Changes in severity of depressive symptoms and mortality: the Italian Longitudinal Study on Aging.

BACKGROUND: Depression is recognized as being associated with increased mortality. However, there has been little previous research on the impact of longitudinal changes in late-life depressive symptoms on mortality, and of their remission in particular. METHOD: As part of a prospective, population-based study on a random sample of 5632 subjects aged 65-84 years, with a 10-year follow-up of vital status, depressive symptoms were assessed by the 30-item Italian version of the Geriatric Depression Scale (GDS). The number of participants in the GDS measurements was 3214 at baseline and 2070 at the second survey, 3 years later. Longitudinal changes in depressive symptoms (stable, remitted, worsened) were examined in participants in both evaluations (n=1941). Mortality hazard ratios (MHRs) according to severity of symptoms and their changes over time were obtained by means of Cox proportional hazards regression models, adjusting for age and other potentially confounding factors. RESULTS: Severity is significantly associated with excess mortality in both genders. Compared to the stability of depressive symptoms, a worsened condition shows a higher 7-year mortality risk [MHR 1.46, 95% confidence interval (CI) 1.15-1.84], whereas remission reduces by about 40% the risk of mortality in both genders (women MHR 0.55, 95% CI 0.32-0.95; men MHR 0.59, 95% CI 0.37-0.93). Neither sociodemographic nor medical confounders significantly modified these associations. CONCLUSIONS: Consistent with previous reports, the severity and persistence of depression are associated with higher mortality risks. Our findings extend the magnitude of the association demonstrating that remission of symptoms is related to a significant reduction in mortality, highlighting the need to enhance case-finding and successful treatment of late-life depression.

CIND and MCI in the Italian elderly: frequency, vascular risk factors, progression to dementia.

OBJECTIVE: To estimate prevalence and progression to dementia of cognitive impairment, no dementia (CIND), mild cognitive impairment (MCI), and relative subtypes, evaluating the relationships with daily functioning, cardiovascular diseases and vascular risk factors. METHODS: We evaluated CIND and MCI in the Italian Longitudinal Study on Aging. The neuropsychological battery assessed global cognitive function, memory and attention. Two thousand eight hundred thirty participants were examined at baseline and after a mean follow-up of 3.9 +/- 0.7 years. RESULTS: The prevalence was 9.5% for CIND and 16.1% for MCI. Prevalence rates for CIND subtypes were 1.8% for amnestic, 2.3% for single nonmemory, 1.5% for multidomain, and 3.9% for CIND defined only on global cognitive function. The prevalence was 7.0% for amnestic, 7.8% for single nonmemory, and 1.3% for multidomain MCI. Incidence of dementia (per 1,000 person-years) was 7.63 in the total sample, 21.37 in CIND, and 13.59 in MCI. In MCI, rates ranged from 8.74 in amnestic to 40.60 in multidomain subtype. The highest incidence of 56.02 per 1,000 person-years was found in multidomain CIND. Both CIND and MCI increased by almost three times the risk of dementia at follow-up. Among baseline variables, only previous stroke and impairment in instrumental activities of daily living significantly increased the risk of dementia at follow-up. CONCLUSIONS: Both cognitive impairment, no dementia and mild cognitive impairment are frequent in the Italian elderly (2,955,000 prevalent cases expected) and significantly predict progression to dementia. Individuation of subgroups with different risk factors and transition rates to dementia is required to plan early and cost-effective interventions.

Epidemiology of distal symmetrical neuropathies in the Italian elderly.

OBJECTIVES: To estimate prevalence and incidence of distal symmetric neuropathies (DSN) in the Italian elderly, and to evaluate the accuracy of our procedure to screen for DSN. METHODS: In eight Italian municipalities, a population-based sample was directly evaluated both at baseline (1992) and after a 3-year follow-up. Cohort members who had died were studied. DSN diagnosis and subtyping were made according to specified diagnostic criteria. RESULTS: Our screening procedure proved accurate (sensitivity 94.7%, specificity 70%, positive predictive value 18.9%), and provided an adjusted prevalence of 7.0 (95% CI, 6.9 to 7.0). Women outnumber men both in the oldest age groups and as a whole. Rates increase with increasing age in both genders. Among the 2,845 individuals re-screened at the follow-up and the 221 deceased subjects with reliable information, we identified 100 incident cases of DSN. Adjusted annual incidence rate (per 1,000 person-years) in the population 65 to 84 years of age is 7.9 (95% CI, 6.3 to 9.5), and for the nondiabetic DSN is 5.76 (95% CI, 4.3 to 7.3). Age significantly predicted the onset of DSN both in diabetic individuals (for every increasing year of age RR = 1.07; 95% CI, 1.01 to 1.14) and in the entire study population (RR = 1.05; 95% CI, 1.02 to 1.09). CONCLUSIONS: We provide the first population-based distal symmetric neuropathies incidence data, as well as prevalence rates from an unselected sample of Italian elderly. Distal symmetric neuropathies are an age-associated condition, but the frequency of diabetic distal symmetric neuropathies declines with age, coincident with an increase in nondiabetic cases.

