Responsiveness of health status measures to change among older adults

OBJECTIVE

This study examines the ability of commonly used self-reported health status measures to detect important changes in health (responsiveness) in older adults.

DESIGN

We compared changes in health status measures over the year among subgroups of a cohort of seniors: those who experienced an intervening illness, hospitalization or increase in drug regimen, and those who didn't. Differences between the two groups in changes in the measures were quantitated using Guyatt's responsiveness statistic and receiver operating characteristic curves (ROC).

SETTING

Staff model HMO.

PARTICIPANTS

1379 senior HMO enrollees who were participants in a health promotion trial and provided complete information at baseline and one year later.

MEASUREMENTS

The following self-reported health status measures were evaluated: restricted activity days, bed disability days, the Medical Outcomes Study physical function scale, self-evaluated health, and a positive affect scale.

MAIN RESULTS

All measures except the positive affect scale were able to discriminate significantly between seniors who were or were not hospitalized and/or reported a major illness in the intervening year. The two disability days measures showed the best responsiveness for all indicators of worsening health and included 70%-80% of the area under the ROC curves for major illness defined by hospitalization or self-report.

CONCLUSIONS

Commonly used, brief self-reported physical health status measures are responsive to intervening illness among relatively healthy seniors supporting their use in longitudinal geriatric research.

Dehydroepiandrosterone sulfate, incidence of myocardial infarction, and extent of atherosclerosis in men

BACKGROUND

Antiatherogenic effects of dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) have been suspected for more than 30 years, yet the available evidence to support or refute such effects in humans is inconclusive. The hypothesis has not been adequately tested in large-scale epidemiological studies.

METHODS AND RESULTS

The present study used a cohort of men initially free of clinically detectable coronary heart disease, stroke, and cancer to compare DHEAS levels measured in sera obtained in 1968-1971 between 238 cases who had definite coronary heart disease during the subsequent 18 years and 476 age-matched controls who survived the follow-up period and remained free of clinically detectable coronary heart disease. In a separate study, the relation of DHEAS levels to extent of atherosclerosis was examined among 82 cohort men who died during the follow-up period and had protocol autopsies. Age-adjusted DHEAS levels were lower among fatal cases of coronary heart disease than among controls (94.7 versus 106.9 micrograms/dl, respectively; p < 0.05). After adjustment for eight coronary risk factors, the odds ratio for fatal coronary heart disease comparing a 100-micrograms/dl difference in DHEAS level was 0.46 (95% confidence intervals, 0.19-1.07). In contrast, age-adjusted DHEAS levels did not significantly differ between nonfatal cases of myocardial infarction and controls (107.2 versus 106.9 micrograms/dl, respectively). Furthermore, DHEAS levels were not related to extent of atherosclerosis at autopsy.

CONCLUSIONS

These findings do not support a role of DHEAS in the development of nonfatal myocardial infarction or the progression of atherosclerosis. The association of DHEAS with fatal coronary heart disease and possibly with death from all causes merits further investigation. These findings suggest continued skepticism that DHEAS has an important role in coronary disease etiology or prevention.

Predictors of smoking cessation and relapse in older adults

We examined longitudinal changes in smoking behavior among older adults in three community cohorts of the Established Populations for Epidemiologic Studies of the Elderly. Smoking prevalence declined from 15% at baseline to 9% during 6 years of follow-up. Annual smoking cessation and relapse rates were 10% and less than 1%, respectively. Interval diagnosis of myocardial infarction, stroke, or cancer increased subsequent smoking cessation but not relapse. Although smoking cessation around diagnosis is increased, primary prevention could yield greater benefits.

Light to moderate alcohol consumption and mortality in the elderly

OBJECTIVE

To determine whether there is a relationship of low to moderate alcohol consumption with cardiovascular mortality in the elderly.

DESIGN

Prospective cohort studies with 5-year mortality follow-up.

SETTING

Three populations of community-dwelling elders.

PARTICIPANTS

Population-based cohorts of men and women, aged 65 or older, in three populations. Subjects with prior myocardial infarction, stroke, or cancer, as well as those lacking alcohol consumption data, were excluded from statistical analyses leaving 2,694 subjects in East Boston, Massachusetts, 2,293 subjects in Iowa, and 1,904 subjects in New Haven, Connecticut.

MAIN OUTCOME MEASUREMENTS

Alcohol consumption, total mortality, cardiovascular mortality, and cancer mortality.

