Weight gain as a barrier to smoking cessation among military personnel.

PURPOSE: To assess the relationships between active-duty military status, military weight standards, concern about weight gain, and anticipated relapse after smoking cessation. DESIGN: Cross-sectional study. SETTING: Hospital-based tobacco cessation program. SUBJECTS: Two hundred fifty-two enrollees, of 253 eligible, to a tobacco cessation program in 1999 (135 men, 117 women; 43% on active duty in the military). MEASURES: Independent variables included gender, body mass index (weight/height2), and military status. Dependent variables included about weight gain with smoking cessation and anticipated relapse. RESULTS: In multivariate regression analyses that controlled for gender and body mass index, active-duty military status was associated with an elevated level of concern about weight gain (1.9-point increase on a 10-point scale; 95% confidence interval [CI], 1.0- to 2.8-point increase), as well as higher anticipated relapse (odds ratio [OR] = 3.6; 95% CI, 1.3 to 9.8). Among subjects who were close to or over the U.S. Air Force maximum allowable weight for height, the analogous OR for active-duty military status was 6.9 (p = .02). CONCLUSIONS: Occupational weight standards or expectations may pose additional barriers for individuals contemplating or attempting smoking cessation, as they do among active-duty military personnel. These barriers are likely to hinder efforts to decrease smoking prevalence in certain groups.

Performance characteristics of digital fundus photography as a screening test for diabetic retinopathy in a low-risk population.

The purpose of the present study was to estimate the sensitivity and specificity of reviewers' assessments of digital fundus photography (DFP) images, using the findings from comprehensive ophthalmologic examination as the criterion standard. One hundred and fifty-two patients with diabetes underwent comprehensive ophthalmologic examination by an active duty U.S. Air Force (ADAF) staff ophthalmologist, and the examination findings were used as the criterion standard for the present study. Eight other ophthalmologists (M.D.'s) and 10 optometrists (O.D.'s), all ADAF providers, each evaluated seven nonstereo standard field DFP images from the left and right eyes (14 images per patient) of the 152 diabetic patients, assessing each patient for evidence of diabetic retinopathy (yes, no, not sure). The sensitivity, specificity, and proportion of "not sure" responses were computed separately for MD and OD reviewers and then compared. O.D. reviewers gave "not sure" responses more frequently than M.D.'s. With "not sure" responses treated as correct (i.e., identical to the results of ophthalmologic examination), both types of providers had 100% sensitivity for cases that are usually treated and 83% sensitivity for cases that are not usually treated but require close follow-up. Specificity (for any diabetic retinopathy) was 92% for M.D.'s and 93% for O.D.'s. With M.D.'s as photographic reviewers, DFP-based screening required only 27 ophthalmologic examinations per 100 patients screened; with O.D.'s as photographic reviewers, 31 ophthalmologic examinations were required per 100 patients screened. Using either ophthalmologists or optometrists as photographic reviewers, DFP-based screening for diabetic retinopathy has very good sensitivity and excellent specificity.

Predictors of low physical fitness in a cohort of active-duty U.S. Air Force members.

BACKGROUND: Each branch of the U.S. armed forces has standards for physical fitness as well as programs for ensuring compliance with these standards. In the U.S. Air Force (USAF), physical fitness is assessed using submaximal cycle ergometry to estimate maximal oxygen uptake (VO2(max)). The purpose of this study was to identify the independent effects of demographic and behavioral factors on risk of failure to meet USAF fitness standards (hereafter called low fitness). METHODS: A retrospective cohort study (N=38,837) was conducted using self-reported health risk assessment data and cycle ergometry data from active-duty Air Force (ADAF) members. Poisson regression techniques were used to estimate the associations between the factors studied and low fitness. RESULTS: The factors studied had different effects depending on whether members passed or failed fitness testing in the previous year. All predictors had weaker effects among those with previous failure. Among those with a previous pass, demographic groups at increased risk were toward the upper end of the ADAF age distribution, senior enlisted men, and blacks. Overweight/obesity was the behavioral factor with the largest effect among men, with aerobic exercise frequency ranked second; among women, the order of these two factors was reversed. Cigarette smoking only had an adverse effect among men. For a hypothetical ADAF man who was sedentary, obese, and smoked, the results suggested that aggressive behavioral risk factor modification would produce a 77% relative decrease in risk of low fitness. CONCLUSIONS: Among ADAF members, both demographic and behavioral factors play important roles in physical fitness. Behavioral risk factors are prevalent and potentially modifiable. These data suggest that, depending on a member's risk factor profile, behavioral risk factor modification may produce impressive reductions in risk of low fitness among ADAF personnel.