Vascular risk factors, incidence of MCI, and rates of progression to dementia.

OBJECTIVE: To estimate prevalence, incidence, and rate of progression of mild cognitive impairment (MCI) to dementia and correlated vascular risk factors with incident MCI and its progression to dementia. METHODS: The authors evaluated 2,963 individuals from the population-based sample of 5,632 subjects 65 to 84 years old, at the first (1992 to 1993) and second survey (1995 to 1996) of the Italian Longitudinal Study on Aging (ILSA), with a 3.5-year follow-up. Dementia, Alzheimer disease (AD), vascular dementia (VaD), other types of dementia, and MCI were classified using current clinical criteria. RESULTS: Among the 2,963 participants, 139 MCI patients were diagnosed at the first ILSA survey. During the 3.5-year follow-up, 113 new events of MCI were diagnosed with an estimated incidence rate of 21.5 per 1,000 person-years. We found a progression rate to dementia (all causes) of 3.8/100 person-years. Specific progression rates for AD, VaD, and other types of dementia were 2.3, 1.3, and 0.3/100 person-years. Furthermore, age was a risk factor for incident MCI (RR: 5.93, 95% CI: 3.17 to 11.10), while education was protective (RR: 0.06, 95% CI: 0.03 to 0.10), and serum total cholesterol evidenced a borderline nonsignificant trend for a protective effect. There was a nonsignificant trend for stroke as a risk factor of progression of MCI to dementia. CONCLUSIONS: In this population, among those who progressed to dementia, 60% progressed to AD and 33% to VaD. Vascular risk factors influence incident mild cognitive impairment and the rate of progression to dementia.

Gender differentials in life expectancy in Italy.

OBJECTIVE: To analyze the gender differential in life expectancy in Italy from 1970 to 1997, to determine which ages and causes of death mostly influenced its trend. METHODS: Pollard method was used to explain which ages and which pathologies give the highest contribution to the gender differential. RESULTS: A slight reduction of the gender differential has been observed since 1980. The olds and very olds became the most relevant contributors to the differential. Cardiovascular diseases were the principal contributing causes of death until 1980 and Cancer replaced them afterwards. CONCLUSION: Relatively recent adoption of unhealthy life styles by women together with an opposite process run by young men towards healthier behaviour, seem to be at the basis of the slight reduction of the gender differential in life expectancy observed since 1980.

Dementia and disability: impact on mortality. The Italian Longitudinal Study on Aging.

Dementia is known to be associated with excess mortality. Physical disability, as a marker of dementia severity, is often considered the last step on the way from disease to death. The objective of this study was to investigate the direct effect of dementia on mortality in a population-based study, carried out in Italy, with a sample of 5,632 individuals aged 65-84 years. At 4-year follow-up, 998 participants had died. The independent predictors of death were: age (75-84 years; HR 2.63, CI = 2.11-3.27), male sex (HR 1.45, CI = 1.22-1.74), coronary heart disease (HR 1.61, CI = 1.34-1.94), moderate and severe instrumental activities of daily living disability (HR 1.98, CI = 1.30-3.03 and HR 3.26, CI = 2.09-5.09, respectively), diabetes in subjects with a survival time greater than 23 months (HR 0.68, CI = 0.43-1.08) and dementia (HR 2.07, CI = 1.62-2.66). These data provide evidence that dementia per se, independently from physical disability, is a strong predictor of death in the elderly.

Drinking pattern and mortality: the Italian Risk Factor and Life Expectancy pooling project.

