Risk factors for sciatica leading to hospitalization.

PURPOSE: To study the known or suspected risk factors for sciatica: Tallness, overweight, smoking, leisure-time physical exercise, self-reported health and occupation, and how they predict hospitalizations due to sciatica. Only a few cohort studies have previously focused on the risk factors for sciatica. METHODS: The 13,095 subjects, free from low back disorders at the baseline in 1973-1976 were followed up to the end of 2011 via the Care Register for Health Care. Along with an invitation to the health examination, a basic questionnaire concerning lifestyle factors was sent to participants. The outcome measure was incident sciatica leading to hospitalization. RESULTS: Altogether 702 incident sciatica cases occurred. Among men, the adjusted hazard ratio (HR) with 95% confidence interval (CI) was 2.57 (95% CI 1.47-4.50) in metal or machine work, and 1.44 (1.06-1.95) in other industrial work, compared to that in white-collar occupations. Among women, the corresponding risk estimates were 1.81 (1.18-2.78) for nurses and related occupations, 1.56 (1.05-2.31) for sales workers, and 1.46 (1.03-2.08) for industrial workers. Among men, physical exercise during leisure predicted a decrease in the risk of sciatica (0.74; 0.55-1.00); this association was significantly pronounced in white-collar occupations (0.38; 0.18-0.88). Among women, the association between body mass index and the risk of sciatica was only modest, but varied greatly between different occupations. CONCLUSIONS: Physically demanding work is a strong risk factor for sciatica. Leisure-time physical activity seems to protect men against sciatica, while overweight is a risk factor among women. However, occupation substantially modifies these associations.

Associations of tissue transglutaminase antibody seropositivity with coronary heart disease: Findings from a prospective cohort study.

BACKGROUND AND AIMS: Clinical experience and observational studies suggest that individuals with coeliac disease are at increased risk of coronary heart disease (CHD), but the precise mechanism for this is unclear. Laboratory studies suggest that it may relate to tissue transglutaminase antibodies (tTGAs). Our aim was to examine whether seropositivity for tTGA and endomysial antibodies (EMAs) are associated with incident CHD in humans. METHODS AND RESULTS: We used data from Mini-Finland Health Survey, a prospective cohort study of Finnish men and women aged 35-80 at study baseline 1978-80. TTGA and EMA seropositivities were ascertained from baseline blood samples and incident CHD events were identified from national hospitalisation and death registers. Cox regression was used to examine the associations between antibody seropositivity and incident CHD. Of 6887 men and women, 562 were seropositive for tTGAs and 72 for EMAs. During a median follow-up of 26 years, 2367 individuals experienced a CHD event. We found no clear evidence for an association between tTGA positivity (hazard ratio, HR: 1.04, 95% confidence interval, CI: 0.83, 1.30) or EMA positivity (HR: 1.16, 95% CI: 0.77, 1.74) and incident CHD, once pre-existing CVD and known CHD risk factors had been adjusted for. CONCLUSION: We found no clear evidence for an association of tTGA or EMA seropositivity with incident CHD outcomes, suggesting that tTG autoimmunity is unlikely to be the biological link between coeliac disease and CHD.

The outcome of short- and long-term psychotherapy 10 years after start of treatment.

BACKGROUND: Empirical evidence on whether patients' mental health and functioning will be more improved after long-term than short-term therapy is scarce. We addressed this question in a clinical trial with a long follow-up. METHOD: In the Helsinki Psychotherapy Study, 326 out-patients with mood or anxiety disorder were randomly assigned to long-term psychodynamic psychotherapy (LPP), short-term psychodynamic psychotherapy (SPP) or solution-focused therapy (SFT) and were followed for 10 years. The outcome measures were psychiatric symptoms, work ability, personality and social functioning, need for treatment, and remission. RESULTS: At the end of the follow-up, altogether 74% of the patients were free from clinically elevated psychiatric symptoms. Compared with SPP, LPP showed greater reductions in symptoms, greater improvement in work ability and higher remission rates. A similar difference in symptoms and work ability was observed in comparison with SFT after adjustment for violations of treatment standards. No notable differences in effectiveness between SFT and SPP were observed. The prevalence of auxiliary treatment was relatively high, 47% in SFT, 58% in SPP and 33% in LPP, and, accordingly, the remission rates for general symptoms were 55, 45 and 62%, respectively. CONCLUSIONS: After 10 years of follow-up, the benefits of LPP in comparison with the short-term therapies are rather small, though significant in symptoms and work ability, possibly due to more frequent use of auxiliary therapy in the short-term therapy groups. Further studies should focus on the choice of optimal length of therapy and the selection of factors predicting outcome of short- v. long-term therapy.

