Thorkild I. A. Sørensen

Long-term influences of body-weight changes, independent of the attained weight, on risk of impaired glucose tolerance and Type 2 diabetes.

Aim  To investigate if weight gain during adulthood has effects on the risk of developing impaired glucose tolerance (IGT) or Type 2 diabetes beyond effect of attained weight.

The Danish National Birth Cohort - its background, structure and aim

Background: It is well known that the time from conception to early childhood has importance for health conditions that reach into later stages of life. Recent research supports this view, and diseases such as cardiovascular morbidity, cancer, mental illnesses, asthma, and allergy may all have component causes that act early in life. Exposures in this period, which infl uence fetal growth, cell divisions, and organ functioning, may have long-lasting impact on health and disease susceptibility. Methods: To investigate these issues the Danish National Birth Cohort (Better health for mother and child) was established. A large cohort of pregnant women with long-term follow-up of the offspring was the obvious choice because many of the exposures of interest cannot be reconstructed with sufficient validity back in time. The study needs to be large, and it is aimed to recruit 100,000 women early in pregnancy, and to continue follow-up for decades. The Nordic countries are better suited for this kind of research than most other countries because of their population-based registers on diseases, demography and social conditions, linkable at the individual level by means of the unique ID-number given to all citizens. Exposure information is mainly collected by computer-assisted telephone interviews with the women twice during pregnancy and when their children are six and 18 months old. Participants are also asked to fill in a self-administered food frequency questionnaire in mid-pregnancy. Furthermore, a biological bank has been set up with blood taken from the mother twice during pregnancy and blood from the umbilical cord taken shortly after birth. Data collection started in 1996 and the project covered all regions in Denmark in 1999. By August 2000, a total of 60,000 pregnant women had been recruited to the study. It is expected that a large number of gene-environmental hypotheses need to be based on case-control analyses within a cohort like this.

The levelling off of the obesity epidemic since the year 1999 - a review of evidence and perspectives.

The purpose was to investigate a possible levelling off in the obesity epidemic, by systematically reviewing literature and web‐based sources. Eligible studies and data sources were required to have at least two measures of obesity prevalence since 1999. A literature and Internet search resulted in 52 studies from 25 different countries. The findings supported an overall levelling off of the epidemic in children and adolescents from Australia, Europe, Japan and the USA. In adults, stability was found in the USA, while increases were still observed in some European and Asian countries. Some evidence for heterogeneity in the obesity trends across socioeconomic status (SES) groups was found. The levelling off was less evident in the lower‐SES groups. No obvious differences between genders were identified. We discussed potential explanations for a levelling off and the utility of investigating obesity trends to identify the driving forces behind the epidemic. It is important to emphasize that the levelling off is not tantamount to calling off the epidemic. Additionally, it is worthwhile to keep in mind that previous stable phases have been followed by further increases in the prevalence of obesity. Therefore, research into the causes, prevention and treatment of obesity should remain a priority.

Type of Alcohol Consumed and Mortality from All Causes, Coronary Heart Disease, and Cancer