RESULTS

Low to moderate alcohol consumption was associated with statistically significant lowered total as well as cardiovascular mortality in East Boston and New Haven. The relative risks of total mortality for low to moderate consumers of alcohol compared to those consuming no alcohol in the previous year were 0.7 (95% CI 0.6-0.8) in East Boston and 0.6 (95% CI 0.5-0.8) in New Haven. For cardiovascular mortality, the RRs were 0.6 in East Boston and 0.5 in New Haven (95% CI's exclude null). These results persisted after control for potential confounding variables. In Iowa, there were no significant differences in total or cardiovascular mortality according to alcohol consumption patterns. For cancer mortality, there were no significant associations with alcohol consumption in any of the three populations.

CONCLUSIONS

These data suggest that the relationship of low to moderate alcohol consumption with reduced total and cardiovascular mortality, which are well documented in middle age, also occur in older populations.

Effects of physical activity on health status in older adults. II. Intervention studies

This review has focused on a specific part of the relationship of exercise to health. The overall evidence supporting the health benefits of exercise is substantial and has been critically reviewed recently (18, 94). Thus, the United States Preventive Services Task Force recommends that all adults exercise regularly (94). The conclusions summarized below regarding older adults do not affect this basic recommendation. There is solid evidence that exercise can improve measures of fitness in older adults, particularly strength and aerobic capacity. These exercise effects occur in chronically ill adults, as well as in healthy adults. Because physical fitness is a determinant of functional status, it is logical to ask whether exercise can prevent or improve impairments in functional status in older adults. The evidence that exercise improves functional status is promising, but inconclusive. Problems with existing studies include a lack of randomized controlled trials, a lack of evidence that effects of exercise can be sustained over long periods of time, inadequate statistical power, and failure to target physically unfit individuals. Existing studies suggest that exercise may produce improvements in gait and balance. Arthritis patients may experience long-term functional status benefits from exercise, including improved mobility and decreased pain symptoms. Nonrandomized trials suggest exercise promotes bone mineral density and thereby decreases fracture risk. Recent studies have generally concluded that short-term exercise does not improve cognitive function. Yet the limited statistical power of these studies does not preclude what may be a modest, but functionally meaningful, effect of exercise on cognition. Future research, beyond correcting methodologic deficiencies in existing studies, should systematically study how functional status effects of exercise vary with the type, intensity, and duration of exercise. It should address issues in recruiting functionally impaired older adults into exercise studies, issues in promoting long-term adherence to exercise, and whether the currently low rate of exercise-related injuries in supervised classes can be sustained in more cost-effective interventions that require less supervision.

Older adults and smoking

This review supports the following conclusions regarding the benefits of smoking cessation in older adults: 1. Older smokers who quit have a reduced risk of death compared with current smokers within 1 to 2 years after quitting. Their overall risk of death approaches that of those who never smoked after 15 to 20 years of abstinence. 2. Smoking cessation in older adults markedly reduces the risks of coronary events and of cardiac deaths within 1 year of quitting, and risk continues to decline more gradually for many years. This is true for older adults both with and without a previous history of coronary disease and symptoms. 3. Risks of dying from several smoking-related cancers are reduced by quitting. Although the decline in risk may be more gradual for older than middle-aged adults, the benefits of cessation are apparent within 5 to 10 years of quitting. 4. Smoking cessation reduces the risk of COPD mortality after 10 to 15 years of abstinence in men and 5 to 10 years of abstinence in women. Within a shorter period of time, quitting reduces the prevalence of respiratory symptoms, slows the rate of decline in pulmonary function, and may reduce functional impairments and improve tolerance for exercise. 5. Smoking cessation may help slow osteoporosis and reduce the risk of hip fractures, but effects on rates of bone loss or subsequent risk of fracture have not yet been studied adequately. 6. Continued smoking in late life is associated with the development and progression of several major chronic conditions, loss of mobility, and poorer physical function. Former smokers appear to have higher levels of physical function and better quality of life than continuing smokers. Physicians and others should encourage older adults to stop smoking. Of course, not every older adult who quits will benefit in all, or indeed any, of these ways. Nevertheless, at the population level, the prospects are excellent that smoking cessation after age 65 will extend both the number of years of life and the quality of life.

