Background: European population cohorts continue to expand our understanding of chronic airways disease and inter-study collaboration may help address the inevitable limitations of study size, duration, era and geography. Towards this aim, CADSET has collated data from ten major general population European cohorts: Asklepios; Copenhagen City Heart Study; Copenhagen General Population Study; ECRHS; HUNT; LEAD; Lifelines, OLIN, Rotterdam Study and WSAS. We included males and females aged 20 to 95 years with baseline demographic and spirometry data.

Results: Data from 262,829 individuals (44% male) from multiple European countries provided good coverage across all adult ages (Fig.1A). Recruitment occurred in every year from 1976 through 2020. 23% were current-smokers and 42% were never-smokers, a pattern varying with advancing age (Fig.1B). The prevalence of airflow limitation varied according to whether lower limit of normal (LLN) or <0.70 thresholds were applied, increasing with age if the latter was used (Fig.1C).

Interpretation: These results fit with previous reports, however the size, geographical reach and span of recruitment provided by this collaboration provides a unique opportunity to explore chronic airways disease development. Together, we are now pursuing research questions previously beyond the scope of individual cohort studies.

Background: Adult onset asthma is poorly studied and there are few long-term clinical follow-up studies.

Aim: To study clinical characteristics of adult onset asthma in a 15-year follow-up.

Method: Within the Obstructive Lung Disease in Northern Sweden (OLIN) studies, a cohort of n=309 subjects with adult onset asthma (aged 20-60 years) was recruited during 1995-99. The cohort was followed up in 2012-14 (n=205). Structured interviews and clinical examinations including spirometry were performed at both recruitment and follow-up. Skin prick tests were performed at recruitment and blood samples for cell counts and IgE at the follow-up. Asthma control was classified according to GINA 2006.

Results: At follow-up n=182 (89%) still had asthma, while n=23 (11%) were in remission. Among individuals with persistent asthma, mean pre-bronchodilator FEV1 percent of predicted was 89.0 at follow-up, similar as recruitment 88.3. At recruitment 16.5% were smokers, and of these, 86.7% had quit smoking at follow-up. At follow-up, 39% had blood neutrophils ≥4.0x109/L, 23% had blood eosinophils ≥0.3x109/L, and 28% had specific IgE>0.35 IU/ml to any airborne allergen. Any respiratory symptoms were reported by 90% and 31% used medium or high dose inhaled corticosteroids (ICS), 20% low dose ICS whereas 20% had no treatment. 55% had controlled asthma, 32% partly controlled and 13% uncontrolled asthma.

Conclusion: In this 15-year follow-up of adult onset asthma, the majority had persistent asthma. Smoking and high proportion using ICS may contribute to the stable lung function. Still, it should be noted that merely around every other had well controlled asthma.

Introduction: The prevalence of COPD seems to have leveled or decreased in some high income countries. Worldwide, population studies using spirometry have found substantial underdiagnosis of COPD.

Aim: To investigate underdiagnosis and misclassification of COPD in Sweden.

Methods: A randomly selected sample (n=1839; age 21-78 years; women 52.6%) was derived from two large scale population studies in Sweden, the West Sweden Asthma Study (WSAS) and the Obstructive Lung Disease in Northern Sweden (OLIN) Studies. Examinations including spirometry and structured interviews were conducted 2009-2012. COPD was defined following GOLD 2017: post-bronchodilator FEV1/FVC<0.70 and presence of respiratory symptoms. The subjects were asked if prior to the study they had been diagnosed with COPD, chronic bronchitis, emphysema, asthma, and if they used medicines for airway diseases.

Results: Of those with COPD (GOLD) identified at the clinical examinations, 24% had, prior to the study, been diagnosed with COPD, emphysema, or chronic bronchitis (COPD-cluster), while 46% had been diagnosed with the COPD-cluster, asthma, or used medicines for airway disease (any OAD). The corresponding figures for those fulfilling the criteria of COPD GOLD grade ≥2 were 39% (COPD-cluster) and 62% (any OAD). Among the subjects who, prior to the study, had been diagnosed with COPD, 78% fulfilled the criteria of COPD, but among those who had been diagnosed as having any of the diseases in the COPD-cluster, 31% fulfilled the criteria of COPD.