Short term effects of cigarette smoking on hospitalisation and associated lost workdays in a young healthy population.

OBJECTIVE: There are relatively few published studies conducted among people of younger ages examining short term outcomes of cigarette smoking, and only a small number with outcomes important to employers. The present study was designed to assess the short term effects of smoking on hospitalisation and lost workdays. DESIGN: Retrospective cohort study. SETTING: Military population. SUBJECTS: 87 991 men and women serving on active duty in the US Army during 1987 to 1998 who took a health risk appraisal two or more times and were followed for an average of 2.4 years. MAIN OUTCOME MEASURES: Rate ratios for hospitalisations and lost workdays, and fraction of hospitalisations and lost workdays attributable to current smoking (population attributable fraction). RESULTS: Compared with never smokers, men and women who were current smokers had higher short term rates of hospitalisation and lost workdays for a broad range of conditions. Population attributable fractions (PAFs) for outcomes not related to injury or pregnancy were 7.5% (men) and 5.0% (women) for hospitalisation, and 14.1% (men) and 3.0% (women) for lost workdays. Evidence suggests that current smoking may have been under reported in this cohort, in which case the true PAFs would be higher than those reported. CONCLUSIONS: In this young healthy population, substantial fractions of hospitalisations and lost workdays were attributable to current smoking, particularly among men.

Effects of a group exercise program on strength, mobility, and falls among fall-prone elderly men.

OBJECTIVES: This randomized controlled trial studied the effects of a low- to moderate-intensity group exercise program on strength, endurance, mobility, and fall rates in fall-prone elderly men with chronic impairments. METHODS: Fifty-nine community-living men (mean age = 74 years) with specific fall risk factors (i.e., leg weakness, impaired gait or balance, previous falls) were randomly assigned to a control group (n = 28) or to a 12-week group exercise program (n = 31). Exercise sessions (90 minutes, three times per week) focused on increasing strength and endurance and improving mobility and balance. Outcome measures included isokinetic strength and endurance, five physical performance measures, and self-reported physical functioning, health perception, activity level, and falls. RESULTS: Exercisers showed significant improvement in measures of endurance and gait. Isokinetic endurance increased 21% for right knee flexion and 26% for extension. Exercisers had a 10% increase (p < .05) in distance walked in six minutes, and improved (p < .05) scores on an observational gait scale. Isokinetic strength improved only for right knee flexion. Exercise achieved no significant effect on hip or ankle strength, balance, self-reported physical functioning, or number of falls. Activity level increased within the exercise group. When fall rates were adjusted for activity level, the exercisers had a lower 3-month fall rate than controls (6 falls/1000 hours of activity vs 16.2 falls/1000 hours, p < .05). DISCUSSION: These findings suggest that exercise can improve endurance, strength, gait, and function in chronically impaired, fall-prone elderly persons. In addition, increased physical activity was associated with reduced fall rates when adjusted for level of activity.

Differences in socioeconomic status and survival among white and black men with prostate cancer.

After diagnosis with prostate cancer, Black men in the United States have poorer survival than White men, even after controlling for differences in cancer stage. The extent to which these racial survival differences are due to biologic versus non-biologic factors is unclear, and it has been hypothesized that differences associated with socioeconomic status (SES) might account for much of the observed survival difference. The authors examined this hypothesis in a cohort study, using cancer registry and US Census data for White and Black men with incident prostate cancer (n = 23,334) who resided in 1,005 census tracts in the San Francisco Bay Area during 1973-1993. Separate analyses were conducted using two endpoints: death from prostate cancer and death from other causes. For each endpoint, death rate ratios (Blacks vs. Whites) were computed for men diagnosed at ages <65 years and at ages > or =65 years. These data suggest that differences associated with SES do not explain why Black men die from prostate cancer at a higher rate when compared with White men with this condition. However, among men with prostate cancer, SES-associated differences appear to explain almost all of the racial difference in risk of death from other causes.

A cost-effectiveness analysis of prescribing strategies in the management of gastroesophageal reflux disease.