Several population studies from different countries have shown a J-shaped relation between intake of alcohol and mortality from all causes (1-6). Studies comparing different countries have found a strong inverse relation between incidence rates of coronary heart disease and wine consumption but a weak or nonexistent relation for consumption of beer or spirits (7-9). The findings that different types of alcoholic beverages have different effects on mortality are indirectly supported by several clinical and experimental studies (10-12). In contrast, prospective studies have shown that beer (13), spirits (14), and wine (15) may have protective effects. However, most of these investigations were based on populations with one predominant type of alcohol consumption; this precluded valid comparison of the effects of the three different types of alcohol. We sought to analyze the effect of intake of different types of alcohol on mortality from all causes, coronary heart disease, and cancer in several large Danish cohort studies. Methods The Copenhagen Centre for Prospective Population Studies is based on three study samples: that of the Copenhagen City Heart Study; that of the Copenhagen County Centre of Preventive Medicine (the former Glostrup Population Studies), which includes six cohorts; and that of the Copenhagen Male Study (16-18). The study samples of the Copenhagen City Heart Study, initiated in 1976, and the Copenhagen County Centre of Preventive Medicine, initiated in 1964, were randomly selected within age strata from the populations residing in defined areas in greater Copenhagen. For the Copenhagen Male Study, initiated in 1971, employees of 14 large companies in Copenhagen were invited to participate. The mean participation rate in all studies was 80% (range, 69% to 88%). The combined study sample comprises 13 064 men and 11 459 women for whom information on alcohol intake and lifestyle-related variables, described below, was complete. Alcohol Intake Participants of the Copenhagen City Heart Study and the studies in the Copenhagen County Centre of Preventive Medicine were asked about their current average weekly intake of beer, wine, and spirits. In the Copenhagen Male Study, participants were asked about their average daily intake of beer, wine, and spirits on weekdays (Monday through Thursday) and weekends (Friday through Sunday); these reports were combined to estimate weekly alcohol consumption. Persons in our study who did not drink alcohol because they were receiving disulfiram or other medication were excluded from the analysis. One bottle of beer contains 11.6 g of alcohol, and 12 g is an approximate average for one serving of wine or spirits. We grouped participants into five categories on the basis of total intake of alcohol: less than 1 drink/wk (nondrinkers), 1 to 7 drinks/wk, 8 to 21 drinks/wk, 22 to 35 drinks/wk, and more than 35 drinks/wk. Intake of beer, wine, and spirits was categorized similarly; however, because of the frequency of end point data, more than 21 drinks/wk is the highest intake category for the individual types of beverages. Smoking Status Participants reported whether they were never-smokers, former smokers, or current smokers. Current smokers reported grams of tobacco smoked per day in the form of cigarettes (1 g/d), small cigars (3 g/d), cigars (5 g/d), and pipe tobacco (50 g/package). Five groups were defined: never-smokers, former smokers, smokers of 1 to 14 g of tobacco daily, smokers of 15 to 24 g of tobacco daily, and smokers of more than 24 g of tobacco daily. Education Participants reported the number of years that they attended school. Three groups were defined: fewer than 8 years, 8 to 11 years, and 12 or more years of school education. Physical Activity Participants reported whether they were physically active during leisure time. Four groups were defined: sedentary (<2 h/wk), light activity (2 to 4 h/wk), moderate activity (>4 h/wk, noncompetitive) and heavy activity (>4 h/wk, competitive). Body Mass Index Body weight and height were measured while the participant was wearing light clothes and no shoes. Body mass index was calculated as weight in kg divided by height in meters squared. Five categories of body mass index were defined: less than 20.0 kg/m2, 20.0 to 24.9 kg/m2, 25.0 to 29.9 kg/m2, 30.0 to 34.9 kg/m2, and 35.0 kg/m2 or more. Changes in Lifestyle-Related Variables When participants were re-examined during follow-up, the newly obtained values for alcohol intake, smoking status, physical activity, and body mass index were used to replace the old values in the statistical analyses. Observation time and vital status were included in the modeling accordingly. Follow-up Participants were followed from date of entry into the study to date of death, loss to follow-up, emigration, or end of follow-up, whichever came first. The vital status of populations was followed by using each participants unique identification number in the national Central Person Register until 9 January 1995. Fewer than 1% of the participants were lost to follow-up. Causes of death were obtained from the National Board of Health and were defined by using codes from International Classifications of Diseases, Eighth Revision (codes 410.0 to 414.0 for coronary heart disease and codes 140.0 to 209.0 for cancer). According to a previous study, the reported diagnoses for these grouped codes have proven to be sufficiently valid (20). Statistical Analysis We performed Poisson regression (21) by using SAS/STAT software (22) to estimate the effect of alcohol intake on the risk for death. These models generate estimates of relative risk that are adjusted for confounders. Each model included the following potential confounders as categorical variables: age, cohort study, sex, education, body mass index, physical activity, and smoking status. Owing to collinearity, it was impossible to include both the amount by type of beverage (beer, wine, or spirits) consumed and total alcohol intake in the same regression. We therefore estimated the influence of alcohol according to number of drinks consumed per week [0, 1 to 7, 8 to 21, 22 to 35,>35] in three regressions: 1) total alcohol consumption in drinks per week, without considering beverage type; 2) alcohol consumption in drinks of each beverage per week, without considering the total intake; and 3) percentage of total alcohol intake consumed as wine (0%, 1% to 30%,>30%). Effects that were insignificant according to the likelihood ratio test (5% level) were removed by backward elimination. A term indicating interaction between total alcohol intake and percentage alcohol consumed as beer, wine, or spirits was included in the analyses to assess whether the effects of beer, wine, and spirits differed at different levels of total alcohol intake; no such effects were identified, as judged from the fit of the model. Likewise, no interaction was found between sex and intake of different types of beverage in terms of mortality. Results A total of 4275 women and 1635 men drank less than 1 drink per week; 64 women and 1032 men drank 35 or more drinks per week. Of 13 613 participants who drank alcohol, 12 846 (69%) included wine in their intake (Table 1). During 257 859 person-years of follow-up, 4833 participants died; of these, 1075 died of coronary heart disease and 1552 died of cancer. Table 1. Baseline Characteristics of the Study Participants Baseline Characteristics Compared with participants who drank alcohol but no wine, those for whom wine made up more than 30% of their total alcohol intake were more likely to be women and have a higher educational level but were less likely to be smokers (Table 1). Participants for whom wine made up more than 30% of their alcohol intake were similar to those who drank no alcohol in terms of smoking habits, body mass index, and physical activity. Across categories of total alcohol intake, mean alcohol intake within the different categories of wine drinking was similar; for example, among participants who drank 8 to 21 drinks/wk, those who drank no wine, those who drank 1% to 30% wine, and those who drank more than 30% wine had a mean alcohol intake of 13.3, 13.7, and 12.8 drinks/wk, respectively. However, among light drinkers (1 to 7 drinks/wk), those who drank 1% to 30% of their alcohol as wine had a slightly higher mean intake than did those who avoided wine and those who drank more than 30% of their alcohol intake as wine. Thus, assessment of the effects of wine intake may not be subject to residual confounding by total alcohol intake when controlled for as specified. Total Alcohol Intake and Mortality We found J-shaped relations between total alcohol intake and all-cause mortality in the three substudies. Pooled analyses also revealed J-shaped relations (Table 2). When nondrinkers were used as the reference group (relative risk, 1.00), intake of 1 to 7 drinks per week carried a relative risk of 0.82 (95% CI, 0.76 to 0.88) and intake of more than 35 drinks per week carried a relative risk of 1.10 (CI, 0.95 to 1.26). Alcohol intake was negatively related to death from coronary heart disease and positively related to death from cancer (Table 2). Table 2. Relative Risk for Death with Regard to Total Alcohol Intake and Intake of Beer, Wine, and Spirits Intake of Beer, Wine, and Spirits and Mortality Light to moderate intake of beer or spirits had a small effect on death from all causes (Table 2). This finding contrasted with the effect of wine intake on mortality: Participants who drank 8 to 21 glasses of wine per week had a relative risk for death from all causes of 0.76 (CI, 0.67 to 0.86). Intake of fewer than 22 drinks of beer, wine, and spirits per week all carried lower risk for death from coronary heart disease; the reduction in risk was of the same magnitude for beer and wine drinking but was smaller and not statistically significant for spirits drinking. Furthermore, light to moderate drinkers of wine had