Conclusion: The proportion of underdiagnosis of COPD is still large in Sweden. Some misclassification of COPD was also found. The results reflect underuse of spirometry in primary care.

Background: Little is known on associations between respiratory symptoms and mortality.

Aim: To study whether respiratory symptoms are risk factors for all-cause, respiratory, cardiovascular (CV), and cancer mortality in Sweden and Norway.

Methods: In 1995-1997, population samples (20-69y) were surveyed about respiratory symptoms, and n=7,104 (85.3% of invited, median age 45y) and n=54,240 (70.1%, 44y) participated within the OLIN Studies in Northern Sweden and the HUNT Study in Norway. Mortality was studied until December 31st 2015. Hazard ratios (HR) for associations between respiratory symptoms and mortality were estimated by Cox regression models adjusted for age, sex, educational level, and smoking habits.

Results: The cumulative 20-year mortality was 14.5% in OLIN and 12.6% in HUNT. Dyspnea (mMRC grade≥2) (HR 1.9, 95%CI 1.6-2.2 in OLIN and 1.6, 1.5-1.7 in HUNT), chronic productive cough (1.5, 1.3-1.8 and 1.5, 1.3-1.6), and wheeze (1.3, 1.1-1.5 and 1.3, 1.2-1.4) were associated with increased risk of all-cause mortality. Women reported dyspnea and wheeze more frequently than men in both countries, but the association with mortality was similar in both sexes. Causes of death were studied in OLIN, where dyspnea associated with increased risk of respiratory (3.6, 2.1-6.1), CV (2.1, 1.6-2.7), and cancer (1.3, 1.0-1.8) mortality. Chronic productive cough was associated with increased risk of respiratory (2.4, 1.3-4.3) and cancer (1.6, 1.2-2.2) mortality, while wheeze was associated with increased risk of respiratory (3.5, 2.1-5.7) and CV (1.3, 1.0-1.6) mortality.

Conclusions: Common respiratory symptoms were similarly associated with increased risk of mortality in adults in Sweden and Norway.

Background: Severe asthma is a considerable challenge for patients, health care professionals and society, but there are few long-term studies on risk factors for severe asthma.

Aim: To identify baseline risk factors of severe asthma in a longitudinal adult asthma cohort study.

Methods: An adult asthma cohort was identified in 1986-2001 by clinical examinations of population samples within the OLIN studies in northern Sweden. The examinations included structured interviews, spirometry with reversibility testing, skin prick testing and metacholine challenge. The cohort was followed up in 2012-2014 when n=1006 participated (mean age 59y). Adjusted Risk Ratios (RR) for baseline factors as risk factors for GINA defined severe asthma (SA) at follow-up (n=51) were estimated by Poisson regression.

Results: Older age, impaired lung function, increased reversibility and hyperreactivity, asthmatic wheeze, persistent wheeze, nighttime awakenings due to respiratory symptoms, and dyspnea were significant baseline risk factors for SA. Allergic sensitization, smoking, occupational groups or BMI did not predict SA. When adjusted for age, sex and smoking, post-bronchodilator FEV1/FVC<0.7, both present at baseline (RR 4.2, 95%CI 1.8-9.9) and developed during follow-up (2.9, 1.6-5.3), increased the risk. Also FEV1<80% at baseline associated with SA (2.9, 1.6-5.2). Triggers for respiratory symptoms at baseline such as tobacco smoke (2.1, 1.2-3.7) and physical activity (3.5, 1.5-81) associated with SA at follow-up, while pollen and furry animals did not.

Conclusion: Among adults with asthma, impaired lung function, wheeze, dyspnea and nighttime awakenings due to respiratory symptoms are important long-term risk factors for severe asthma.