OBJECTIVE: Patients who have uncomplicated gastroesophageal-reflux disease (GERD) typically present with heartburn and acid regurgitation. We sought to determine the cost-effectiveness of H2-receptor antagonists (H2RAs) and proton-pump inhibitors (PPIs) as first-line empiric therapy for patients with typical symptoms of GERD. METHODS: Decision analysis comparing costs and benefits of empirical treatment with H2RAs and PPIs for patients presenting with typical GERD was employed. The six treatment arms in the model were: 1) Lifestyle therapy, including antacids; 2) H2RA therapy, with endoscopy performed if no response to H2RAs; 3) Step up (H2RA-PPI) Arm: H2RA followed by PPI therapy in the case of symptomatic failure; 4) Step down arm: PPI therapy followed by H2RA if symptomatic response to PPI, and antacid therapy if response to H2RA therapy; 5) PPI-on-demand therapy: 8 wk of treatment for symptomatic recurrence, with no more than three courses per year; and 6) PPI-continuous therapy. Measurements were lifetime costs, quality-adjusted life years (QALYs) gained, and incremental cost effectiveness. RESULTS: Initial therapy with PPIs followed by on-demand therapy was the most cost-effective approach, with a cost-effectiveness ratio of $20,934 per QALY gained for patients with moderate to severe GERD symptoms, and $37,923 for patients with mild GERD symptoms. This therapy was also associated with the greatest gain in discounted QALYs. The PPI-on-demand strategy was more effective and less costly than the H2RA followed by PPI strategy or the other treatment arms. The results were not highly sensitive to cost of therapy, QALY adjustment from GERD symptoms, or the success rate of the lifestyle arm. However, when the success rate of the PPI-on-demand arm was < or =59%, the H2RA-PPI arm was the preferred strategy. CONCLUSION: For patients with moderate to severe symptoms of GERD, initial treatment with PPIs followed by on-demand therapy is a cost-effective approach.

Race, prostate cancer survival, and membership in a large health maintenance organization.

BACKGROUND: Population-based cancer registry data have shown that black men with prostate cancer have poorer stage-specific survival than white men, while studies in equal-access health care systems have not found racial differences in stage-specific survival. This study was designed to test the hypothesis that black men and white men with prostate cancer have equal stage-specific survival in equal-access health care systems. METHODS: We conducted a cohort study using cancer registry data from all incident cases of prostate cancer occurring in a five-county San Francisco Bay Area region. Incident cases occurred among members (5263 cases, from January 1973 through June 1995) and nonmembers (16,019 cases, from January 1973 through December 1992) of the Kaiser Permanente Medical Care Program, a large health maintenance organization. Death rate ratios (DRRs, black men versus white men) for Kaiser members and nonmembers were computed for all stages combined (adjusting for age and stage) and for each stage (adjusting for age). RESULTS: Among Kaiser members, adjusted DRRs comparing black men with white men were as follows: all stages combined, 1.28 (95% confidence interval [CI] = 1.14-1.44); local stage, 1.23 (95% CI = 1.01-1.51); regional stage, 1.30 (95% CI = 0.97-1.75); and distant stage, 1.27 (95% CI = 1.07-1.50). Corresponding DRRs for nonmembers were as follows: all stages combined, 1.22 (95% CI = 1.14-1.30); local stage, 1.24 (95% CI = 1.09-1.41); regional stage, 1.48 (95% CI = 1.29-1.68); and distant stage, 1.01 (95% CI = 0.91-1.12). CONCLUSIONS: These results show poorer prostate cancer survival for black men compared with white men in an equal-access medical care setting. The findings are most consistent with the hypothesis of increased tumor virulence in blacks.

Regional differences in known risk factors and the higher incidence of breast cancer in San Francisco.

BACKGROUND: The age-adjusted incidence of breast cancer in the San Francisco Bay Area has consistently been higher than that in other regions of the United States. The distribution of established risk factors for breast cancer (i.e., parity, age at first full-term pregnancy, breast-feeding, age at menarche, and age at menopause) and probable risk factors (e.g., alcohol consumption) also differs across geographic regions. PURPOSE: A study was planned to explore the extent to which differences in the regional distribution of established and probable risk factors could explain the increased incidence of breast cancer in the San Francisco Bay Area. METHODS: Age-adjusted breast cancer incidence rates for January 1978 through December 1982 were obtained for the San Francisco Bay Area and other regions from the Surveillance, Epidemiology, and End Results (SEER) Program. Risk factor data from January 1980 through December 1982 were computed from the Cancer and Steroid Hormone Study, a population-based, case-control study of women 22-55 years of age who resided in eight SEER regions. Two different statistical methods were used to compute the relative risk (RR) of breast cancer associated with residence in the San Francisco Bay Area versus other regions, after adjusting for regional differences in known risk factors. RESULTS: Substantial differences in the distribution of breast cancer risk factors were found between the San Francisco Bay Area and other regions. Nearly all of these differences would be expected to lead to an elevated incidence of breast cancer in the San Francisco Bay Area. With the use of incidence rates adjusted only for age, the RR for San Francisco Bay Area residence from January 1978 through December 1982 compared with residence in seven other SEER areas was 1.14 for white women and 1.10 for black women. Depending on the statistical method used, the RR was reduced to approximately 0.96-0.99 for white women and 0.75-0.83 for black women, after further adjusting for established and probable risk factors (parity, age at first full-term pregnancy, breast-feeding, age at menarche, age at menopause, and alcohol consumption). Without adjustment for alcohol consumption, the corresponding results were 0.97-1.02 for white women and 0.77-0.88 for black women. CONCLUSIONS: Among both white women and black women, the elevated breast cancer incidence rate in the San Francisco Bay Area can be completely accounted for by regional differences in known risk factors.