Factors associated with participation in a senior health promotion program

The health status and life-style characteristics of participants in a senior health promotion program were compared with those of nonparticipants from the same HMO enrollee population. Nonparticipation was associated with lower income, less education, and lower involvement in community organizations. Although nonparticipants smoked more and evaluated their health less favorably than did participants, other risky behaviors and health status indicators differed little between the groups.

Smoking and mortality among older men and women in three communities

BACKGROUND

Although cigarette smoking is the leading avoidable cause of premature death in middle age, some have claimed that no association is present among older persons.

METHODS

We prospectively examined the relation of cigarette-smoking habits with mortality from all causes, cardiovascular causes, and cancer among 7178 persons 65 years of age or older without a history of myocardial infarction, stroke, or cancer who lived in one of three communities: East Boston, Massachusetts; Iowa and Washington counties, Iowa; and New Haven, Connecticut. At the time of the initial interview, prevalence rates of smoking in the three communities ranged from 5.2 to 17.8 percent among women and from 14.2 to 25.8 percent among men. During five years of follow-up there were 1442 deaths, 729 due to cardiovascular disease and 316 due to cancer.

RESULTS

In both sexes, rates of total mortality among current smokers were twice what they were among participants who had never smoked. Relative risks, as adjusted for age and community, were 2.1 among the men (95 percent confidence interval, 1.7 to 2.7) and 1.8 among the women (95 percent confidence interval, 1.4 to 2.4). Current smokers had higher rates of cardiovascular mortality than those who had never smoked (as adjusted for age and community, the relative risk was 2.0 [95 percent confidence interval, 1.4 to 2.9] among the men and 1.6 [95 percent confidence interval, 1.1 to 2.3] among the women), as well as increased rates of cancer mortality (relative risk, 2.4 [95 percent confidence interval, 1.4 to 4.1] among the men and 2.4 [95 percent confidence interval, 1.4 to 3.9] among the women). In both sexes, former smokers had rates of cardiovascular mortality similar to those of the participants who had never smoked, regardless of age at cessation, whereas the rates for all cancers, as well as smoking-related cancers, remained elevated among men who had once smoked.

CONCLUSIONS

Our prospective findings indicate that the mortality hazards of smoking extend well into later life, and suggest that cessation will continue to improve life expectancy in older people.

Morbidity and disability in older persons in the years prior to death

BACKGROUND

A large proportion of the disease and disability which affects older persons occurs in the years just prior to death. Little prospective evidence is available which quantifies the burden of morbidity and disability during these years.

METHODS

In three community-based cohorts of persons age 65 and older, chronic conditions and disability were evaluated for the three years prior to death in 531 persons who had three annual assessments and then died within one year of the third assessment. Number of chronic conditions, prevalence of disability in activities of daily living (ADLs), and prevalence of disability on a modified Rosow-Breslau scale were determined for these decedents and compared to 8821 members of the cohorts known to have survived.

RESULTS

Prevalence rates of disease and disability increased during the follow-up for both decedents and survivors, with decedents generally having higher rates than survivors. Disability rates prior to death, but not the number of diseases, increased with increasing age at death. The odds ratio for disability in ADLs at any of the three assessments for decedents versus survivors ranged from 3.0 to 4.2 in the three communities. In each community the odds ratio for ADL disability was higher in women decedents versus survivors than in men decedents versus survivors.

CONCLUSIONS

These results have important implications for disability levels in future older populations in which death is projected to occur at increasingly higher ages.

Tracking progress toward national health objectives in the elderly: what do restricted activity days signify?

Restricted activity days is the measure by which the 1990 health objectives for prevention of functional disability in older adults will be evaluated. Yet its significance in older populations is poorly understood. We evaluated its use as an outcome measure for a randomized trial designed to impact upon physical function in elderly HMO enrollees. As predicted, restricted activity days was more correlated with physical disability measures than with other health status measures. Distributional properties and rates of missing data were shortcomings.

Cognitive impairment and mortality: a study of possible confounders

Baseline neuropsychological function was assessed in 2,123 Framingham Heart Study participants and was related to mortality over an 8- to 10-year follow-up period. During that time, 573 persons died. Using Cox proportional hazards models, the authors showed poor cognitive function to be consistently associated with an increased risk of death. This association persisted after adjustment for the confounding effects of age, education, and illness. Subjects scoring below the 26th percentile of performance were at increased risk of mortality compared with high scorers (the relative risk was 1.3 for the 11th percentile-25th percentile and 1.7 for the 1st percentile-10th percentile).

