Registration of Diagnoses in the Danish National Registry of Patients

Examining selection bias in a population-based cohort study of 522 children with familial high risk of schizophrenia or bipolar disorder, and controls: The Danish High Risk and Resilience Study VIA 7

PURPOSE

Knowledge about representativity of familial high-risk studies of schizophrenia and bipolar disorder is essential to generalize study conclusions. The Danish High Risk and Resilience Study (VIA 7), a population-based case-control familial high-risk study, creates a unique opportunity for combining assessment and register data to examine cohort representativity.

METHODS

Through national registers, we identified the population of 11,959 children of parents with schizophrenia (FHR-SZ) or bipolar disorder (FHR-BP) and controls from which the 522 children participating in The VIA 7 Study (202 FHR-SZ, 120 FHR-BP and 200 controls) were selected. Socio-economic and health data were obtained to compare high-risk groups and controls, and participants versus non-participants. Selection bias impact on results was analyzed through inverse probability weights.

RESULTS

In the total sample of 11,959 children, FHR-SZ and FHR-BP children had more socio-economic and health disadvantages than controls (p < 0.001 for most). VIA 7 non-participants had a poorer function, e.g. more paternal somatic and mental illness (p = 0.02 and p = 0.04 for FHR-SZ), notifications of concern (FHR-BP and PBC p < 0.001), placements out of home (p = 0.03 for FHR-SZ), and lower level of education (p ≤ 0.01 for maternal FHR-SZ and FHR-BP, p = 0.001 for paternal FHR-BP). Inverse probability weighted analyses of results generated from the VIA Study showed minor changes in study findings after adjustment for the found selection bias.

CONCLUSIONS

Familial high-risk families have multiple socio-economic and health disadvantages. In The VIA 7 Study, although comparable regarding mental illness severity after their child's birth, socioeconomic and health disadvantages are more profound amongst non-participants than amongst participants.

Validity of Algorithm for Classification of In- and Outpatient Hospital Contacts in the Danish National Patient Registry

Purpose

Following the implementation of the 3rd version of the Danish National Patient Register (DNPR-3), information on whether hospitalizations were inpatient, outpatient, or emergency room (ER) contacts was no longer readily available. This study examined the positive predictive values (PPV) of a common algorithm to characterize hospitalizations as inpatient, outpatient, or emergency room (ER) contacts in both DNPR-2 and DNPR-3.

Patients and Methods

All hospital contacts in North Denmark Region were identified in the DNPR within a 1-year window of the implementation of DNPR-3 in early 2019. An algorithm based upon proportion of overnight (±50%) and elective (±50%) contacts for each hospital department was developed. Next, PPVs of these categorizations were computed using manual characterization of all departments and clinics by two experienced clinicians as reference. Second, the reliability of various time intervals to join department contacts and subsequent categorization of overnight hospital stays as proxies for inpatient contacts was explored.

Results

The algorithm yielded PPVs of 91% and 89% for hospital units and related contacts categorized as inpatient in DNPR-2 and 100% for both parameters in DNPR-3. In outpatient units, the PPVs were 99% in both DNPR-2 and DNPR-3, whereas the corresponding PPVs were 99.6% and 99% on the contact level. In contrast, the PPV for ERs was 33% in DNPR-2 and 56% in DNPR-3, primarily due to misclassification of outpatient clinics. Still, the proportion of correctly categorized ER contacts was 87% in DNPR-2 and 85% in DNPR-3. Using time intervals from 0 to 12 hours to join department contacts showed that overnight hospitalizations comprised inpatient contacts in 97% in DNPR-2 and 98% in DNPR-3. However, the sensitivity was moderate at 76-78% for all inpatient hospitalizations in DNPR-2 and DNPR-3.

Conclusion

This algorithm accurately categorized hospitalizations as inpatient, outpatient, or ER contacts in both DNPR-2 and DNPR-3.

Opioid use among Danish patients with severe knee osteoarthritis: a drug utilization study

PURPOSE

Knee osteoarthritis (OA) is one of the most common musculoskeletal diseases. Opioids have been increasingly used in the treatment of severe knee OA-related pain, particularly in the USA. Less is known about the patterns of use of opioids among Danish patients with severe knee OA. We investigated opioid use among Danish patients with severe knee OA in the 5 years preceding knee replacement surgery (KR).

METHODS

We identified adults who had undergone KR from January 1st, 2005, to December 31st, 2018, using the Danish National Patient Register. These patients were considered to have severe knee OA in the 5-year period leading up to KR. Individual-level data on prescribed opioids were retrieved from the Danish National Prescription Registry.

RESULTS

We identified 77,168 severe knee OA patients (mean age 66 years). The prevalence of opioid users increased from 21% 5 years before KR to 40% 1 year before. Total use of opioids increased each year and doubled from 3254 mg oral morphine equivalents (OMEQ)/1000 individuals/day 5 years before to 6396 mg OMEQ/1000 individuals/day the year before KR corresponding to an increase of 3141 mg OMEQ (95% confidence interval 3010 to 3273). Tramadol was the most frequently used opioid. About 10% of the population accounted for 90% of the total opioid use.

CONCLUSION

Among patients with severe knee OA, the prevalence and total use of opioids doubled during the 5 years before KR. In addition, 10% of the study population was responsible for 90% of the opioids used.

The association between life satisfaction, vitality, self-rated health, and risk of cancer

PURPOSE

Only few prospective studies have been conducted on the contribution of quality of life-related factors to the risk of cancer. The aim of this study was to investigate the prospective associations of three quality of life-related factors with the risk of cancer; life satisfaction, vitality, and self-rated health.

METHODS

In 2009-2011, 7189 participants in the Copenhagen Aging and Midlife Biobank were asked to rate their life satisfaction, their vitality, and their health. The study population was followed until the end of 2015 for registration of cancer in the Danish National Patient Register.

RESULTS

During the follow-up period, cancer was diagnosed in 312 individuals. Life satisfaction was not associated with the risk of cancer. Vitality was significantly associated with the risk of cancer, but the association became non-significant after adjustment for age, sex, socioeconomic position, and lifestyle factors. However, when additionally adjusting for life satisfaction, individuals who rated their vitality as low had a hazard ratio of 1.46 (95% confidence interval [CI] 1.04-2.07) for the development of cancer. Individuals who rated their health as poor had a hazard ratio of 1.70 (95% CI 1.27-2.26) for the development of cancer, compared with individuals with good, very good, or excellent self-rated health. The association remained significant after adjustment for basic confounders, life satisfaction, and vitality.