PURPOSE: To analyze the relationship between an aspect of drinking pattern (i.e., drinking with or without meals) and risk of all-cause and specific-cause mortality. METHODS: The Risk Factors and Life Expectancy Study, is a pooling of a series of epidemiological studies conducted in Italy. Eight-thousand six-hundred and forty-seven men and 6521 women, age 30-59 at baseline, and free of cardiovascular disease, were followed for mortality from all causes, cardiovascular and noncardiovascular, during an average follow-up of 7 years. RESULTS: Drinkers of wine outside meals exhibited higher death rates from all causes, noncardiovascular diseases, and cancer, as compared to drinkers of wine with meals. This association was independent from the cardiovascular disease (CVD) risk factors measured at baseline and the amount of alcohol consumed and seemed to be stronger in women as compared to men. CONCLUSIONS: The present results indicate that drinking patterns may have important health implications, and attention should be given to this aspect of alcohol use and its relationship to health outcomes. The relationship between alcohol consumption and disease has been the focus of intensive scientific investigation (1-9). Most studies to date, however, have limitations. A major drawback is that limited information has been collected regarding the complex issue of alcohol consumption. In many studies, ascertainment of alcohol consumption frequently focused only on quantity of alcohol consumed without considering the many different components of alcohol consumption, particularly drinking pattern (10-12). It has been hypothesized, and preliminary data support the notion, that drinking pattern could have important influences on determining the health effects of alcohol (13,14). The present study examines the relationship between one aspect of drinking pattern (drinking wine outside meals) and mortality in a large cohort of men and women.

[Smoking and prevention of cardiovascular diseases in Italy. Research Group of the Cardiovascular Epidemiologic Observatory]. / Fumo e prevenzione delle malattie cardiovascolari in Italia. Gruppo di Ricerca dell'Osservatorio Epidemiologico Cardiovascolare.

BACKGROUND: Cigarette smoking is one of the most important risk factors for chronic diseases. Major epidemiological studies provide consistent evidence that cardiovascular risk is 3-fold among smokers compared to non-smokers, incidence increases with the number of cigarettes smoked daily and cardiovascular risk is less among subjects who have stopped smoking. METHODS: To support evidence of a time decreasing trend of smoking habits in Italy, data from two observational studies carried out in different Italian regions, the RIFLE Project (1978-1987), the Cardiovascular Epidemiological Observatory (1998), and data collected from the "Multiscopo" ISTAT Investigation on Italian families (1980-1997) have been analyzed. RESULTS: During the 1950's, the prevalence of smokers among adult men was about 80%, it decreased to 50% in the 1980's and now is about 30-40%. Contemporarily, the prevalence of smokers among adult women has increased, from 17% in the 1980's to 23% nowadays. The habit is more common in Central-Southern Italy compared to the North and is less frequent among married subjects, and male graduates. Learned women smoke more than those who have not received any education. Men begin smoking at 17-18 years, and this trend seems to have remained stable; on the contrary, women have changed their habits and now start smoking at 17 years compared to 25-30 in previous generations. Almost all smokers would like to stop smoking, more than 60% have tried at least once, but on average have lasted only for about 1 year. Passive smoking exposure varies from 1 hour in Northern Italy to 2 hours in the South. CONCLUSIONS: Although smoking habits, at least in men, reveal a decreasing trend, about 30-40% of adult men continue to smoke. Preventive intervention, especially focused on the younger generations, could be useful to contrast this attitude at an early stage.

Alcohol and survival in the Italian rural cohorts of the Seven Countries Study.

BACKGROUND: Middle-aged men who regularly drink a moderate amount of alcohol have lower mortality rates from all causes in comparison with abstainers and heavy drinkers. This cohort study looks at the relationship between alcohol consumption and long-term survival, adjusting for smoking habit and physical activity. METHODS: In 1965, a total of 1536 Italian males aged 45-65 years underwent an examination which included: a general questionnaire, anthropometric measurements, an overall physical examination, ECG recording, blood pressure and serum cholesterol measurements and measurement of food consumption including alcohol. The cohort was followed for total mortality from 1965 to 1995. RESULTS: During a period of 30 years 1096 deaths occurred. Age-adjusted life expectancy for men assuming a mean daily quantity of 63 g of alcohol (range 4-7 drinks per day) was 21.6 +/- 0.4 years, roughly 2 years more than men taking a mean quantity of 3.7 g (10 drinks per day. Taking smoking habit into account, the longest survival of 22.4 +/- 0.5 years was observed in non-smokers drinking 4-7 drinks daily; the lowest, 18.5 +/- 0.7 years, in smokers drinking >10 drinks. Stratifying for physical activity, the longest survival (23.4 +/- 0. 7 years) was experienced by men engaged in heavy physical activity at work drinking 1-4 drinks per day. CONCLUSIONS: The relationship between life expectancy and alcohol consumption (97% wine in this Italian cohort and mostly red wine) is confirmed to be non-linear. Men aged 45-64 at entry drinking about 5 drinks per day have a longer life expectancy than occasional and heavy drinkers.