Body mass, fat-free body mass, and prognosis in patients with chronic obstructive pulmonary disease from a random population sample: findings from the Copenhagen City Heart Study.

RATIONALE Low body mass index (BMI) is a marker of poor prognosis in chronic obstructive pulmonary disease (COPD). In the general population, the harmful effect of low BMI is due to the deleterious effects of a low fat-free mass index (FFMI; fat-free mass/weight(2)). OBJECTIVES We explored distribution of low FFMI and its association with prognosis in a population-based cohort of patients with COPD. METHODS We used data on 1,898 patients with COPD identified in a population-based epidemiologic study in Copenhagen. FFM was measured using bioelectrical impedance analysis. Patients were followed up for a mean of 7 yr and the association between BMI and FFMI and mortality was examined taking age, sex, smoking, and lung function into account. MAIN RESULTS The mean FFMI was 16.0 kg/m(2) for women and 18.7 kg/m(2) for men. Among subjects with normal BMI, 26.1% had an FFMI lower than the lowest 10th percentile of the general population. BMI and FFMI were significant predictors of mortality, independent of relevant covariates. Being in the lowest 10th percentile of the general population for FFMI was associated with a hazard ratio of 1.5 (95% confidence interval, 1.2-1.8) for overall mortality and 2.4 (1.4-4.0) for COPD-related mortality. FFMI was also a predictor of overall mortality when analyses were restricted to subjects with normal BMI. CONCLUSIONS FFMI provides information in addition to BMI and assessment of FFM should be considered in the routine assessment of COPD.