CONCLUSION

A better grasp of the significance of quality of life-related factors for the risk of cancer may be of great importance to population-based cancer prevention that aims to target early risk factors for development of cancer across widespread cancer sites.

The Birth and Evolution of a Discipline Devoted to Information in Biomedicine and Health Care

Background: The journal Methods of Information in Medicine, founded in 1962, has now completed its 50th volume. Its publications during the last five decades reflect the formation of a discipline that deals with information in biomedicine and health care. Objectives: To report about 1) the journal‘s origin, 2) the individuals who have significantly contributed to it, 3) trends in the journal’s aims and scope, 4) influential papers and 5) major topics published in Methods over the years.

Methods: Methods included analysing the correspondence and journal issues in the archives of the editorial office and of the publisher, citation analysis using the ISI and Scopus databases, and analysing the articles’ Medical Subject Headings (MeSH) in MEDLINE.

Results: In the journal’s first 50 years 208 editorial board members and/or editors contributed to the journal’s development, with most individuals coming from Europe and North America. The median time of service was 11 years. At the time of analysis 2,456 articles had been indexed with Me SH. Topics included computerized systems of various types, informatics methodologies, and topics related to a specific medical domain. Some MeSH topic entries were heavily and regularly represented in each of the journal‘s five decades (e.g. information systems and medical records), while others were important in a particular decade, but not in other decades (e.g. punched-card systems and systems integration). Seven papers were cited more than 100 times and these also covered a broad range of themes such as knowledge representation, analysis of biomedical data and knowledge, clinical decision support and electronic patient records. Conclusions: Methods of Information in Medicine is the oldest international journal in biomedical informatics. The journal’s development over the last 50 years correlates with the formation of this new discipline. It has and continues to stress the basic methodology and scientific fundamentals of organizing, representing and analysing data, information and knowledge in biomedicine and health care. It has and continues to stimulate multi-disciplinary communication on research that is devoted to high-quality, efficient health care, to quality of life and to the progress of biomedicine and the health sciences.

Does vital exhaustion increase the risk of type 2 diabetes? A prospective study

BACKGROUND

There is evidence that both stress and depression have a causal relationship with type 2 diabetes suggesting that vital exhaustion (VE) too could be a risk factor. The association between VE and type 2 diabetes has, however, not been investigated prospectively.

AIM

To prospectively investigate whether VE is associated with an increased risk of type 2 diabetes in a Danish population.

METHODS

A prospective cohort study based on the Copenhagen City Heart Study (1991-1993). The degree of VE was measured among 9075 participants without type 1 or 2 diabetes at baseline. To detect type 2 diabetes in the follow-up period, two different approaches were used: In the first substudy, type 2 diabetes was defined based on blood samples and questionnaires from a follow-up study in 2001-2003 (N=4708). The second substudy was register-based, and the study population was linked to the Danish Hospital Discharge Register to detect registrations with type 2 diabetes until 2014.

RESULTS

A high degree of VE was associated with an increased risk of developing type 2 diabetes in both substudies. In the first substudy, the OR for developing type 2 diabetes was 2.56 (95% CI, 1.53; 4,29, P<0,001) among the quartile of participants reporting the highest degree of VE. In the second substudy, the OR was 1.31 (95% CI, 0.99; 1.72, P=0.053) for this group.

CONCLUSION

The results indicate that VE may be a useful measure in clinical practice in order to discover individuals at risk of type 2 diabetes.

Self-rated health and cancer risk - a prospective cohort study among Danish women

BACKGROUND

Self-rated health (SRH) has been shown to be a strong predictor of mortality from a number of major chronic diseases, however, the association with cancer remains unclear. The aim of this study was to investigate a possible association between change in SRH and cancer incidence.

MATERIALS AND METHODS

SRH and information on lifestyle and other risk factors were obtained for 13-636 women in the Danish Nurse Cohort. Cancers that developed during 12 years of follow-up were identified in the National Patient Registry. An association between SRH and cancer was examined in a Cox proportional hazards model with adjustment for age, smoking, alcohol, marital status, physical activity, body mass index and estrogen replacement therapy.

RESULTS

No significant association was found between SRH and overall cancer incidence in the age-adjusted Cox proportional hazards model (1.04; 95% CI 0.93-1.16), even after adjustment for potential confounding factors (HR 1.08; 95% CI 0.96-1.21). Likewise, there was no significant association between SRH and breast cancer (HR 1.09; 95% CI 0.89-1.33), lung cancer (HR 1.03; 95% CI 0.71-1.49) or colon cancer (HR 1.08; 95% CI 0.75-1.54).

CONCLUSION

SRH is not significantly associated with the incidence of all cancers or breast, lung or colon cancer among Danish female nurses. Women who reported a decrease in SRH between 1993 and 1999 had the same risk for cancer as those who reported unchanged or improved SRH.

Trends in One-Year Outcomes of Dialysis-Requiring Acute Kidney Injury in Denmark 2005-2012: A Population-Based Nationwide Study

BACKGROUND

Dialysis-requiring acute kidney injury (AKI) is associated with substantial mortality and risk of end-stage renal disease (ESRD). Despite considerable growth in incidence of severe AKI, information pertaining to trends in outcomes remains limited. We evaluated time trends in one year risks of ESRD and death in patients with dialysis-requiring AKI over an eight year period in Denmark.

METHODS

In a retrospective nationwide study based on national registers, all adults requiring acute renal replacement therapy between 2005 and 2012 were identified. Patients with preceding ESRD were excluded. Through individual-level cross-referencing of administrative registries, information pertaining to comorbidity, preceding surgical interventions, and concurrent other organ failure and sepsis was ascertained. Comparisons of period-specific one year odds ratios for ESRD and death were calculated in a multiple logistic regression model.

RESULTS

A total of 13,819 patients with dialysis-requiring AKI were included in the study. Within one year, 1,017 (7.4%) patients were registered with ESRD, and 7,908 (57.2%) patients died. The one-year rate of ESRD decreased from 9.0% between 2005 and 2006 to 6.1% between 2011 and 2012. Simultaneously, the one-year mortality rate decreased from 58.2% between 2005 and 2006 to 57.5% between 2011 and 2012. Consequently, the adjusted odds ratios for the period 2011-2012 (with the period 2005-2006 as reference) were 0.75 (0.60-0.95, p = 0.015) and 0.87 (95% CI 0.78-0.97, p = 0.010) for ESRD and death, respectively.

CONCLUSIONS

In a nationwide retrospective study on time trends in one year outcomes following dialysis-requiring AKI, risk of all-cause mortality and ESRD decreased over a period of 8 years.