Demographic and health characteristics of elderly smokers: results from established populations for epidemiologic studies of the elderly

Although smoking has been shown to be associated with excess morbidity and mortality, most studies have focused on young and middle-aged rather than elderly smokers. We examined the demographic characteristics and physical and psychological health of elderly cigarette smokers in four population-based studies (N = 3,673, 3,811, 2,811, and 4,165) of persons 65 years of age and older. Although there was substantial geographic variation in the percentages of smokers, the demographic and health characteristics of smokers were similar across the sites. Most women had never smoked, but most men were former or current smokers. The percentage of never smokers was highest in the "oldest old." Among the elderly ever smokers, men were more likely to have quit than women, and the relative percentage of former smokers increased with age. Current smokers were generally more likely to consume alcohol than never smokers. In this age group, cigarette smoking was typically associated with higher rates of physical disease and symptoms, poorer self-perceived health status, and higher levels of depressive symptoms. Based on these four large geographically diverse population surveys, cigarette smoking remains an important health burden and public health challenge among the elderly.

Thiazide diuretic agents and the incidence of hip fracture

Thiazide diuretic agents lower the urinary excretion of calcium. Their use has been associated with increased bone density, but their role in preventing hip fracture has not been established. We prospectively studied the effect of thiazide diuretic agents on the incidence of hip fracture among 9518 men and women 65 years of age or older residing in three communities. At base line, 24 to 30 percent of the subjects were thiazide users. In the subsequent four years, 242 subjects had hip fractures. The incidence rates of hip fracture were lower among thiazide users than nonusers in each community; the Mantel-Haenszel relative risk of hip fracture, adjusted for community and age, was 0.63 (95 percent confidence interval, 0.46 to 0.86). The protective effect of the use of thiazides was independent of sex, age, impaired mobility, body-mass index, and current and former smoking status; the multivariate adjusted relative risk of hip fracture was 0.68 (95 percent confidence interval, 0.49 to 0.94). Furthermore, the protective effect was specific to thiazide diuretic agents, since there was no association between the use of antihypertensive medications other than thiazides and the risk of hip fracture. These prospective data suggest that in older men and women the use of thiazide diuretic agents is associated with a reduction of approximately one third in the risk of hip fracture.

Chest pain and coronary heart disease mortality among older men and women in three communities

Angina pectoris is a manifestation of coronary heart disease, yet little is known from clinical or epidemiologic studies about its prognosis in older populations. We investigated the relation of uncomplicated angina symptoms to risk of coronary heart disease mortality within 3 years in a prospective study of 8,359 people aged 65 and older residing in three communities. From baseline (1981-1983) to the third year of follow-up (1984-1986), there were 245 deaths from coronary heart disease. Three classifications of chest pain were defined using the Rose Questionnaire: nonexertional chest pain, chest pain on exertion (including angina), and angina. Exertional chest pain was a strong, independent predictor of coronary heart disease death for older men and women. There were no differences in the prognostic implications of this symptom between the sexes; the relative risks being 2.4 (95% confidence interval, 1.4-4.4) in men and 2.7 (1.7-4.2) in women. The risk of coronary heart disease mortality for those reporting chest pain on exertion was at least as high as that for participants whose symptoms met the Rose Questionnaire criteria for angina. The association between exertional chest pain and coronary heart disease mortality was independent of other coronary risk factors. The relation was specific for deaths from coronary heart disease, as there was no association between exertional chest pain and noncoronary causes of death. Chest pain on exertion conveys important prognostic information about risk of coronary death in older populations, regardless of gender.

The prediction of midlife coronary heart disease and hypertension in young adults: the Johns Hopkins multiple risk equations

Educating medical students about the identification of risk factors for coronary disease and hypertension should be enhanced by exercises in which medical students identify their own risk factors and visualize the impact of current risk status on future risk of disease. A cohort of 1,130 former Johns Hopkins medical students were examined in medical school and followed annually from 1948 to 1964 to identify youthful factors associated with the development of coronary heart disease and hypertension in midlife. In the ensuing years through 1984, 51 cases of coronary heart disease and 114 cases of hypertension developed. Multiple risk equations using Cox proportional hazards regression were developed to predict these endpoints. Incidence of coronary heart disease was predicted best by an equation containing age, serum cholesterol at baseline, cigarette smoking at baseline, and paternal history of coronary disease. Hypertension was predicted best by an equation containing age, systolic blood pressure at baseline, paternal history of hypertension, and Quetelet index. These equations were applied to a class of present-day medical students to demonstrate the considerable variability in 30-year risk of coronary disease or hypertension. Thus, coronary heart disease and hypertension in midlife can be predicted by factors identified in youth. The Johns Hopkins multiple risk equations may be valuable as tools in preventive cardiology education to illustrate risk assessment and the importance of risk factor interventions.