Drug-gene interactions and the search for missing heritability: a cross-sectional pharmacogenomics study of the QT interval.

Variability in response to drug use is common and heritable, suggesting that genome-wide pharmacogenomics studies may help explain the 'missing heritability' of complex traits. Here, we describe four independent analyses in 33 781 participants of European ancestry from 10 cohorts that were designed to identify genetic variants modifying the effects of drugs on QT interval duration (QT). Each analysis cross-sectionally examined four therapeutic classes: thiazide diuretics (prevalence of use=13.0%), tri/tetracyclic antidepressants (2.6%), sulfonylurea hypoglycemic agents (2.9%) and QT-prolonging drugs as classified by the University of Arizona Center for Education and Research on Therapeutics (4.4%). Drug-gene interactions were estimated using covariable-adjusted linear regression and results were combined with fixed-effects meta-analysis. Although drug-single-nucleotide polymorphism (SNP) interactions were biologically plausible and variables were well-measured, findings from the four cross-sectional meta-analyses were null (Pinteraction>5.0 × 10(-8)). Simulations suggested that additional efforts, including longitudinal modeling to increase statistical power, are likely needed to identify potentially important pharmacogenomic effects.

A cohort study on diet and the risk of Parkinson's disease: the role of food groups and diet quality.

Previous studies on individual foods and nutrients and Parkinson's disease (PD) risk have been inconsistent. Furthermore, only one study has examined the association between the quality of diet and PD. We investigated the prediction of food groups and diet quality on PD in the Finnish Mobile Clinic Survey (1966-72). The population comprised 4524 individuals, aged 40-79 years and free from PD at baseline. Data collection included health examinations, a questionnaire and a 1-year dietary history interview. A modified Alternate Healthy Eating Index was formed to assess diet quality. Statistical analyses were based on Cox's model. During a 41-year follow-up, eighty-five incident cases of PD occurred. No statistically significant associations were found between PD incidence and most of the food groups examined. A few exceptions were fruits and berries in men and milk in women, which showed positive associations. An inverse association between the intake of meat products and PD was found in women. The diet quality index did not predict PD, the adjusted relative risk between the highest and lowest quartiles being 1.83 (95 % CI 0.65, 5.18) in men and 0.97 (95 % CI 0.38, 2.48) in women. The present study suggests that since most of the single food groups or the quality of diet did not predict PD occurrence, the role of diet is apparently rather modest.

Association of 25-hydroxyvitamin D with the incidence of knee and hip osteoarthritis: a 22-year follow-up study.

OBJECTIVE: In the light of conflicting results from previous studies on the role of vitamin D, we studied serum 25-hydroxyvitamin D [25(OH)D] with regard to its prediction of incident knee and hip osteoarthritis (OA). METHODS: The study population (n = 805) consisted of participants of a national health examination survey who had undergone baseline and follow-up clinical examinations at intervals of 20-23 years. Knee and hip OA were diagnosed on the basis of a standardized clinical examination by physicians with the same diagnostic criteria at baseline and follow-up. Information on covariates, including age, sex, season of blood draw, education, body mass index (BMI), physical workload, leisure time physical activity, smoking history, and previous injuries, was collected at baseline. Serum 25(OH)D concentrations were determined from baseline serum samples kept frozen at -20°C. RESULTS: We found no significant association between serum 25(OH)D level and the risk of incident knee or hip OA. However, a statistically significant interaction between season of blood draw and serum 25(OH)D emerged when predicting the development of definite knee OA (p = 0.004). After adjusting for all the covariates, the relative odds (95% confidence interval) of developing definite knee OA per increment of 1 SD (20.7 ng/mL) in winter season 25(OH)D was 1.57 (1.10-2.27), whereas for summer season sera the corresponding rate was 0.53 (0.28-1.00). CONCLUSION: The results do not support the hypothesis that a low level of serum 25(OH)D contributes to the development of OA. Instead, our study suggests that season is a potent effect modifier of 25(OH)D, which merits attention in future research.

Parity and risk of hip fracture in postmenopausal women.