Dialysis-Requiring Acute Kidney Injury in Denmark 2000-2012: Time Trends of Incidence and Prevalence of Risk Factors-A Nationwide Study

Introduction

Dialysis-requiring acute kidney injury is a severe illness associated with poor prognosis. However, information pertaining to incidence rates and prevalence of risk factors remains limited in spite of increasing focus. We evaluate time trends of incidence rates and changing patterns in prevalence of comorbidities, concurrent medication, and other risk factors in nationwide retrospective cohort study.

Materials and Methods

All patients with dialysis-requiring acute kidney injury were identified between January 1^st 2000 and December 31^st 2012. By cross-referencing data from national administrative registries, the association of changing patterns in dialysis treatment, comorbidity, concurrent medication and demographics with incidence of dialysis-requiring acute kidney injury was evaluated.

Results

A total of 18,561 adult patients with dialysis-requiring AKI were identified between 2000 and 2012. Crude incidence rate of dialysis-requiring AKI increased from 143 per million (95% confidence interval, 137–144) in 2000 to 366 per million (357–375) in 2006, and remained stable hereafter. Notably, incidence of continuous veno-venous hemodialysis (CRRT) and use of acute renal replacement therapy in elderly >75 years increased substantially from 23 per million (20–26) and 328 per million (300–355) in 2000, to 213 per million (206–220) and 1124 per million (1076–1172) in 2012, respectively. Simultaneously, patient characteristics and demographics shifted towards increased age and comorbidity.

Conclusions

Although growth in crude incidence rate of dialysis-requiring AKI stabilized in 2006, continuous growth in use of CRRT, and acute renal replacement therapy of elderly patients >75 years, was observed. Our results indicate an underlying shift in clinical paradigm, as opposed to unadulterated growth in incidence of dialysis-requiring AKI.

Social network as predictor for onset of alcohol use disorder: A prospective cohort study

OBJECTIVE

Social network has been linked to alcohol use disorder in several studies. However, since the majority of such findings are cross-sectional, causal interpretation is difficult. The aim of the present study was to test if social network characteristics predict alcohol use disorder in a prospective design.

METHODS

Information on social network and covariates was obtained from 9589 men and women aged 21-99 years in the Copenhagen City Heart Study, followed for registration of alcohol use disorder in the Danish National Patient Registry and the WINALCO database.

RESULTS

Men who lived alone, were separated or divorced or widowers had a higher risk of developing alcohol use disorder: HR among men living alone vs. men not living alone was 2.28 (95% CI: 1.59-3.27), and HR among separated/divorced men vs. married men was 2.55 (95% CI: 1.33-4.89). No such associations were found among women. Frequency of contact with friends was associated with risk of developing alcohol use disorder among both sexes. For example, the HRs were 1.72 (CI 95%: 0.99-3.01) and 2.59 (95% CI: 1.42-4.71) among women who had contact with friends a couple of times per week and daily, respectively, compared with more rarely. Frequency of contact with family was not associated with risk of developing alcohol use disorder among either sex.

CONCLUSION

Living alone and not being married or cohabiting with a partner were predictors of developing alcohol use disorder among men. Further, frequent contact with friends was associated with higher risk of alcohol use disorder among both sexes.

Cirrhosis is Associated with an Increased 30-Day Mortality After Venous Thromboembolism

Objectives:

Patients with cirrhosis are at increased risk of venous thromboembolism (VTE), but the impact of cirrhosis on the clinical course following VTE is unclear. In a nationwide cohort study, we examined 30-day mortality among patients with cirrhosis and VTE.

Methods:

We used Danish population-based health-care databases (1994–2011) to identify patients with incident VTE, i.e., deep venous thrombosis (DVT), pulmonary embolism (PE), and portal vein thrombosis (PVT). Among these, we identified 745 patients with cirrhosis and 3647 patients without cirrhosis (matched on gender, year of birth, calendar year of VTE diagnosis and VTE type). We assessed the 30-day mortality risk among VTE patients with and without cirrhosis, and the mortality rate ratios (MRRs), using an adjusted Cox model with 95% confidence interval. We obtained information on immediate cause of death for patients who died within 30 days after VTE.

Results:

The 30-day mortality risk for DVT was 7% for patients with cirrhosis and 3% for patients without cirrhosis. Corresponding PE-related mortality risks were 35% and 16%, and PVT-related mortality risks were 19% and 15%, respectively. The adjusted 30-day MRRs were 2.17 (1.24–3.79) for DVT, 1.83 (1.30–2.56) for PE, and 1.30 (0.80–2.13) for PVT. Though overall mortality was higher in patients with cirrhosis than patients without cirrhosis, the proportions of deaths due to PE were similar among patients (25% and 24%, respectively).

Conclusions:

Cirrhosis is a predictor for increased short-term mortality following VTE, with PE as the most frequent cause of death.

A prospective cohort study of quality of life and ischemic heart disease

AIMS

To examine the role of self-rated quality of life, as a whole, on the risk of developing ischemic heart disease.

METHODS

Information on self-rated quality of life and covariates was obtained from 4497 people from the Copenhagen Perinatal Cohort. The study population, consisting of both men and women aged 31-33 years at baseline, was followed for registration of ischemic heart disease in national registers. Two different follow-up periods were examined.

RESULTS

Using a follow-up of 13 years, the basis-adjusted hazard ratio among women who rated their quality of life as poor was 2.93 (95% CI: 1.09-7.86), compared with women who rated their quality of life as very good. The hazard ratio of women who rated their quality of life as good, compared to very good, was 2.99 (95% CI: 1.22-7.29). When analyzing 16 years of follow-up, estimates among women were lower and some of the hazard ratios were insignificant. All hazard ratios among men were below two and none were statistically significant.

CONCLUSIONS

Among women, the risk for ischemic heart disease was two to four times greater for those who rated their quality of life as poor or good, than for women with the most positive rating. For men, the association was not statistically significant.

Smoking and risk of liver cirrhosis: a population-based cohort study

OBJECTIVE

Alcohol is the most acknowledged risk factor for liver cirrhosis. Smoking is rarely considered to be a cause of liver cirrhosis even though a few studies have suggested the opposite. The aim of this study was to assess the independent effect of smoking on alcoholic liver cirrhosis and liver cirrhosis in general.