The association of age-related macular degeneration and lens opacities in the aged

Data from 3,087 persons age 45 or older in the National Health and Nutrition Survey, 1971-74, showed that subjects with lens opacifying disease had an increased odds for age-related macular degeneration (AMD) compared to those who had no lens opacities. The crude odds ratio for aphakic patients was 4.6 (95% CI = 2.5, 8.6). The association remained after controlling for age, sex, and systolic blood pressure (a common risk factor) in a logistic regression model. These data are consistent with the hypothesis that light-induced damage may contribute to both lens and retinal disease and suggest that cataract extraction without implantation of ultra-violet/blue light absorbing intraocular lens may place subjects at increased risk of AMD.

Antihypertensive drug therapy and survival by treatment status in a national survey

The National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Follow-up Study, an investigation of a cohort originally examined during the period 1971-1975, provided an opportunity to assess the frequency of antihypertensive drug therapy in the United States during the period 1982-1984. For most age-sex-race subgroups, the frequency of medication use during 1982-1984 was higher than that observed during 1976-1980 based on the NHANES II. In the interval 1982-1984, diuretic agents were the most frequent medications prescribed (47% of drugs prescribed), and beta-blockers were second (17%). At the time of the initial survey in 1971-1975, participants had their blood pressures measured and a history of diagnosis and treatment of hypertension ascertained. Follow-up for vital status was 93% complete by 1984 (average length of follow-up, 9 years). In white men and women aged 50 years and older, the relative risk of death increased steadily, from those with elevated blood pressure (systolic blood pressure greater than or equal to 160 mm Hg or diastolic blood pressure greater than or equal to 95 mm Hg) but no history of hypertension to those treated for hypertension but whose blood pressure was still elevated. Regardless of history or treatment, those with an elevated blood pressure had about a 25-30% excess risk of death. Evidence from these national studies shows a high frequency of antihypertensive drug therapy in 1982-1984 and suggests the importance of adequate blood pressure control for optimal survival.

Race and sex differentials in the impact of hypertension in the United States. The National Health and Nutrition Examination Survey I Epidemiologic Follow-up Study

Hypertension was evaluated longitudinally in a nationally representative sample of the US population. This study, based on the data from the National Health and Nutrition Examination Survey NHANES I Epidemiologic Follow-up Study, analyzed changes in blood pressure and frequency of treatment, hypertension incidence, and ten-year survival of the cohort relative to hypertension status at baseline. Higher prevalence rates for each older age group, especially in women, as previously reported on data from community studies were confirmed. However, this analysis found minimal differences in the incidence of hypertension between men and women for all age groups. Incidence rates for blacks were at least twice the rates for whites for almost every age-sex group. Decreased survival in older hypertensive men probably explained the higher prevalence in older women. Treatment and location of measurement in clinic or household must be major considerations in the calculation of incident cases.

Rose Questionnaire angina among United States black, white, and Mexican-American women and men. Prevalence and correlates from The Second National and Hispanic Health and Nutrition Examination Surveys

The prevalence of Rose Questionnaire angina and its association with coronary heart disease risk factors and manifestations were investigated in representative samples of the US population. The study populations included 1,135 black and 8,323 white subjects aged 25-74 years examined in the Second National Health and Nutrition Examination Survey, 1976-1980, and 2,775 Mexican-American subjects aged 25-74 years examined in the Hispanic Health and Nutrition Examination Survey, Mexican-American portion, 1982-1983. Age-adjusted prevalence rates of Rose angina were similar among black, white, and Mexican-American women (6.8%, 6.3%, and 5.4%, respectively). An excess in the prevalence of Rose angina was observed in women compared with men for white and Mexican-American persons under age 55 years, but not for those over age 55. Electrocardiographic evidence of myocardial infarction and self-reported heart attack were strongly associated with prevalent Rose angina among white men and women aged 55 years and over, but not among those below age 55. Serum cholesterol, body mass index (weight (kg)/height (m)2), current cigarette smoking, and dyspnea were independently associated with an increased risk of prevalent angina in multivariate logistic models for white women, excluding those with a prior heart attack. Because many younger women with chest pain who may consult physicians are likely to have elevations in cardiovascular risk factors, their self-reported chest pain can be used as an opportunity to intervene and reduce their future risk of cardiovascular disease.