UNLABELLED: Hip fracture risk was assessed according to parity among postmenopausal women. Compared with nulliparous women, the fracture risk was lower in women with three or more births. INTRODUCTION: Parity was assessed for long-term prediction of hip fracture in postmenopausal women. METHODS: Postmenopausal women (n= 2,028) aged 45 or over with no history of hip fracture were studied. From 1978 to 1980, all of them had participated in a comprehensive health survey based on a nationally representative population sample. Emerging cases of hip fracture were identified from the National Hospital Discharge Register during a follow-up period extending up to 17 years. RESULTS: The risk of hip fracture was lower among parous women compared with nulliparous women. The model adjusted for age showed a significant inverse association between parity as a continuous variable and the risk of hip fracture [RR = 0.74; 95% confidence interval (CI), 0.61-0.90] per an increment of one standard deviation (2.4 births). Adjusted for age, menopausal age, level of education, body mass index, vitamin D status, alcohol consumption, smoking history, leisure time physical activity, and self-rated health, the relative risk was 0.50 (95% CI, 0.32-0.79) for women with three or more births and 0.85 (95% CI, 0.55-1.32) for women with one to two births as compared with nulliparous women. CONCLUSION: Parity, three or more births in particular, predicts a lowered risk of hip fracture in the long run.

Severity of vertebral fracture and risk of hip fracture: a nested case-control study.

UNLABELLED: Severe vertebral fractures strongly predicted subsequent hip fracture in this population-based study. Such high-risk patients should be provided with clinical evaluation and care for osteoporosis. INTRODUCTION: Vertebral fractures are commonly osteoporotic and known to predict hip fracture. The aim of this study was to evaluate associations between the severity of vertebral fractures and the risk of subsequent hip fracture. METHODS: Chest radiographs were obtained of 7,095 Finnish men and women aged 30 years or over in the Mini-Finland Health Survey in 1978-1980. Record linkage to the National Hospital Discharge Register identified 182 subjects from the survey who had subsequently been hospitalized for primary treatment of hip fracture by the end of 1994. A nested case-control setting was adopted, where three controls individually matched for age, gender, and place of residence were drawn for 169 subjects with hip fracture from the same cohort. Baseline vertebral fractures were identified at levels T3 to T12, and their morphology was categorized to mild, moderate, or severe according to Genant's classification. RESULTS: Severe vertebral fracture (>40% reduction in vertebral body height) strongly predicted hip fracture. After controlling for education, physical activity, smoking, alcohol consumption, and self-rated general health, the adjusted relative odds was 12.06 (95% confidence interval, 3.80-38.26). Mild to moderate fracture grades and the number of compressed vertebral bodies showed no prediction for hip fracture. CONCLUSIONS: The presence of a severe vertebral fracture in the thoracic spine strongly predicts subsequent hip fracture. Such high-risk patients should be clinically evaluated and provided with care for osteoporosis and measures to reduce the risk of falling as required.

Estimation of population attributable fraction (PAF) for disease occurrence in a cohort study design.

The population attributable fraction (PAF) is a useful measure for describing the expected change in an outcome if its risk factors are modified. Cohort studies allow researchers to assess the predictive value of the risk factor modification on the incidence of the outcome during a certain follow-up. Estimation of PAF for both mortality and morbidity in cohort studies with censored survival data has been developed in the recent years. So far, however, censoring due to death in the estimation of PAF for morbidity has been ignored, resulting in estimation of a quantity which is not relevant in practice as some people are likely to die during the follow-up. The risk factors related to the disease incidence may also be related to mortality, and modification of these risk factors is likely to delay the occurrence of both events. Thus, censoring due to death and the impact of risk factor modification must be considered when estimating PAF for disease incidence. We consider both and introduce two measures of disease burden: PAF for the incidence of disease during lifetime and PAF for the prevalence of disease in the population at a certain time. We demonstrate how consideration of censoring due to death changes the estimated PAF for disease incidence and its confidence interval. This underlines the importance of choosing a correct PAF measure depending on the outcome of interest and the risk factors of interest to obtain accurate and interpretable results.

Malignancies in cases with screening-identified evidence of coeliac disease: a long-term population-based cohort study.