MATERIALS AND METHODS

The authors used data from the Copenhagen City Heart Study including a cohort of 9889 women and 8590 men from the Danish general population. Smoking and covariates were assessed at four consecutive examinations in 1976-78, 1981-83, 1991-94 and 2001-03. Updated measures were used in the analyses. Information on incident cases of alcoholic liver cirrhosis and liver cirrhosis was obtained from national hospital registries. Data were analyzed by means of Cox regression.

RESULTS

A total of 225 cases of alcoholic liver cirrhosis and 431 cases of liver cirrhosis occurred during follow-up. The hazard ratios (HR) of alcoholic liver cirrhosis was 3.9, 95% confidence interval (CI): 1.6, 9.4 for women and 1.6, 95% CI: 0.9, 3.0 for men smoking >10 g of tobacco per day compared with never-smokers. For liver cirrhosis, corresponding HRs were 2.2, 95% CI: 1.4, 3.4 for women and 1.4, 95% CI: 0.9, 2.2 for men. The HRs were adjusted for age, alcohol intake, education and body mass index.

CONCLUSIONS

Smoking was associated with an increased risk of liver cirrhosis independent of alcohol intake.

Social and psychological predictors of onset of anxiety disorders: results from a large prospective cohort study

PURPOSE

The vast majority of studies investigating the association between social and psychological factors and anxiety disorders have been cross-sectional, making it difficult to draw causal conclusions. The purpose of the study was to investigate in a prospective longitudinal study whether social and psychological factors are associated with the later risk of being admitted to a hospital and receive a diagnosis of anxiety disorders.

METHOD

The study population comprised 4,497 members of The Copenhagen Perinatal Cohort (CPC) who in 1993 answered a mailed questionnaire containing questions on a range of social and psychological factors. In 2007, the study population was linked to The Danish Hospital Discharge Register and the Danish Psychiatric Central Register to obtain information on registration with anxiety disorders. Multiple Cox regression analysis was used to analyze the risk of anxiety disorders according to social and psychological factors.

RESULTS

A total of 5.3% of the study population had lifetime registration with an anxiety disorder diagnosis. The risk of admission for anxiety disorders was significantly associated with previous: discontentedness with partner-status, loneliness, self-rated low intelligence, not feeling part of a whole, unhappiness, low quality of life, and low meaningfulness. Estimates were adjusted for income and current diseases.

CONCLUSION

The present study demonstrated that in a population without previous registration with anxiety disorders, contentment with social relations and a range of beneficial psychological factors reduced the later risk of being hospitalized with anxiety disorders.

A prospective association between quality of life and risk for cancer

AIM

The contributions of social and especially of psychological factors to cancer development have been questioned. The goal of this study was to investigate, in a longitudinal setting, the prospective associations between self-reported measures of social relations, subjective health (physical and mental) and quality of life and the risk for cancer.

METHODS

In 1993, 4493 people in the Copenhagen Perinatal Cohort were asked to rate their social relations, their physical and mental health and their quality of life. The study population was followed until the end of 2006 for registration of cancer in the Danish Hospital Discharge Register.

RESULTS

During the follow-up period, cancer was diagnosed in 102 people. When compared with people with very good quality of life, those who rated their quality of life as poor had a hazard ratio of 1.90 (95% confidence interval [CI], 1.1-3.4) for cancer, and those who rated their quality of life as good had a hazard ratio of 1.31 (95% CI, 0.8-2.2), after adjustment for age, sex, income, lifestyle factors and number of health problems. Self-rated physical and mental health were significantly associated with the risk for cancer, but the estimates became non-significant after adjustment for confounding factors. Social relations were not associated with the risk for cancer.

CONCLUSION

Broad assessment of general well-being, self-rated as global quality of life, appears to be a better predictor of cancer risk than more specific information on social relations and health.

Binge drinking, drinking frequency, and risk of ischaemic heart disease: a population-based cohort study

INTRODUCTION

Light-to-moderate alcohol drinking is associated with a decreased risk of ischaemic heart disease (IHD). However, drinking heavily and in binges has been suggested to increase IHD risk. This complexity makes the issue of binge drinking within the light-to-moderate alcohol range an important area for further investigation.

METHODS

This population-based cohort study included 26,786 men and women who participated in the Danish National Cohort Study in 1994, 2000, and 2005. Binge drinking (defined >5 drinks/day) and risk of IHD and all-cause mortality was investigated among light-to-moderate drinkers (defined ≤21 and ≤14 drinks/week for men and women, respectively). In the entire study population, we investigated the association between drinking frequency, separately and combined with total weekly alcohol intake, and risk of IHD and all-cause mortality.

RESULTS

1136 individuals developed IHD during a mean follow up of 6.9 years. Among male light-to-moderate drinkers reporting occasional binge drinking, the hazard ratio (HR) of IHD was 0.81 (95% CI 0.61-1.08) compared to male light-to-moderate drinkers reporting no binge drinking. Corresponding HR for women was 0.97 (95% CI 0.54-1.76). For women drinking 5-6 days/week, the risk of IHD was lower than for women drinking 1-2 days/week (HR 0.54, 95% CI 0.32-0.90). We did not observe any patterns when looking at combinations of total weekly alcohol intake and drinking frequency.

CONCLUSIONS

Among light-to-moderate alcohol drinkers, binge drinking was not associated with risk of IHD and all-cause mortality. Overall, drinking frequency did not appear to be an important determinant of the risk of IHD and all-cause mortality.

Alcohol consumption and later risk of hospitalization with psychiatric disorders: prospective cohort study

The potential effects of alcohol intake upon the risk of psychiatric disorders have not often been investigated. The purpose of this study was to investigate, in a population sample, the association between self-reported amount of alcohol intake and the later risk of being registered in a Danish hospital with a psychiatric disorder. The prospective cohort study, the Copenhagen City Heart Study (n=18,146), was used, containing three updated sets of alcohol intake and lifestyle covariates and up to 26 years follow-up. Alcohol intake was measured by self-report while psychiatric disorders were measured through registers. For women, the overall pattern showed that drinking above the sensible limits increased the risk of psychiatric disorders in general, especially for anxiety disorders where women drinking above the sensible drinking limits had a risk of 2.00 (confidence interval: 1.31-3.04) compared to women drinking below the sensible drinking limits. For men, the risk functions were slightly U-shaped; thus, a weekly low or moderate alcohol intake seemed to have a protective effect towards developing psychiatric disorders. The findings suggest sex differences in the association between alcohol consumption and risk of psychiatric disorders.

Tobacco smoking as a risk factor for depression. A 26-year population-based follow-up study

BACKGROUND

A key question regarding the documented association between smoking and depression is whether it reflects a causal influence of smoking on depression; however, only a limited number of longitudinal studies exist in the literature, all of which have relatively short time frames. The purpose was to prospectively assess the risk of depression according to daily tobacco consumption in a Danish longitudinal study.