The impact of the major causes of death on life expectancy in Italy.

BACKGROUND: This study aims to evaluate the contribution of the reduction in major cardiovascular diseases (CVD), malignant neoplasms (MN), accidents and AIDS mortality to the gains in life expectancy observed during the decade 1985-1994, as well as to calculate and compare the potential gains due to the partial or total elimination of these causes. METHODS: Mortality data from the Italian Mortality Data Base were analysed by the method of decomposition of changes in life expectancy and the partial multiple decrement life table technique. RESULTS: In Italy, considering the decade 1985-1994, the gain in life expectancy at birth was 2.27 years for men and 2.16 for women. The major contribution to this increase was the reduction in CVD mortality followed by fewer deaths from accidents and MN. Conversely, AIDS caused a loss in the length of life of adults. Major potential gains in life expectancy at birth could be obtained by the elimination or even partial reduction of CVD and MN mortality. When working life (15-64 years) is considered, the relative importance of the causes changes. The elimination of accidents and AIDS would result in increased life expectancy longer than that associated with a 50% reduction in CVD. CONCLUSIONS: The findings of this study provide useful information which could contribute to a more effective allocation of resources for research activity and public health programmes.

Premature mortality in Italy during the first decade of the AIDS epidemic: 1984-1993.

BACKGROUND: AIDS has become a leading cause of premature mortality in many countries, owing to the decline in other major causes of premature death and the increase in AIDS itself. This study was carried out to determine the trends in premature mortality due to selected causes in Italy. METHODS: Data from the Italian Mortality Data Base, for the ten years from 1984 to 1993 (the first decade of the AIDS epidemic) were analysed. Premature mortality was measured in terms of years of potential life lost before the age of 70 years (YPLL), excluding infant mortality. Trends in premature mortality due to AIDS were compared with those of the principal causes of premature death: lung cancer, colon-rectum cancer, stomach cancer, leukaemia, female breast cancer, uterine cancer, myocardial infarction, stroke, liver diseases, suicide, road accidents and overdose. RESULTS: In this period there has been a marked increase in premature mortality from AIDS both among males aged 1-69 years (from a rate of YPLL of 0.01 per 1000 in 1984 to 3.71 in 1993) and females of the same age group (from 0 deaths in 1984 to a rate of YPLL of 1.02). Throughout the same period all the other causes of premature death have been declining, with the exception of suicide and overdose among males, and overdose and lung cancer among females. For people aged 25-44 years, AIDS has become the greatest cause of premature death. The increasing trend in premature mortality due to AIDS is most pronounced in the northern and central areas of Italy. CONCLUSIONS: AIDS is the leading cause of death among males aged 25-44 years in Italy and is having an important impact on premature mortality among females in the same age group.

Helical CT angiography in gastrointestinal bleeding of obscure origin.

OBJECTIVE: The aim of the study was to verify the usefulness of helical CT angiography for diagnosis of gastrointestinal hemorrhage of obscure origin. SUBJECTS AND METHODS: Eighteen consecutive patients underwent catheterization of the abdominal aorta followed by helical CT angiography before and after intraarterial injections of a contrast medium. Helical CT angiography revealed the site of hemorrhage as an extravasation of contrast medium resulting in a hyperdense area in the intestinal lumen. All patients then underwent conventional angiography that was selective for the 13 patients in whom helical CT angiography was positive and standard for the remaining five patients with negative helical CT angiograms. RESULTS: Helical CT angiography revealed the site of hemorrhage in 13 (72%) of 18 patients. Diagnosis of bleeding site was confirmed, and the cause was established at surgery in 11 of these 13 patients. The site of bleeding for the remaining two patients was confirmed by angiography as angiodysplasia of the jejunum (n = 1) and of the colon (n = 1). Of the five patients with negative helical CT angiograms, location of bleeding was revealed by conventional angiography in two patients; one of the remaining three patients underwent exploratory laparotomy; and bleeding ceased in the remaining two patients, who were treated without surgery. Conventional angiography was negative for two of the 13 patients in whom helical CT angiography was positive. CONCLUSION: Despite our limited experience, helical CT angiography proved to be an easier and faster technique than conventional angiography for localizing gastrointestinal bleeding of obscure origin and useful as a guide for subsequent selective conventional angiography.