The reliability and validity of an obstacle course as a measure of gait and balance in older adults.

The use of an obstacle course to quantify gait, balance and functional mobility in elderly persons, particularly to assess objectively changes following exercise and rehabilitation interventions, has not been extensively developed or tested. In this study, we describe an 18-item obstacle course developed as an outcome measure for an exercise intervention among fall-prone elderly men. Reliability and validity of the obstacle course was tested in a group of 58 community-living elderly men (mean age = 75 years). Each subject's performance was videotaped and timed. The videotapes were scored by a physical therapist and a physician. Inter-rater reliability between the raters was high (Kappa = 0.96, p < 0.0001). Both the obstacle course score and time correlated significantly with gait velocity, a 6-minute walk test, and a performance-oriented instrument of gait and balance. Obstacle course scores showed significant improvement among the most impaired subjects, but not among higher functioning subjects following a 3-month exercise intervention. These results suggest that an obstacle course may be a useful and valid method for measuring outcomes related to mobility tasks in selected elderly populations. Further work is needed to determine in which populations, and for which outcomes, an obstacle course is better than simpler performance-based measures.

Restructuring VA ambulatory care and medical education: the PACE model of primary care.

The Veterans Health Administration (VHA) Western Region and associated medical schools formulated a set of recommendations for an improved ambulatory health care delivery system during a 1988 strategic planning conference. As a result, the Department of Veterans Affairs (VA) Medical Center in Sepulveda, California, initiated the Pilot (now Primary) Ambulatory Care and Education (PACE) program in 1990 to implement and evaluate a model program. The PACE program represents a significant departure from traditional VA and non-VA academic medical center care, shifting the focus of care from the inpatient to the outpatient setting. From its inception, the PACE program has used an interdisciplinary team approach with three independent global care firms. Each firm is interdisciplinary in composition, with a matrix management structure that expands role function and empowers team members. Emphasis is on managed primary care, stressing a biopsychosocial approach and cost-effective comprehensive care emphasizing prevention and health maintenance. Information management is provided through a network of personal computers that serve as a front end to the VHA Decentralized Hospital Computer Program (DHCP) mainframe. In addition to providing comprehensive and cost-effective care, the PACE program educates trainees in all health care disciplines, conducts research, and disseminates information about important procedures and outcomes. Undergraduate and graduate trainees from 11 health care disciplines rotate through the PACE program to learn an integrated approach to managed ambulatory care delivery. All trainees are involved in a problem-based approach to learning that emphasizes shared training experiences among health care disciplines. This paper describes the transitional phases of the PACE program (strategic planning, reorganization, and quality improvement) that are relevant for other institutions that are shifting to training programs emphasizing primary and ambulatory care.

Evaluation of the VA's Pilot Program in Institutional Reorganization toward Primary and Ambulatory Care: Part I, Changes in process and outcomes of care.