METHODS

A prospective cohort study, the Copenhagen City Heart Study (n = 18,146), was analyzed with up to 26 years of follow-up. It contains three updated data sets on tobacco consumption and potential confounding factors. The study population was linked to Danish hospital registers to detect registrations with depression. Individuals with depression before baseline were excluded.

RESULTS

Women smoking more than 10 g of tobacco per day were at significantly increased risk of depression compared to women who did not smoke. The adjusted risk of depression among women smoking 11-20 g per day was 1.74 (CI:1.33-2.27) and 2.17 (CI:1.45-3.26) among women smoking more than 20 g per day. For men, there was an increased risk of depression for those smoking more than 20 g per day (HR = 1.90; CI:1.05-3.44). All tests for linear trend were significant (all p < 0.01). The estimates remained significant in sensitivity analyses aiming to eliminate reverse causation, and in analyses based on a reduced sample without individuals with chronic diseases or psychiatric disorders other than depression.

CONCLUSION

The study suggests that smoking is associated with increased risk of developing depression. This underlines the potentially harmful consequences of smoking for mental health and supports efforts to prevent and stop smoking.

Alcohol Intake, Myocardial Infarction, Biochemical Risk Factors, and Alcohol Dehydrogenase Genotypes

Background— The risk of myocardial infarction is lower among light-to-moderate alcohol drinkers compared with abstainers. We tested associations between alcohol intake and risk of myocardial infarction and risk factors and whether these associations are modified by variations in alcohol dehydrogenases.

Methods and Results— We used information on 9584 men and women from the Danish general population in the Copenhagen City Heart Study. During follow-up, from 1991 to 2007, 663 incident cases of myocardial infarction occurred. We observed that increasing alcohol intake was associated with decreasing risk of myocardial infarction, decreasing low-density lipoprotein cholesterol and fibrinogen, increasing diastolic and systolic blood pressure and high-density lipoprotein cholesterol, and with U-shaped nonfasting triglycerides. In contrast, ADH1B and ADH1C genotypes were not associated with risk of myocardial infarction or with any of the cardiovascular biochemical risk factors, and there was no indication that associations between alcohol intake and myocardial infarction and between alcohol intake and risk factors were modified by genotypes.

Conclusions— Increasing alcohol intake is associated with decreasing risk of myocardial infarction, decreasing low-density lipoprotein cholesterol and fibrinogen, increasing diastolic and systolic blood pressure and high-density lipoprotein cholesterol, and U-shaped nonfasting triglycerides. These associations were not modified by ADH1B and ADH1C are genotypes.

Alcohol intake, alcohol dehydrogenase genotypes, and liver damage and disease in the Danish general population

OBJECTIVES

We tested the hypothesis that alcohol, alone and in combination with alcohol dehydrogenase (ADH) 1B and ADH1C genotypes, affects liver damage and disease in the general population.

METHODS

Information on alcohol intake and on liver disease was obtained from 9,080 men and women from the Copenhagen City Heart Study. Biochemical tests for the detection of liver damage were specific for alanine aminotransferase (ALT), aspartate aminotransferase (AST)-to-ALT ratio (AST/ALT), gamma-glutamyl transpeptidase (gamma-GT), albumin, bilirubin, alkaline phosphatase, coagulation factors, and erythrocyte volume.

RESULTS

Increasing alcohol intake was associated with increasing erythrocyte volume, AST/ALT, and levels of ALT, gamma-GT, albumin, bilirubin, coagulation factors, and with decreasing levels of alkaline phosphatase. Multifactorially adjusted hazard ratios for alcoholic liver disease overall were 0.9 (95% confidence interval (CI), 0.6-1.4), 1.4 (0.8-2.5), 1.8 (0.9-3.5), and 4.1 (2.5-7.0) for an alcohol intake of 1-13, 14-20, 21-27, and > or = 28 drinks per week, respectively, compared with drinking <1 drink per week (P for trend<0.0001); the corresponding hazard ratios for alcoholic liver cirrhosis were 1.7 (0.6-4.7), 2.0 (0.8-7.1), 6.5 (2.0-21), and 13 (4.6-37) (P for trend<0.0001). ADH1B and ADH1C genotypes were not associated with and did not modify the effect of alcohol on biochemical tests or risk of liver disease.

CONCLUSIONS

Increasing alcohol intake from none to low (1-6 drinks per week) through to moderate (7-20 drinks per week) and excessive intake (> or = 21 drinks per week) leads to stepwise increases in signs of liver damage with no threshold effect, and to an increased risk of liver disease. The minor changes in biochemical tests for low alcohol intake may not account for subclinical liver disease.

Comorbidity and temporal ordering of alcohol use disorders and other psychiatric disorders: results from a Danish register-based study

BACKGROUND

Understanding the comorbidity of alcohol use disorders (AUD) and other psychiatric disorders may have important implications for treatment and preventive interventions. However, information on the epidemiology of this comorbidity is lacking. The objective of this study was to present results on lifetime psychiatric comorbidity of AUD in a large Danish community population.

METHODS

A prospective cohort study was used, comprising 3 updated measures of sets of lifestyle covariates and 26 years of follow-up data on 18,146 individuals from the Copenhagen City Heart Study, Denmark. The study population was linked to national Danish hospital registers and a greater Copenhagen alcohol unit treatment register to detect registrations with AUD and other psychiatric disorders.

RESULTS

Of the individuals invited to the study, 7.6% were registered with AUD, and among these, 50.3% had a lifetime comorbid psychiatric disorder. Personality disorders were the most common comorbid disorders (24%) together with mood disorders (16.8%) and drug abuse (16.6%). The risk of developing a psychiatric disorder in individuals who were already registered with AUD was larger than the risk of developing AUD in individuals who were already registered with another psychiatric disorder; these differences in risk were especially noticeable for anxiety disorders, personality disorders, and drug abuse.

CONCLUSIONS

AUD is frequently comorbid with other psychiatric disorders, and it is likely that AUD is both an etiologic factor in other mental disorders and a consequence of mental disease. However, in interpreting these complex and perhaps circular causal links, it is important to consider that AUD is registered before a comorbid psychiatric diagnosis more often than the reverse temporal order.