AIDS: a major health problem among young Italian women.

OBJECTIVE: To estimate the extent of the HIV/AIDS epidemic among Italian women and to assess its impact on mortality. Further, to assess the incidence of AIDS and the main characteristics of women affected by AIDS: with particular attention to those aged 25-34 years, and to identify differences in these characteristics compared with men infected with AIDS. METHODS: Incidence data were derived from the Italian National Register of AIDS Cases (RAIDS). The most recent complete data refer to cases diagnosed to the end of 1994. Mortality data were collected from the Italian mortality database, which is held by the Italian National Census Bureau (ISTAT) and processed by the Laboratory of Epidemiology and Biostatistics at the National Institute of Health of Italy (Istituto Superiore di Sanitá; ISS). The most recent mortality data refer to 1992, but projections were made to obtain mortality data up to 1994. All the rates were standardized by age and sex. RESULTS: A total of 26 957 cases of AIDS, 21 441 men (79.5%) and 5516 women (20.5%), diagnosed before the end of 1994 were reported to RAIDS. The majority of cases and deaths occurred in individuals aged 25-34 years. The incidence of AIDS among women increased from 2.1 per 100 000 in 1987 to 17.2 per 100 000 in 1994. The sex ratio (men:women) fell from 5.2 in 1987 to 2.9 in 1994. In 1992, AIDS was the second most common cause of death (slightly behind cancer and ahead of traffic accidents) in Italy, but was the most common cause in three regions of northern Italy (Liguria, Lombardia and Emilia). CONCLUSIONS: Although the AIDS epidemic in the past has predominantly affected young men, AIDS is now rapidly becoming a major health problem for young women in Italy. Moreover, as these women are at reproductive age, this may have important consequences in terms of increased morbidity and mortality among children.

[Spiral computed tomography with arteriography in the diagnosis of digestive system hemorrhages of obscure origin]. / La Tomografia Computerizzata spirale con arteriografia nella diagnosi delle emorragie dell'apparato digerente con origine oscura.

Some 5-20% of gastrointestinal (GI) bleeding cases are of obscure origin because bleeding is difficult to diagnose or locate with conventional imaging methods. The authors suggest arterial spiral CT (ASCT) of the abdomen as an original tool to study GI bleeding of obscure origin. This diagnostic method consists of two subsequent phases: first, after abdominal aorta catheterization, the catheter is positioned in the celiac artery; then, CT of the abdomen is performed before and after intra-arterial contrast agent injection. In 6 of 10 patients with GI bleeding of obscure origin, ASCT depicted small high-intensity areas in the intestinal lumen, which indicate a source of bleeding. The site of bleeding was studied only with topographic criteria according to scanning planes, and the morphological features of the intestinal segments were investigated. All patients were submitted to surgery: when ASCT identified a bleeding site, angiography was limited to the involved arterial district. Angiography alone could locate the source of bleeding in 5 patients; angiography and ASCT were in agreement in 4 patients. The combined techniques located the source of bleeding in 7/10 patients and the nature of bleeding was diagnosed in 3/10 patients (angiodysplasia). Even though their experience is limited, the authors believe that ASCT can be useful in locating GI bleeding of obscure origin, as well as in guiding subsequent selective angiography.

Relationship between eating patterns meeting recommendations and subsequent mortality in 20 years.

OBJECTIVE: Study of the relationships of conformity to dietary recommendations and subsequent rates of total mortality and mortality from specific causes. DESIGN: Prospective investigation of risk factors related to cardiovascular disease. SETTING: Two Italian rural cohorts of the Seven Countries Study. SUBJECTS: The study populations are defined samples from two villages, Crevalcore in Northern Italy and Montegiorgio in Central Italy. The examination in 1965, which included an individual dietary survey, involved 1538 men aged 45-64 years. The 20-year follow-up system was able to track every participant. RESULTS: Conformity to recommended levels of intake of carbohydrates (55-75% of calories) is associated to the relative risks: 0.84 (0.68-1.03) for total mortality, 0.73 (0.50-1.07) for cancer mortality, 1.34 (0.90-1.99) for coronary heart disease mortality. Relative risks for other nutrients are reported. CONCLUSIONS: Overall, recommended levels of intake of macronutrients are associated with lower total mortality, but are not equally appropriate for specific causes of death, cancer and coronary heart disease. Diet-associated differences in mortality persisted after adjustment for confounding by age, smoking habit and physical activity. In this population, intakes meeting the principal recommendations (total carbohydrate and total fat) seem to involve automatically meeting the other recommendations. SPONSORSHIP: Research partly supported by the project 'ACRO-clinical uses of Oncological Research' (contract 93.02289.PF39) and by the project 'FATMA-Risk Factors and Disease Control' (contract 93.00773.PF41) of the Italian National Research Council, CNR.