BACKGROUND AND AIMS: The association between diagnosed coeliac disease and malignancy has been established. The present study was conducted to determine whether previously unrecognised and thus untreated adults with screening-identified evidence of coeliac disease carry an increased risk of malignancies. METHODS: A Finnish population-based adult-representative cohort of 8000 individuals was drawn in 1978-1980. Stored sera of the participants with no history of coeliac disease or any malignancy were tested for immunoglobulin A (IgA) class tissue transglutaminase antibodies (Eu-tTG) in 2001. Positive sera were further analysed by another tissue transglutaminase antibody test (Celikey tTG) and for endomysial antibodies (EMAs). Malignant diseases were extracted from the nationwide database and antibody-positive cases were compared with negative cases during a follow-up of nearly 20 years. RESULTS: Altogether 565 of all the 6849 analysed serum samples drawn in 1978-80 were Eu-tTG positive. In further analyses, 202 (2.9%) of the participants were Celikey tTG positive and 73 (1.1%) were EMA positive. The overall risk of malignancy was not increased among antibody-positive cases in the follow-up of two decades; the age- and sex-adjusted relative risk was 0.91 (95% CI 0.60 to 1.37) for those who were Celikey tTG positive and 0.67 (95% CI 0.28 to 1.61) for those who were EMA positive. CONCLUSIONS: The prognosis of adults with unrecognised coeliac disease with positive coeliac disease antibody status is good as regards the overall risk of malignancies. Thus, current diagnostic practice is sufficient and there is no need for earlier diagnosis of coeliac disease by mass screening on the basis of the findings of this study.

Risk factors for the development of hip osteoarthritis: a population-based prospective study.

OBJECTIVES: Potential risk factors and their roles in the aetiology of hip OA are poorly understood. We analysed several alleged risk factors predisposing to hip OA in a 22-yr prospective study. METHODS: A comprehensive health survey was carried out in 1978-80 in a nationally representative sample of adult Finns. In 2000-01, 1286 participants in that survey were invited for re-examination, and 909 agreed to participate. After excluding those with hip OA at the baseline and those who were no longer working, a total of 840 subjects constituted the present study population. Hip OA was diagnosed on the basis of a standardized clinical examination by physicians who applied uniform criteria both at the baseline and at the re-examination phase. RESULTS: After 22 yrs of follow-up, hip OA was diagnosed in 41 subjects (4.9%). Heavy manual labour predicted the risk of developing hip OA [adjusted odds ratio (OR) 6.7; 95% CI 2.3, 19.5]. Permanent damage as a consequence of any musculoskeletal injury was also an independent predictor of hip OA (adjusted OR 5.0; 95% CI 1.9, 13.3). BMI, smoking, alcohol intake and leisure time physical activity were not factors which were predictive for hip OA. CONCLUSION: Heavy physical stress at work and major musculoskeletal injuries are associated with an increased risk of developing clinically diagnosed hip OA.

Self-report dieting and long-term changes in body mass index and waist circumference.

OBJECTIVE: This prospective study explores whether dieting attempts and previous changes in weight predict changes in body mass index (BMI) and waist circumference (WC). METHODS: The study was based on the representative Finnish Health 2000 Survey and on its follow-up examination 11 years later. The sample included 2,785 participants, aged 30-69. BMI and WC were determined at health examinations. Information on dieting attempts and previous changes in weight was collected using a questionnaire including questions on whether participant had tried to lose weight (no/yes), gained weight (no/yes) or lost weight (no/yes) during the previous year. RESULTS: At baseline, 32.8% were dieters. Of these, 28.4% had lost weight during the previous year. Dieters had higher BMI and WC than non-dieters. During the follow-up, the measures increased more in dieters and in persons with previous weight loss. The mean BMI changes in non-dieters versus dieters were 0.74 (standard deviation [SD] 2.13) kg/m2 and 1.06 (SD 2.77) kg/m2 (P = 0.002), respectively. The corresponding numbers for those with no previous weight change versus those who had lost weight were 0.65 (SD 2.07) kg/m2 and 1.52 (SD 2.61) kg/m2. The increases in BMI and WC were most notable in dieters with initially normal weight. CONCLUSIONS: The increases in BMI and WC were greater in dieters than in non-dieters, suggesting dieting attempts to be non-functional in the long term in the general population.

Physical work and chronic shoulder disorder. Results of a prospective population-based study.