Alcohol use disorders increase the risk of completed suicide--irrespective of other psychiatric disorders. A longitudinal cohort study

Knowledge of the epidemiology of suicide is a necessary prerequisite for developing prevention programs. The aim of this study was to analyze the risk of completed suicide among individuals with alcohol use disorders (AUD), and to assess the role of other psychiatric disorders in this association. A prospective cohort study was used, containing three updated sets of lifestyle covariates and 26 years follow-up of 18,146 individuals between 20 and 93 years of age from the Copenhagen City Heart Study in Denmark. The study population was linked to four different registers in order to detect: Completed suicide, AUD, Psychotic disorders, Anxiety disorders, Mood disorders, Personality disorders, Drug abuse, and Other psychiatric disorders. Individuals registered with AUD were at significantly increased risk of committing suicide, with a crude hazard ratio (HR) of 7.98 [Confidence interval (CI): 5.27-12.07] compared to individuals without AUD. Adjusting for all psychiatric disorders the risk fell to 3.23 (CI: 1.96-5.33). In the stratified sub-sample of individuals without psychiatric disorders, the risk of completed suicide was 9.69 (CI: 4.88-19.25) among individuals with AUD. The results indicate that individuals registered with AUD are at highly increased risk of completed suicide, and that registered co-morbid psychiatric disorders are neither sufficient nor necessary causes in this association.

Increased rate of treatment with antidepressants in patients with multiple sclerosis

The prevalence of depression and anxiety is increased in patients with multiple sclerosis, but it has not been investigated whether these conditions are treated in clinical practice. The objective of this study was to investigate whether the rate of treatment with antidepressants is increased in patients with multiple sclerosis compared with patients with other chronic illnesses and compared with the general population. By linkage of nationwide case registers, all patients were identified, who had received a main diagnosis of multiple sclerosis or osteoarthritis at first admission or during outpatient contact in the period 1995-2000 in Denmark. Rates of subsequent purchase of antidepressants for these patients were calculated. In total, 417 patients with a main diagnosis of multiple sclerosis and 12 127 patients with a main diagnosis of osteoarthritis, at first discharge from hospital or outpatient contact, were included. Patients with a diagnosis of multiple sclerosis had a 3.21 [95% confidence intervals (CI): 2.56-4.03] times increased rate of subsequently purchasing antidepressants compared with patients with a first diagnosis of osteoarthritis, and a 4.75 times (95% CI: 3.91-5.76) increased rate when compared with the rate among a gender-matched, age-matched, and calendar-matched sample of the general population. The rates were increased in all subgroups of patients regardless of gender, age, socioeconomic group, and time elapsed since diagnosis.

Validity of dementia diagnoses in the Danish hospital registers

BACKGROUND

The validity of dementia diagnoses in the Danish nationwide hospital registers was evaluated to determine the value of these registers in epidemiological research about dementia.

METHODS

Two hundred patients were randomly selected from 4,682 patients registered for the first time with a dementia diagnosis in the last 6 months of 2003. The patients' medical journals were reviewed to evaluate if they fulfilled ICD-10 and/or DSM-IV criteria for dementia and specific dementia subtypes. The patients who were still alive in 2006 were invited to an interview.

RESULTS

One hundred and ninety-seven journals were available for review and 51 patients were interviewed. A registered diagnosis of dementia was found to be correct in 169 (85.8%) cases. Regarding dementia subtypes, the degree of agreement between the registers and the results of the validating process was low with a kappa of 0.36 (95% CI 0.24-0.48).

CONCLUSION

The validity of dementia syndrome in the Danish hospital registers was high and allows for epidemiological studies about dementia. Alzheimer's disease, although underregistered, also had a good validity once the diagnosis was registered. In general, other ICD-10 dementia subtypes in the registers had a low validity and are less suitable for epidemiological research.

Alcoholism and alcohol drinking habits predicted from alcohol dehydrogenase genes

Alcohol drinking habits and alcoholism are partly genetically determined. Alcohol is degraded primarily by alcohol dehydrogenase (ADH) wherein genetic variation that affects the rate of alcohol degradation is found in ADH1B and ADH1C. It is biologically plausible that these variations may be associated with alcohol drinking habits and alcoholism. By genotyping 9080 white men and women from the general population, we found that men and women with ADH1B slow vs fast alcohol degradation drank more alcohol and had a higher risk of everyday drinking, heavy drinking, excessive drinking and of alcoholism. For example, the weekly alcohol intake was 9.8 drinks (95% confidence interval (CI): 9.1-11) among men with the ADH1B.1/1 genotype compared to 7.5 drinks (95% CI: 6.4-8.7) among men with the ADH1B.1/2 genotype, and the odds ratio (OR) for heavy drinking was 3.1 (95% CI: 1.7-5.7) among men with the ADH1B.1/1 genotype compared to men with the ADH1B.1/2 genotype. Furthermore, individuals with ADH1C slow vs fast alcohol degradation had a higher risk of heavy and excessive drinking. For example, the OR for heavy drinking was 1.4 (95% CI: 1.1-1.8) among men with the ADH1C.1/2 genotype and 1.4 (95% CI: 1.0-1.9) among men with the ADH1B.2/2 genotype, compared with men with the ADH1C.1/1 genotype. Results for ADH1B and ADH1C genotypes among men and women were similar. Finally, because slow ADH1B alcohol degradation is found in more than 90% of the white population compared to less than 10% of East Asians, the population attributable risk of heavy drinking and alcoholism by ADH1B.1/1 genotype was 67 and 62% among the white population compared with 9 and 24% among the East Asian population.

Increased rate of treatment with antidepressants in patients with epilepsy

The risk of depressive and anxiety disorders is increased among patients with epilepsy, but it is not known whether patients are treated with antidepressants in clinical practice. In a nationwide case register linkage study, all patients who received a main diagnosis of epilepsy or osteoarthritis on first admission or outpatient contact during the period 1995 to 2000 in Denmark were identified, and rates of subsequent purchases of antidepressants were calculated. Comparisons were also made with a gender-, age-, and calendar-matched sample of the general population. Patients diagnosed with epilepsy had a 1.73 (95% CI: 1.45-2.06) times increased rate of subsequently purchasing antidepressants, compared with patients diagnosed with osteoarthritis and a 2.44 times (95% CI: 2.12-2.81) increased rate compared with the rate among the general population. It is concluded that patients with epilepsy have a moderately increased probability of being prescribed antidepressants in clinical practice.

Parkinson's disease and antidepressant drug treatment: a case-register study

We estimated the nationwide prevalence rate of antidepressant drug treatment in Parkinson's disease (PD) patients.

BACKGROUND

Very few studies exist on the frequency of antidepressant drug treatment in patients with PD.