The relation of psychosocial dimensions of work with coronary heart disease risk factors: a meta-analysis of five United States data bases

The relation of job psychologic demands and decision latitude to four coronary heart disease risk factors (cholesterol, smoking, and systolic and diastolic blood pressures) was tested among 12,555 men in five investigations conducted in the United States during the period 1959-1980 (National Health and Nutrition Examination Surveys I and II, National Health Examination Survey, Western Collaborative Group Study, and Exercise Heart Survey). Using an imputation strategy, the authors attached measures of the two job characteristics above to persons in each data base by occupation. In 19 possible tests, decision latitude was related (p less than 0.05) to cholesterol and smoking in two instances in the predicted direction in the Exercise Heart Survey, when controlling for Type I error rate. Psychologic demands were not related to any of the risk factors. When a meta-analysis was performed across data bases, all relations were in the predicted direction except for the relation of psychologic demands to systolic pressure, and two of these were statistically significant (p less than 0.05): the relation of job decision latitude to smoking and to systolic pressure. The interaction of psychologic demands and decision latitude was not related to any of the risk factors when two common forms of an interaction were tested. These results indicate that psychosocial aspects of work, in particular the decision latitude of the job, may be related to some cardiovascular risk factors.

Occupational strain and the incidence of coronary heart disease

The hypothesis that men in high "strain" occupations have an increased risk of developing coronary heart disease was tested during an 18-year follow-up study from 1965-1983 of a cohort of 8,006 men of Japanese ancestry in Hawaii. There were no significant associations between the incidence of coronary heart disease and the individual job components of high psychologic demands and low job control or for the high strain interaction of these two characteristics. There were, in fact, trends of associations opposite to that predicted by the job strain model which were of borderline significance in multivariate analyses. Stratified analyses by level of acculturation showed similar inverse associations of job strain and coronary heart disease for the more Westernized men and no association for the more traditional men. There were also no significant associations among the various job characteristics and the major risk factors for coronary heart disease in this cohort. The disagreement of these results with those from other studies may be due to methodologic differences of using men whose usual and current occupations were the same in this study compared with using only current occupation in the other studies, the use of different methods of measuring job strain, or the possibility that men in this cohort perceive or react to occupational strain differently.

Prospective study of pneumonia hospitalizations and mortality of U.S. older people: the role of chronic conditions, health behaviors, and nutritional status

Mortality and hospitalization rates for pneumonia have increased among older Americans during recent years (1979-86), despite a national commitment to the reduction of premature deaths from pneumonia. A prospective study of deaths and hospitalizations attributable to pneumonia was conducted among 5,474 subjects ages 55 and older who participated in the NHANES I Epidemiologic Followup Study. Prevalent chronic conditions, health behaviors, and nutritional status indicators, measured at baseline, were examined in relation to pneumonia hospitalization and death during 12 years of followup. Mortality and hospitalization rates for pneumonia were higher among men than women, and higher among those ages 65 and older than among those 55-64 of both sexes. Risk of pneumonia death was higher among subjects with a history of congestive heart failure, stroke, cancer, or diabetes. Risk of pneumonia hospitalization was higher among subjects with a history of chronic obstructive pulmonary disease and among men who were current smokers. Daily alcohol consumption did not increase risk of pneumonia in this study population. Four measures of nutritional status were examined taking age, prevalent chronic conditions, and cigarette smoking into account: body mass index, arm muscle area, and serum albumin and hemoglobin levels. Risk of pneumonia death was 2.6 times higher in men in the lowest quartile, compared with men in the highest quartile, of body mass index. Similarly, the risk was 4.5 times higher among men in the lowest quartile of arm muscle area. Risk of death from pneumonia was 3.6 times higher among women in the lowest quartile of serum albumin levels compared with women in the highest quartile. Relative risks for these nutritional status indicators remained elevated after adjusting for age and the medical history risk factors. These risk factors should be taken into account when designing and evaluating pneumonia vaccination trials and community prevention programs.