OBJECTIVES: A study of whether occupational physical load predicted subsequent chronic shoulder disorders. METHODS: A comprehensive national survey was carried out among a representative sample (n = 7217) of the Finnish adult population in 1977-80. Twenty years later, 1286 participants from the previous survey were invited to be re-examined, and 909 (71%) participated. After excluding those with diagnosed shoulder disorders at baseline, 883 subjects were available for the analyses. RESULTS: At follow-up, a physician diagnosed chronic shoulder disorders in 63 subjects (7%) using a standardised protocol. Work exposure to repetitive movements and vibration at baseline increased the risk of shoulder disorder: adjusted ORs 2.3 (95% CI 1.3 to 4.1) and 2.5 (1.2 to 5.2), respectively. Exposure to several physical factors increased the risk further, the adjusted OR was nearly 4 for at least three exposures. The adverse effects of physical work were seen even among those older than 75 years at follow-up. The statistically significant risk factors differed between genders: for men vibration and repetitive movements, and for women lifting heavy loads and working in awkward postures. Age and body mass index modified the effects of the physical exposures. The results remained similar after excluding those with any shoulder pain at baseline. CONCLUSIONS: This is the first prospective study in a general population showing that occupational physical loading increases the risk of a subsequent clinical shoulder disorder and the effects seem to be long-term. Early preventive measures at the workplace may have long-lasting health benefits for the shoulder.

Prospective study of coffee consumption and risk of Parkinson's disease.

OBJECTIVE: To examine the prediction of coffee consumption on the incidence of Parkinson's disease. SUBJECTS AND METHODS: The study population comprised 6710 men and women, aged 50-79 years and free from Parkinson's disease at the baseline. At baseline, enquiries were made about coffee consumption in a self-administered questionnaire as the average number of cups per day. During a 22-year follow-up, 101 incident cases of Parkinson's disease occurred. Parkinson's disease cases were identified through a nationwide registry of patients receiving medication reimbursement, which is based on certificates from neurologist. RESULTS: After adjustments for age, sex, marital status, education, community density, alcohol consumption, leisure-time physical activity, smoking, body mass index, hypertension and serum cholesterol, the relative risk for subjects drinking 10 or more cups of coffee per day compared with non-drinkers was 0.26 (95% confidence interval 0.07-0.99, P-value for trend=0.18). The association was stronger among overweight persons and among persons with lower serum cholesterol level (P-value for interaction=0.04 and 0.03, respectively). CONCLUSIONS: The results support the hypothesis that coffee consumption reduces the risk of Parkinson's disease, but protective effect of coffee may vary by exposure to other factors.

IgA antibodies in persisting Helicobacter pylori infection in Finnish adults.

Most individuals infected with Helicobacter pylori have elevated levels of specific IgG antibodies, but only in about two-thirds of cases does the IgA titre exceed the cut-off level. The aim of this study was to determine whether H. pylori-infected subjects with elevated IgG levels would subsequently produce IgA antibodies, and whether elevated IgA levels increased during infection. Paired sera were available from 336 adults who took part in a large population-based health survey in 1977-1980 and a follow-up study on asthma and atopic diseases in 1997-1998 (series A). Data on paired sera from 224 adults who participated in a population-based health survey in Vammala, Finland in 1973 and who gave a follow-up blood sample in 1994 (series B) were also re-analysed. H. pylori IgG and IgA levels were determined with commercially available (series A) and in-house (series B) enzyme immunoassays. Twenty-one (35%) of the 60 subjects who initially had elevated levels of IgG antibodies only were found to be IgA-positive at follow-up. In those subjects whose baseline and follow-up samples were IgG- and IgA-positive, the median IgA levels increased by 48% and 22% in series A and B (p < 0.0001 and p 0.0241), respectively, whereas the median IgG levels did not change significantly in either series. During H. pylori infection, an increase in specific IgA was reflected by the increase in the number of responders and by the rise in titres.

Serum alpha-tocopherol concentration in relation to subsequent colorectal cancer: pooled data from five cohorts.

BACKGROUND: Vitamin E is an antioxidant that inhibits mutagenesis and cell transformation. Previous findings in five prospective epidemiologic studies suggested that the level of serum alpha-tocopherol, the predominant form of vitamin E in the blood, was lower in subjects who subsequently developed colorectal cancer than in control subjects. However, the difference was neither obvious nor statistically significant in any one of these five studies. PURPOSE: To evaluate in greater detail the association between serum alpha-tocopherol concentration and risk of colorectal cancer, we pooled and analyzed the original data from the five studies. Our analyses were designed to (a) test the hypothesis with greater statistical power, (b) examine the association after adjustment for serum cholesterol levels, and (c) evaluate the association after uniform exclusion of cases diagnosed shortly after blood specimens were drawn. METHODS: Data for individual subjects were analyzed. To make the design of the component investigations uniformly nested case-control studies with individual matching, we matched controls to cases in two of the cohorts. Subjects were categorized according to study-specific quartile of serum alpha-tocopherol level within the study. The pooled analysis included 289 cases of colorectal cancer and 1267 matched controls. RESULTS: For cancers of the colon and rectum combined, the matched odds ratio (OR) for the highest quartile of serum alpha-tocopherol concentration compared with the lowest was 0.6 (95% confidence interval [CI] = 0.4-1.0). Adjustment for serum cholesterol level attenuated the OR to 0.7 (95% CI = 0.4-1.1). CONCLUSION: The results suggest that serum alpha-tocopherol concentration may be inversely related to risk of colorectal cancer. It is unclear whether an association exists, however, because the association between serum alpha-tocopherol level and decreased risk of colorectal cancer was modest, the CIs were wide, and, overall, the tests for trend in effect were not significant. IMPLICATIONS: Larger observational studies with concurrent dietary data are needed to determine whether vitamin E has a modest but potentially important protective effect against colorectal cancer.