AIDS as a leading cause of death among young adults in Italy.

The objective of this study is to describe the impact of AIDS on the mortality of young adult (aged 25 to 44 years) in Italy, at both the national and regional level. We analyzed the official mortality data for Italy: the most recent data available being from 1990. General mortality trends show that while mortality among young women is still decreasing (i.e. from a standardized rate of 83.8 per 100,000 in 1980 to 68.4 in 1990), mortality among young men began to rise in the mid-1980s, after a steady decrease over many years. Among the 25-34 year olds, however, this reversal in trend is more marked, notwithstanding a decrease or stabilization in most major causes of death. In fact it coincides with the appearance and spread of AIDS in Italy, which has affected young men in particular. (The peak age group for AIDS deaths is the 25-34 year olds). Mortality data from 1990 reveal that AIDS is the fourth leading cause of death in Italy among men between the ages of 25 and 44 years. Among 25-34 year-old men, however, AIDS is the second leading cause of death, after road accidents. AIDS also contributes greatly to the general mortality in individual regions, both among 25-44 year-old men (Lombardy, Liguria, Lazio, Emilia-Romagna, Tuscany), and especially among 25-34 year-old men (Lombardy, Liguria, Lazio, Sardinia, where it is the number one cause of death, Emilia-Romagna, Tuscany, and Veneto, where it is the number two cause of death).

The prediction of coronary heart disease mortality as a function of major risk factors in over 30 000 men in the Italian RIFLE pooling Project. A comparison with the MRFIT primary screenees. The RIFLE research group.

BACKGROUND: Few risk functions for the prediction of coronary heart disease mortality have been produced in Italy. This study used a large population sample to evaluate the effect of major risk factors on coronary mortality. METHODS: Coronary deaths in 45 cohorts of men (n = 31317, aged 30-69 years) were studied and related to selected cardiovascular risk factors. RESULTS: After 6 years, 1089 men had died, of whom 239 were coronary fatalities. Univariate and multivariate (Cox model) analyses conducted on each age group (30-39, 40-49, 50-59, and 60-69 years) showed a positive association between coronary deaths and systolic blood pressure, serum cholesterol level and cigarette smoking, with few exceptions. A multiple logistic model was produced for men aged 35-57 years, assessing the role of age, serum cholesterol, cigarettes smoked per day and diastolic instead of systolic blood pressure, using the same endpoint as that employed in a similar model published from the analysis of MRFIT primary screenees in the USA to facilitate valid comparison. The coefficients in the present study were similar to those in the US cohort: no statistically significant differences could be detected when comparing the pairs of coefficients. CONCLUSION: Coefficients relating cholesterol, blood pressure and cigarette smoking to coronary mortality in Italian men are similar to those in American men from the same age groups.

Is diet an independent risk factor for mortality? 20 year mortality in the Italian rural cohorts of the Seven Countries Study.

The relation of diet to mortality is examined using the data of the Italian rural cohorts of the Seven Countries Study, a prospective investigation of factors related to cardiovascular disease. The present analysis includes 1536 men aged 45-64 years, whose dietary habits and food consumption, including alcoholic beverages, were measured in 1965. Of the 1536, 668 (43.5%) died during a follow-up period of 20 years. Large differences in survival probabilities were observed for different dietary patterns (i.e. for different intakes of energy and nutrients). The dietary pattern that corresponded to the lowest mortality rate (27% after 20 years) was: more than 2800 kcal/d (11.7 MJ/d), with more than 41% of the calories coming from carbohydrates, more than 9% from proteins, between 16% and 23% from unsaturated lipids, and between 13% and 19% from alcohol. As the number of disagreements from the previous conditions increases, the mortality rate increases to reach 69% in 20 years in the worst case. Differences in mortality persisted after adjustment for confounders and some major established mortality risk factors.