone counseling.   Combined Pharmacological/Behavioral Treatments - Studies to identify and test combined pharmacological and behavioral interventions, such as the combined use of nicotine replacement and biofeedback, specifically for the treatment of physiological addiction to nicotine.     Community and State Intervention Research   Research studies that test group and population-based interventions and the dissemination of group and population-based interventions to prevent and reduce tobacco use.   Prevention Research - Studies to test and implement interventions among groups of individuals that are designed to prevent the initiation of tobacco use.   Cessation Research - Studies to test and implement interventions among groups of individuals that are designed to stop current use of tobacco.   Combined Prevention/Cessation Research - Studies to test and implement interventions among groups of individuals that are designed to both prevent the initiation of tobacco use and stop current tobacco use.     Policy Research   Research studies that assess the effect of policies, including worksite, community, state and Federal policies, on tobacco initiation and use.     Basic Biological Research   Laboratory-based research studies that assess the effects of tobacco and tobacco products on cell biology and cellular mechanisms of action, including the role of tobacco and tobacco byproducts in the initiation and promotion of cancer and the biological and health effects of exposures to tobacco.   Genetics - Studies that seek to identify genes that are related to the tobacco-induced carcinogenic process or that can conceivably be related to cancer risk associated with tobacco use. These include the identification of genetic susceptibility genes associated with tobacco-induced carcinogenesis or tobacco addiction. They do not include those genes being studied in the context of an epidemiological investigation.   Carcinogenesis - Studies to assess the role of tobacco and tobacco byproducts in DNA damage, carcinogen metabolism, and carcinogenesis and/or studies seeking to identify biomarkers as indicators of DNA damage or carcinogenesis.   Animal Models - Studies that seek to identify or that use animals as models of the biologic and/or carcinogenic effects of tobacco and tobacco byproducts.     Epidemiology   Population-based research studies that assess associations between tobacco use, behavioral and biological factors, and cancer.   Descriptive Epidemiology - Population and patient-based studies that describe trends and associations between risk factors, tobacco, and cancer.   Molecular Epidemiology - Population and patient-based studies that assess trends and associations between molecular and genetic factors, tobacco, and cancer.     Surveillance   Research studies that assess trends in tobacco-related risk factors, behaviors and health services to determine changes over time and the influence of these trends on cancer incidence, morbidity, mortality and survival.     Chemoprevention   Studies that assess the administration of specific chemicals and/or dietary modifications to reverse or suppress the carcinogenic effects of tobacco and prevent the development of invasive cancer.   RESULTS The tobacco research portfolio analysis revealed support for a wide spectrum of tobacco-related research. The NCI supported 176 research projects across both intramural and extramural divisions, excluding projects with 10 percent or less relevance to tobacco. The total FY97 annual budget for tobacco-related research was $76.2 million. As shown in the figure above, epidemiology, treatment of tobacco addiction, and community and state intervention research accounted for the largest number of research projects, about 20 percent each of the tobacco research portfolio funded by the NCI. In contrast, only one tobacco-related surveillance project was funded in FY97. This picture differed when assessed by the allocation of funding. Community and state intervention research accounted for more than 40 percent of the FY97 funds for tobacco-related research. Research on the treatment of tobacco addiction, which represented the next largest allocation of funds, accounted for 18 percent of the funds expended. This proportion was roughly comparable to the proportion of projects addressing treatment of nicotine dependence. When assessed either by the number of projects or the amount of funding, biobehavioral, prevention and policy research each accounted for less than 10 percent of the tobacco portfolio. Tobacco-related surveillance was the most under supported area of tobacco research in 1997. The tobacco research portfolio also was evaluated in terms of the funding mechanisms used to support tobacco research. Fifteen different types of funding mechanisms were used to support tobacco research. The most frequently used funding mechanism was the R01 research grant; 48 percent of the tobacco-related research projects were supported by R01s. This mechanism also accounted for a large proportion of the total tobacco funding, with 36 percent of tobacco-related research dollars being provided to investigators through the R01 mechanism. Thirty-five percent of funds supported research through contracts (N01). These two mechanisms (R01 and N01) combined accounted for almost 70 percent of the tobacco research funding. CONCLUSIONS The tobacco research portfolio analysis revealed support for a broad spectrum of tobacco-related research. Proportional to tobacco's cancer burden, however, tobacco research reflects a disproportionately small portion of the NCI budget. While tobacco use accounts for almost 30 percent of cancer deaths, the total annual budget for tobacco-related research represents only 3 percent of the NCI budget. This analysis reflects current tobacco research priorities. Cancer Control Research priorities focus on community and state interventions, treatment of nicotine addiction and epidemiological investigations. Biobehavioral, prevention and policy research on the other hand, appear under-emphasized, and tobacco surveillance research receives almost no attention. The analysis provides a clear picture of current tobacco research across topic areas and the distribution of funding across projects and mechanisms. It provides a strong foundation from which research priorities for the next 5-7 years can be established.