PURPOSE: To evaluate the impact of the reorganization of an academic Veterans Affairs medical center toward primary and ambulatory care--including the implementation of a medical-center-wide interdisciplinary firm system and ambulatory care training program--on the quality of primary ambulatory care. METHOD: Randomly selected male veterans visiting the Veterans Affairs Medical Center in Sepulveda, California, were surveyed in 1992, early in the implementation of the program, and in 1993, after the program had been fully implemented. Two surveys were used: one before the veterans saw their primary care providers (practice-based survey) and the other immediately after patient visits (visit-based survey). Survey-participant data were then linked to computerized utilization and mortality data. Survey topics were mapped to the medical center's strategic plan and goals for ambulatory care, and focused on patients' reports about the care they had received in terms of continuity, access, preventive care, and other aspects of the biopsychosocial model of care. Administrative computer data were then used to evaluate effects on medical center workload. Statistical analyses included analysis of variance, analysis of covariance, chi-square, and logistic regression. RESULTS: For practice-based comparisons, complete data were available for 1,262 veterans in 1992 and 1,373 in 1993. For visit-based comparisons, complete data were available for 1,407 veterans in 1992 and 643 in 1993. Results included statistically significant improvements in continuity of care and detection of depression as well as increased rates of preventive care counseling (smoking and exercise). The proportion of veterans reporting being seen by physicians increased, as did the proportion of patients seen for check-ups rather than for acute problems. Fewer patients were seen in subspecialty clinics than in general medicine clinics. Patient satisfaction increased, hospitalizations decreased, and death rates decreased. Alcohol counseling and access to care for acute symptoms declined. Workload shifted from subspecialists to generalists and from inpatient care to outpatient care. CONCLUSION: The institutional reorganization toward primary and ambulatory care succeeded in substantially improving the quality of ambulatory care, reflecting improvements in the system of care and of health care provider training in ambulatory care.

Evaluation of the VA's Pilot Program in Institutional Reorganization Toward Primary and Ambulatory Care: Part II, A study of organizational stresses and dynamics.

BACKGROUND: Many academically affiliated hospitals are moving from an inpatient, subspecialty orientation in their patient care and educational programs toward a greater emphasis on ambulatory and primary care. Few studies have focused on the organizational, staffing, and management issues involved in implementing these changes. METHOD: The authors carried out a qualitative evaluation of the process of change in an academic Department of Veterans Affairs hospital during implementation of a major ambulatory primary care program. They interviewed four top managers individually and 59 top and middle managers, house officers, and patients in focus groups in the spring of 1992, nine months after implementation of the key components of the program. Four raters independently evaluated written transcripts of focus-group sessions and identified themes. RESULTS: The main problems identified were difficulty with administrative integration between inpatient and outpatient services; need for training, retraining, and orientation; tensions due to changes in roles and organizational culture; and inefficiency due to the need for frequent negotiations in daily work life. These four problems reflected tensions associated with new demands imposed by matrix management, changing job descriptions, policies and procedures, and changing patterns of communication and record keeping. CONCLUSION: During the process of implementation of a primary care focus throughout a medical center, extra demands upon staff are inevitable and should be anticipated and planned for. Twelve key factors for successful organizational change are discussed.

Increasing the impact of quality improvement on health: an expert panel method for setting institutional priorities.

BACKGROUND: Successful implementation of modern ongoing quality improvement (QI) methods requires investment of institutional resources, but can produce significant improvements in medical care. A health care organization's goals and objectives for improving care are expressed in strategic plan documents, which could provide a framework for planning quality improvement initiatives. However, institutional strategic planning processes are often not well linked to QI staff and resources. We developed the Quality Action Program (QAP) to connect QI to strategic planning. HISTORY: In 1991, Sepulveda VHAMC implemented a major primary care initiative, documented in a comprehensive strategic plan. The QAP was developed to enable the initiative to be evaluated within a QI context. THREE-ROUND EXPERT PANEL PROCESS: To carry out the QAP, members of an institution's quality council engage in a structured consensus process. The first round involves reading educational materials and filling out a quality action survey the second round includes participation in an expert panel meeting, and the third round involves making final priority rankings. EIGHT-STEP QAP IMPLEMENTATION PLAN: QI staff carry out activities to prepare for and carry out the three-round expert panel process. RESULTS: QAP induced significant institutional QI activity directed toward achieving the top-ranked QI criterion--ensuring continuity of care. Continuity of care improved significantly over time between the pre- and post-QAP periods. CONCLUSIONS: Expert panel methods can be used to link strategic plan goals and objectives to QI efforts.

Expert ratings of primary care goals and objectives.