METHOD

Patients with a PD diagnosis at first hospital contact were identified and followed for up to 6 years. The subsequent probability of antidepressant drug treatment was estimated and compared to a control group of patients with osteoarthritis.

RESULTS

The probability of antidepressant drug treatment was 3.98 [95% CI: 3.23-4.91) times higher for PD patients than for controls.

CONCLUSION

Patients with PD have higher rates of actual antidepressant drug treatment than a control group with osteoarthritis. Still, however, undertreatment of depressive states may be the case.

No Increased Risk of Developing Alzheimer Disease in Patients With Glaucoma

PURPOSE

To investigate whether primary open-angle glaucoma (POAG) is associated with increased risk of developing Alzheimer disease (AD).

METHODS

In a nationwide case register linkage study of patients with hospital admission or outpatient contact during the period from 1977 to 2001 in Denmark, the rate of subsequent AD for patients with a diagnosis of POAG was compared with the rate for patients with primary angle-closure glaucoma (PACG), cataract, and osteoarthritis (OA) and with the rate for the general population.

RESULTS

A total of 11,721 patients with a diagnosis of POAG (including normal tension glaucoma), 5975 patients with PACG, 162,640 patients with cataract, and 230,208 patients with OA were identified in the registers. Patients with POAG did not have increased rate of subsequent AD compared to patients with PACG, cataract, or OA or compared with the general population.

CONCLUSIONS

POAG was not associated with increased risk of developing AD. It cannot be excluded that this negative finding is due to diagnostic misclassification as register data were used.

Increased risk of developing diabetes in depressive and bipolar disorders?

A few studies have suggested that the prevalence of diabetes is increased for patients with depression and for patients with bipolar disorder compared with the general population. However, no study has been published comparing the risk of getting a diagnosis of diabetes for patients with affective disorders with the risk for patients with other medical illnesses. It was the aim of the present study to investigate whether patients hospitalised for depressive or bipolar disorders are at increased risk of getting a diagnosis of diabetes at readmission compared to patients previously admitted for osteoarthritis. In a nationwide case register study, all patients who got a discharge diagnosis of depression, mania/bipolar disorder or osteoarthritis at first admission in a period from 1977 to 1997 were identified. The probability of getting readmitted and discharged with a diagnosis of diabetes was estimated with competing risks models in survival analysis. In total, 29,035 patients with a diagnosis of depression at first discharge, 6683 patients with mania/mixed episode and 108,525 patients with a diagnosis of osteoarthritis were identified. The risk of getting readmitted with diabetes was not increased for patients who had previously been admitted with depression or mania/bipolar disorder compared to patients with osteoarthritis. There was no difference in the risks of developing Type 1 and Type 2 diabetes.

No increased risk of developing depression in diabetes compared to other chronic illness

Several studies have found that the prevalence of depression in patients with diabetes is higher than in the general population but it is unclear whether patients with diabetes have an increased risk of developing depression compared with patients with other chronic illnesses. In a nationwide case register study, all patients who had a discharge diagnosis of diabetes or of osteoarthritis at first admission in a period from 1977 to 1997 were identified. The probability of being readmitted and discharged with a diagnosis of depression was estimated with competing risks models in survival analysis. In total, 91,507 patients with a diagnosis of diabetes at first discharge and 108,487 patients with a diagnosis of osteoarthritis were identified. The median age of the populations was high. No difference in the risk of being readmitted with a diagnosis of depression was found between the two groups of patients. There was no difference in the risk for Type 1 and Type 2 diabetes. It is concluded that older patients with diabetes do not seem to have an increased risk of developing severe depression compared with patients with other chronic illness.

Alcohol use as predictor for infertility in a representative population of Danish women

BACKGROUND

Our aim was to examine the association between use of alcohol and subsequent incidence of primary infertility.

METHODS

The study subjects were chosen from a population-based cohort of Danish women aged 20-29 years. Eligible women were nulliparous and not pregnant (n = 7760). Information on alcohol intake and potential confounders (age, education, marital status, diseases in the reproductive organs, and cigarette smoking) was assessed at enrollment. The incidence of fertility problems during follow-up was obtained by record linkage with the Danish Hospital Discharge Register and the Danish Infertility Cohort Register. Main outcome measures were hazard ratios of infertility according to alcohol intake at baseline estimated in a multivariate Cox proportional hazards model.

RESULTS

During a mean follow-up of 4.9 years, 368 women had experienced infertility. Alcohol intake at baseline was unassociated with infertility among younger women, but was a significant predictor for infertility among women above age 30. In this age group, the adjusted hazard ratio for consuming seven or more drinks per week was 2.26 (95% confidence interval: 1.19-4.32) compared with women consuming less than one drink per week.

CONCLUSIONS

These findings suggest that alcohol intake is a predictor for infertility problems among women in the later reproductive age group.

Affective disorders in neurological diseases: a case register-based study

OBJECTIVE

To investigate the temporal relationships between a range of neurological diseases and affective disorders.

METHOD

Data derived from linkage of the Danish Psychiatric Central Register and the Danish National Hospital Register. Seven cohorts with neurological index diagnoses and two control group diagnoses were followed for up to 21 years. The incidences of affective disorders in the different groups were compared with the control groups, using competing risks to consider the risk of affective disorder and the risk of death in the same analysis.

RESULTS

We found an increased incidence of affective disorders in dementia, Parkinson's disease, epilepsy, stroke and intracerebral haemorrhage compared with control groups. The association was found to be the strongest for dementia and Parkinson's disease. In hospitalized patients, with incident multiple sclerosis, the incidence of affective disorder was lower than the incidence in the control groups.

CONCLUSION

In neurological diseases there seems to be an increased incidence of affective disorders. The elevated incidence was found to be particularly high for dementia and Parkinson's disease (neurodegenerative diseases).

Major depressive disorder in Parkinson's disease: a register-based study

OBJECTIVE

To investigate whether patients with Parkinson's disease (PD) were at an increased risk of developing major depression compared with patients having other medical illnesses with a comparable degree of disability.

METHOD

Case register linkage study of Danish Psychiatric Central Register (DPCR)and Danish National Hospital Register (DNHR). Three study cohorts were identified: all patients with PD, osteoarthritis, and diabetes. The rate of discharge diagnosis of depression on re-admission was estimated using competing risks models in survival analyses. The rates for patients with PD were compared with those of patients with osteoarthritis, and diabetes.

RESULTS

The study sample identified 211 245 patients in the hospital registers with one of the index diagnoses. An increased incidence of developing depression was found for women and men throughout their lifetime when this incidence was compared with the control groups.