Serum selenium and subsequent risk of cancer among Finnish men and women.

The association between the serum selenium level and the subsequent incidence of cancer was investigated in a longitudinal study of 39,268 men and women participating in the Social Insurance Institution's Mobile Clinic Health Examination Survey in Finland. The baseline examinations, including the collection of blood samples, were performed in 1968-1972. During a median follow-up of 10 years, 1,096 new cancer cases were identified from the files of the Finnish Cancer Registry. Selenium concentrations were measured from the stored serum samples collected from these cancer cases and from two controls per case, matched for sex, municipality, and age. The mean serum selenium level was 59.1 micrograms/L among all male cancer cases and 62.5 micrograms/L among controls. The difference was statistically significant (P less than .001). Corresponding values among women were 63.6 and 63.9 micrograms/L, respectively. Low serum selenium levels were associated with an increased risk of developing cancer at several sites, especially cancers of the stomach and lung among men. The relative risk of lung cancer between the highest and lowest decile of serum selenium was 0.11, and it differed significantly from unity (P less than .001). These findings are in agreement with the hypothesis that low selenium intake may increase the risk of some cancers among men.

Food consumption and the incidence of type II diabetes mellitus.

OBJECTIVE: The consumption of different foods was studied for their ability to predict type II diabetes mellitus. DESIGN: The study design was a cohort study, based on the Finnish Mobile Clinic Health Examination Survey. SETTING: A total of 30 communities from different parts of Finland. SUBJECTS: A total of 4304 men and women, 40-69 y of age and free of diabetes at baseline in 1967-1972 and followed up for incidence of diabetes medication during 23 y (383 incident cases). RESULTS: Higher intakes of green vegetables, fruit and berries, oil and margarine, and poultry were found to predict a reduced risk of type II diabetes. The relative risks of developing type II diabetes between the extreme quartiles of the intakes were 0.69 (95% confidence interval (CI) = 0.50-0.93; P for trend (P) = 0.02) for green vegetables, 0.69 (CI = 0.51-0.92; P = 0.03) for fruit and berries, 0.71 (CI = 0.52-0.98; P = 0.01) for margarine and oil, and 0.71 (CI = 0.54-0.94; P = 0.01) for poultry. CONCLUSION: The results suggest that prevention of type II diabetes might be aided by consumption of certain foods that are rich in nutrients with hypothesized health benefits.

Hyperhomocystinemia: a risk factor or a consequence of coronary heart disease?

BACKGROUND: Mild hyperhomocystinemia has been suggested as an indicator of an increased risk of cardiovascular disease. OBJECTIVE: To examine whether serum homocysteine concentration is a predictor of coronary heart disease (CHD) events. METHODS: A case-control study, nested in a population-based cohort study was used. During a follow-up of 13 years, 166 major coronary events (death from CHD or nonfatal myocardial infarction) occurred in men with evidence of heart disease at baseline and 272 events in men without a history of heart disease. Two controls per case were selected by individual matching. RESULTS: Among men with known heart disease at baseline, the relative risk (95% confidence interval) of CHD events adjusted for age, smoking, hypertension, diabetes mellitus, serum cholesterol level, body mass index, and alcohol consumption was 2.23 (95% confidence interval, 1.03-4.85) in the highest serum homocysteine quintile compared with the lowest quintile. Among the men free of heart disease at baseline, the corresponding relative risk was 0.90 (95% confidence interval, 0.51-1.60). CONCLUSIONS: This prospective study does not support the hypothesis that a high concentration of serum homocysteine is a risk factor for coronary events in a population free of heart disease. However, it does suggest that mild hyperhomocystinemia predicts secondary coronary events in men with heart disease, possibly as a consequence of atherosclerotic changes.