Increased Risk of Intestinal Cancer in Crohn's Disease: A Meta-Analysis of Population-Based Cohort Studies

OBJECTIVES:The risk of intestinal malignancy in Crohns disease (CD) remains uncertain since risk estimates vary worldwide. The global CD population is growing and there is a demand for better knowledge of prognosis of this disease. Hence, the aim of the present study was to conduct a meta-analysis of population-based data on intestinal cancer risk in CD.METHODS:The MEDLINE search engine and abstracts from international conferences were searched for the relevant literature by use of explicit search criteria. All papers fulfilling the strict inclusion criteria were scrutinized for data on population size, time of follow-up, and observed to expected cancer rates. STATA meta-analysis software was used to perform overall pooled risk estimates (standardized incidence ratio (SIR), observed/expected) and meta-regression analyses of the influence of specific variables on SIR.RESULTS:Six papers fulfilled the inclusion criteria and reported SIRs of colorectal cancer (CRC) in CD varying from 0.9 to 2.2. The pooled SIR for CRC was significantly increased (SIR, 1.9; 95% CI 1.4–2.5), as was the risk for colon cancer separately (SIR, 2.5; 95% CI 1.7–3.5). Regarding small bowel cancer, five studies reported SIRs ranging from 3.4 to 66.7, and the overall pooled estimate was 27.1 (95% CI 14.9–49.2).CONCLUSIONS:The present meta-analysis of intestinal cancer risk in CD, based on population-based studies only, revealed an overall increased risk of both CRC and small bowel cancer among patients with CD. However, some of the available data were several decades old, and future studies taking new treatment strategies into account are required.

Growth patterns and the risk of breast cancer in women

Adult height and body mass index (BMI) influence the risk of breast cancer in women. Whether these associations reflect growth patterns of the fetus or growth during childhood and adolescence is unknown. We investigated the association between growth during childhood and the risk of breast cancer in a cohort of 117,415 Danish women. Birth weight, age at menarche, and annual measurements of height and weight were obtained from school health records. We used the data to model individual growth curves. Information on vital status, age at first childbirth, parity, and diagnosis of breast cancer was obtained through linkages to national registries. During 3,333,359 person-years of follow-up, 3340 cases of breast cancer were diagnosed. High birth weight, high stature at 14 years of age, low BMI at 14 years of age, and peak growth at an early age were independent risk factors for breast cancer. Height at 8 years of age and the increase in height during puberty (8-14 years of age) were also associated with breast cancer. The attributable risks of birth weight, height at 14 years of age, BMI at 14 years of age, and age at peak growth were 7%, 15%, 15%, and 9%, respectively. No effect of adjusting for age at menarche, age at first childbirth, and parity was observed. Birth weight and growth during childhood and adolescence influence the risk of breast cancer.

Prospective evaluation of alcohol abuse and alcoholic liver injury in men as predictors of development of cirrhosis

258 alcohol-abusing men, free from cirrhosis on primary liver biopsy, were followed for 10-13 years during which cirrhosis developed in 38, corresponding to a rate of 2% per year. The likelihood of cirrhosis developing proved to be independent of duration of abuse and of daily consumption before the primary biopsy. This indicates that the effect of alcohol abuse is not cumulative over time, but rather establishes conditions for the development of cirrhosis. The rate of cirrhosis increased stepwise with degree of steatosis in the primary biopsy, and in those with alcoholic hepatitis was nine times higher than in those with no steatosis. This finding, together with the results on alcohol abuse, indicates that steatosis and alcoholic hepatitis, despite their reversibility, are causally associated with cirrhosis rather than epiphenomena of alcohol abuse. In the alcohol-abusing man, a liver biopsy provides more information than alcoholic history about the likelihood of future cirrhosis.

Concordance of Inflammatory Bowel Disease among Danish Twins: Results of a Nationwide Study