CONCLUSION

The findings support the hypothesis that depression in patients with PD is a consequence of brain dysfunction.

Psychopharmacological treatment and psychiatric morbidity in 390 cases of suicide with special focus on affective disorders

OBJECTIVE

The aim of this study was, on the basis of data from health-care registers, to describe the adequacy of psychopharmacological treatment in suicides.

METHOD

Data on consecutive suicides in a Danish County (Funen) in the period of 1 April 1991-31 December 1995 were identified in the Danish Psychiatric Central Register, the National Patient Register, the National Health Insurance and Odense University Pharmacoepidemiological Database.

RESULTS

Twenty-five per cent of the suicides previously hospitalized due to affective disorders and 3% of the suicides without psychiatric hospitalizations at all, received an apparently adequate treatment with antidepressants the month before suicide.

CONCLUSION

The most striking finding was the insufficiency of treatment with antidepressants in the group of suicides without psychiatric hospitalization, in particular in light of the fact that depression is assumed to be present in at least 50% of all suicides.

Smoking and alcohol intake as risk factors for bleeding and perforated peptic ulcers: a population-based cohort study

Both the incidence of and mortality from bleeding and perforated peptic ulcers are increasing. We assessed the association between smoking, intake of alcohol (including type of alcoholic beverage), and risk of a complicated peptic ulcer in a population-based study of 26,518 Danish subjects followed up for an average of 13.4 years. There were 214 cases of incident bleeding and 107 cases with perforated ulcers. We estimated relative risks (RRs) for incident bleeding and perforated peptic ulcers using Poisson regression analysis. Smoking more than 15 cigarettes per day compared with never smoking increased the risk of a perforated ulcer more than threefold [RR = 3.5; 95% confidence interval (CI) = 1.7-7.1)]. Drinking more than 42 drinks per week increased the risk of a bleeding ulcer fourfold (RR = 4.4; 95% CI = 2.3-8.3) compared with drinking less than one drink per week. Using the same comparison group, subjects who drank more than 21 drinks per week but no wine were at a higher risk of a bleeding ulcer (RR = 8.8; 95% CI = 2.2-35) than drinkers of the same amount of alcohol, but with more than 25% of their intake as wine (RR = 2.4; 95% CI = 1.0-6.0).

Tobacco smoking and risk of hip fracture in men and women

BACKGROUND

Previous findings suggest that tobacco smoking increases the risk of hip fracture in women. A similar adverse effect of smoking is suspected to be present in men, but bone mineral density studies have raised the concern that men may be more sensitive to the deleterious effect of smoking on bone than women. In this study we prospectively determined the influence of current, previous, and cumulative smoking history on risk of hip fracture in men and women and addressed the issue of possible gender difference in the susceptibility to tobacco smoking.

METHODS

Pooled data from three population studies conducted in Copenhagen with detailed information on smoking habit. A total of 13,393 women and 17,379 men, initially examined between 1964 and 1992, were followed until 1997 for first admission due to hip fracture. The relative risks (RR) of hip fracture associated with smoking were estimated by means of multiplicative Poisson regression models.

RESULTS

During follow-up, 722 hip fractures were identified in women, and 447 in men. After adjustment for potential confounders, including body mass index, female current smokers had an RR of hip fracture of 1.36 (95% CI: 1.12-1.65) and male smokers 1.59 (95% CI: 1.04-2.43) relative to never smokers. In both sexes, the RR of hip fracture gradually increased by current and accumulated tobacco consumption. The RR were consistently higher in men than in women, but the test for interaction between sex and tobacco smoking was insignificant. After 5 years, male ex-smokers had an adjusted RR of 0.73 (95% CI: 0.55-0.98) relative to current smokers, while no significant decrease in risk was observed in female ex-smokers (RR = 0.91; 95% CI: 0.72-1.17)). Approximately 19% of all hip fractures in the present study population were attributable to tobacco smoking.

CONCLUSION

Tobacco smoking is an independent risk factor for hip fracture in men and women, and there appears to be no gender differences in smoking related risk. Smoking cessation reduces the risk of hip fracture in men after 5 years, while the deleterious effect of smoking seems to be more long-lasting in female ex-smokers.

Femoral fractures in postmenopausal breast cancer patients treated with adjuvant tamoxifen

The anti-estrogen tamoxifen is the prevalent endocrine treatment in postmenopausal breast cancer patients. However, nothing is known about the long-term effects of the drug on the skeleton as assessed by the occurrence of fractures. We investigated the occurrence of fractures of the femur in patients from a Danish Breast Cancer Cooperative Group (DBCG) trial initiated in 1977 by a linkage of data from the Danish National Registry of Patients with data from the DBCG registry. 1716 postmenopausal women with high-risk breast cancer were randomized to local radiotherapy with or without tamoxifen, 30 mg daily for 1 year. Fifty-one patients in the control group had one femoral fracture and 64 tamoxifen treated patients had one femoral fracture. Eleven patients in the control group had one trochanteric fracture compared to 27 patients in the tamoxifen group (logrank = 5.28. P = 0.022; hazard ratio = 2.12, 95% CL 1.12, 4.01). The results could not be explained by a longer survival in the tamoxifen group nor by bone metastases with pathological fractures. In conclusion, our study suggests that tamoxifen does not seem to offer protection against fractures in old age and may even increase the risk of fractures at particular sites. This hypothesis needs to be disproved or confirmed in other trials.

Baseline characteristics are not sufficient indicators of non-response bias follow up studies

STUDY OBJECTIVE

The aim was to examine whether baseline characteristics from a cross sectional survey provided sufficient information regarding non-response bias in a follow up study when compared with information on hospital admissions in the intervening years.

DESIGN

This was an 11 year follow up study of a cohort selected in 1974 with register information on hospital admissions during follow up.

SETTING

The study was based on a sample of cement workers from a particular Portland cement factory with suitable controls from other occupations.

PARTICIPANTS

A total of 1404 men participated in the first survey in 1974, including a questionnaire and lung function tests. In 1985 1070 men were alive and of these, 928 men (87%) responded to a postal questionnaire.

MAIN RESULTS

Non-responders in 1985 did not differ markedly from responders when smoking habits, respiratory symptoms, and lung function were examined in 1974. During follow up, non-responders had twice as high rates of hospital admission due to respiratory diseases as responders. These differences remained present after adjusting for minor differences in age and smoking habits.

CONCLUSIONS

Equal distributions of baseline characteristics among responders and non-responders in a follow up study do not preclude non-response bias.