OBJECTIVE: To develop consensus on proficiencies internal medicine residents should master in the area of primary and managed care. DESIGN: A draft compendium of primary care educational objectives including important clinical topics was developed at the Sepulveda Veterans Health Administration Medical Center Pilot Ambulatory Care and Education (PACE) Program as part of a local and regional primary care curricular review. Fifty-one experts, including leaders in the Society of General Internal Medicine, the Association of Program Directors in Internal Medicine, the American College of Physicians, general internal medicine division chiefs, and Veterans Affairs (VA) associate chiefs of staff for ambulatory care rated the compendium. MEASUREMENTS AND MAIN RESULTS: Eleven objectives and nine clinical topics were rated "critically important" (4.7 or above on a five-point scale). General internal medicine chiefs and associate chiefs of staff for ambulatory care judged them to be covered adequately in fewer than half of the 17 VA Western Region-affiliated internal medicine programs. Forty-five objectives and 77 clinical topics were considered at least somewhat important to the education of general internal medicine residents in primary care. The VA raters reported that in the prior academic year, their housestaffs had spent between 21% (postgraduate year I) and 33% (postgraduate year III) of their time in ambulatory care settings. CONCLUSION: With the emphasis on primary and managed care, there is a need for national consensus on educational objectives in primary care general internal medicine. This review provides educators with a benchmark to test the adequacy of their institutions' curricula in primary care internal medicine.

Helping practices reach primary care goals. Lessons from the literature.

We reviewed rigorous evaluations of programs to enhance the quality and economy of primary care. We identified 36 evaluations published from 1980 through 1992. We abstracted data on objectives, setting(s), patients and processes, outcomes, and costs of care. We identified successful programs, as well as significant gaps in our knowledge of how to improve aspects of care. In specific, computer reminders and social influence-based methods fostered preventive and economic care. Nurse implementation of prevention protocols increased their performance. Multidisciplinary teams improved access and economy. Regional organization of practices or telephone management improved access; regionalization also reduced emergency care. Improvements were not found in continuity, comprehensiveness, humanistic process, physical environment, or health outcomes. Primary care practices can implement several programs to continuously improve prevention and access, and to reduce costs and use of unnecessary services. Research documenting how to accomplish other major goals, including health outcome changes, in different practice types is needed.

Falls in the nursing home.

OBJECTIVE: To review the epidemiology and causes of falls and fall-related injuries in nursing homes and to provide clinicians with a structured framework to evaluate and treat nursing home residents at risk for falls. DATA SOURCES: All large-scale published studies documenting incidence, causes, risk factors, and preventive strategies for falls in nursing homes were reviewed. RESULTS: The mean incidence of falls in nursing homes is 1.5 falls per bed per year (range, 0.2 to 3.6 falls). The most common precipitating causes include gait and balance disorders, weakness, dizziness, environmental hazards, confusion, visual impairment, and postural hypotension. The most important underlying risk factors for falls and injuries include some of these same items and others, such as lower-extremity weakness, gait and balance instability, poor vision, cognitive and functional impairment, and sedating and psychoactive medications. Many strategies for the prevention of falls have been tried, with mixed success. The most successful consider the multifactorial causes of falls and include interventions to improve strength and functional status, reduce environmental hazards, and allow staff to identify and monitor high-risk residents. Strategies that reduce mobility through use of restraints have been shown to be more harmful than beneficial and should be avoided. CONCLUSIONS: A focused history and physical examination after a fall can usually determine both the immediate underlying causes of the fall and contributing risk factors. In addition, regular evaluations in the nursing home can help identify patients at high risk who can then be targeted for specific treatment and prevention strategies.

Cigarette smoking and stroke in a cohort of U.S. male physicians.

OBJECTIVE: To examine the association between cigarette smoking and the risk for stroke in men. DESIGN: Prospective cohort study. SETTING: Participants in the Physicians' Health Study, a randomized trial of aspirin and beta-carotene among U.S. male physicians. PATIENTS: 22,071 men, 40 to 84 years of age at entry, free from self-reported myocardial infarction, stroke, and transient ischemic attack; followed for an average of 9.7 years; and classified as never-smokers, current smokers, and former smokers based on self-report. MEASUREMENTS: Incidence rates of total, ischemic, and hemorrhagic stroke. RESULTS: With never-smokers as the reference group (relative risk, 1.00), relative risks (adjusted for age and treatment assignment) for total nonfatal stroke (n = 312) were as follows: former smoking, 1.20 (95% CI, 0.94 to 1.53); currently smoking fewer than 20 cigarettes daily, 2.02 (CI, 1.23 to 3.31); and currently smoking 20 or more cigarettes daily, 2.52 (CI, 1.75 to 3.61) (P for trend, < 0.0001). For participants who had total fatal stroke (n = 28), the risk for stroke was not increased with smoking (P > 0.2). In proportional-hazards models that controlled simultaneously for other risk factors, these associations were not materially altered. CONCLUSIONS: Current but not former cigarette smoking was significantly associated with an increased risk for stroke in men. Smoking may account for a substantial amount of stroke-associated morbidity and mortality.