BACKGROUND Previous studies have shown an increased risk of inflammatory bowel disease (IBD) among relatives of patients with Crohn disease and ulcerative colitis. In the present study the probandwise concordance rates for ulcerative colitis and Crohn disease among mono- and dizygotic twins were estimated. Further we aimed to evaluate whether smoking habits might influence the concordance, and to look for clinical characteristics of concordant versus discordant twin pairs. METHODS Among the 38,507 identified twins born in Denmark from 1953 to 1982, a questionnaire was sent to the 34,076 who previously had accepted to participate in studies. For twins reporting IBD, the diagnosis was verified by applying standard criteria to records requested from hospitals or practitioners. RESULTS Among the 29,421 (86.3%) twins answering the questionnaire, 103 pairs had at least one twin who suffered from IBD. In the Crohn disease group five of 10 monozygotic pairs, but none of 27 dizygotic pairs were concordant. In the ulcerative colitis group three of 21 monozygotic, and two of 44 dizygotic pairs were concordant. The probandwise concordance rate among monozygotic pairs was 58.3% for Crohn disease and 18.2% for ulcerative colitis; among the dizygotic pairs the rates were 0 and 4.5%, respectively. The frequency of smokers was higher among twins with Crohn disease and lower among twins with ulcerative colitis compared to the frequency in the twin register. Furthermore, smoking habits were found to be of significance for discordance for disease. Regarding the clinical characteristics no homogenous pattern was observed within the concordant pairs and the differences between concordant and discordant pairs were not significant. CONCLUSION The observation of a significantly higher concordance rate among monozygotic than among dizygotic twin pairs strongly points to a genetic influence on occurrence of IBD, which seems to be more pronounced with regard to Crohn disease than to ulcerative colitis. Differences in smoking habits among the members of the discordant twin pairs may influence the discordance.Background: Previous studies have shown an increased risk of inflammatory bowel disease (IBD) among relatives of patients with Crohn disease and ulcerative colitis. In the present study the probandwise concordance rates for ulcerative colitis and Crohn disease among mono- and dizygotic twins were estimated. Further we aimed to evaluate whether smoking habits might influence the concordance, and to look for clinical characteristics of concordant versus discordant twin pairs. Methods: Among the 38,507 identified twins born in Denmark from 1953 to 1982, a questionnaire was sent to the 34,076 who previously had accepted to participate in studies. For twins reporting IBD, the diagnosis was verified by applying standard criteria to records requested from hospitals or practitioners. Results: Among the 29,421 (86.3%) twins answering the questionnaire, 103 pairs had at least one twin who suffered from IBD. In the Crohn disease group five of 10 monozygotic pairs, but none of 27 dizygotic pairs were concordant. In the ulcerative colitis group three of 21 monozygotic, and two of 44 dizygotic pairs were concordant. The probandwise concordance rate among monozygotic pairs was 58.3% for Crohn disease and 18.2% for ulcerative colitis; among the dizygotic pairs the rates were 0 and 4.5%, respectively. The frequency of smokers was higher among twins with Crohn disease and lower among twins with ulcerative colitis compared to the frequency in the twin register. Furthermore, smoking habits were found to be of significance for discordance for disease. Regarding the clinical characteristics no homogenous pattern was observed within the concordant pairs and the differences between concordant and discordant pairs were not significant. Conclusion: The observation of a significantly higher concordance rate among monozygotic than among dizygotic twin pairs strongly points to a genetic influence on occurrence of IBD, which seems to be more pronounced with regard to Crohn disease than to ulcerative colitis. Differences in smoking habits among the members of the discordant twin pairs may influence the discordance.

Childhood overweight after establishment of the gut microbiota: the role of delivery mode, pre-pregnancy weight and early administration of antibiotics

Objective:To investigate whether delivery mode (vaginal versus by caesarean section), maternal pre-pregnancy body mass index (BMI) and early exposure to antibiotics (<6 months of age) influence childs risk of overweight at age 7 years, hence supporting the hypotheses that environmental factors influencing the establishment and diversity of the gut microbiota are associated with later risk of overweight.Design:Longitudinal, prospective study with measure of exposures in infancy and follow-up at age 7 years.Methods:A total of 28 354 mother–child dyads from the Danish National Birth Cohort, with information on maternal pre-pregnancy BMI, delivery mode and antibiotic administration in infancy, were assessed. Logistic regression analyses were performed with childhood height and weight at the 7-year follow-up as outcome measures.Results:Delivery mode was not significantly associated with childhood overweight (odds ratio (OR):1.18, 95% confidence interval (CI): 0.95–1.47). Antibiotics during the first 6 months of life led to increased risk of overweight among children of normal weight mothers (OR: 1.54, 95% CI: 1.09–2.17) and a decreased risk of overweight among children of overweight mothers (OR: 0.54, 95% CI: 0.30–0.98). The same tendency was observed among children of obese mothers (OR: 0.85, 95% CI: 0.41–1.76).Conclusion:The present cohort study revealed that a combination of early exposures, including delivery mode, maternal pre-pregnancy BMI and antibiotics in infancy, influences the risk of overweight in later childhood. This effect may potentially be explained by an impact on establishment and diversity of the microbiota.