Background: Quality-adjusted life-years (QALYs) is commonly used in health-economic evaluations. QALY-weights combine health status and time into one measure.

Aims: To investigate the association between multimorbidity and QALY-weights among adults with asthma and COPD.

Methods: Within the OLIN-studies in northern Sweden, a random sample was invited to a postal questionnaire survey. A random sample of 1016 responders was invited to clinical examinations and interviews in 2009 (737 participated, ages 21-86 years), of which 605 completed the health-related quality of life (HRQL) questionnaire SF-36. QALY-weights were derived from the SF-36 data using the SF-6D tool via the standard gamble method. The SF-6D scores are equivalent to QALY-weights with low values representing poor health and the score one representing perfect health.

Results: Of the 605 participants, 74 had current asthma, 81 had COPD (FEV1/FVC<0.7), 66 had heart disease, 30 had diabetes, 30 had rheumatic disease, and 160 had hypertension. There was an association between an increasing number of morbid conditions and lower QALY-weights (p<0.001). The mean QALY-weight tended to be lower among subjects with asthma compared to COPD, 0.77 and 0.81, respectively (p=0.078). Subjects with COPD and two or more non-respiratory conditions had significantly lower QALY-weights compared to subjects with COPD alone (0.75 vs.0.83, p=0.016). No significant difference in QALY-weights was found among asthmatics with versus without other non-respiratory conditions.

Conclusions: Subjects with asthma tended to have lower QALY-weights compared to subjects with COPD. Having two or more non-respiratory conditions affected the QALY-weight negatively among subjects with COPD but not among subjects with asthma.

Background: Ischemic heart disease (IHD) is a common multimorbidity in individuals with COPD. High sensitive cardiac troponin I (hs-cTnI) has been shown to predict short-term mortality, but longer follow-ups has rarely been performed in population-cohorts.

Aim: To evaluate the predictive value of elevated hs-cTnI on mortality among individuals with COPD compared with normal lung function (NLF).

Methods: In 2002-04, subjects with FEV1/VC <0.70 (COPD, n=993) and age and sex-matched referents withoutCOPD were identified from OLIN’s population-based cohorts. In 2005, structured interviews, post-bronchodilator spirometry, blood sampling and ECG were performed in individuals with COPD (n=599) and NLF (n=756). Hs-cTnI was analysed in serum and concentrations ≥5 ng/L were defined as elevated. Mortality data were collected until 2016.

Results: In 2005, the prevalence of reported IHD and elevated hs-cTnI was higher in COPD than NLF (16.2% vs 11.9% p=.02 and 31.1% vs 25.0% p=.01). The cumulative mortality was higher in COPD than NLF, both overall (36.5% vs 19.2% p<.001), and when restricting comparison to individuals with hs-cTnI≥5 (59.1% vs 34.9% p<.001). In a Cox-regression model adjusting for common confounders including reported IHD and ischemic ECG changes, hs-cTnI≥5 was associated with an increased risk for death in COPD (HR 1.41, 95%CI 1.03-1.93), but not in NLF (HR 0.84 95%CI 0.58-1.22). The increased risk remained after adjusting for FEV1% predicted.

Conclusion: Elevated hs-cTnI was associated with increased mortality over a 11 -year follow-up among individuals with COPD, but not among those with NLF in this population-based study. The use of troponin could identify individuals with stable COPD at the highest risk of death.

Background: During the second half of the 20th century the prevalence of asthma has increased and smoking habits and socioeconomy has changed in Sweden.

Aim: To analyze risk factor patterns for current asthma in 1996, 2006 and 2016.

Methods: Three cross-sectional random samples from the same area of Sweden, 20-69 years, participated in population surveys with the same validated questionnaire in 1996 (n=7104 participants, 85% response rate), 2006 (n=6165, 77%) and 2016 (n=5466, 53%). Current asthma was defined as physician-diagnosed asthma with symptoms common in asthma and/or asthma medication use during the last 12 months. Socioeconomic classification was based on occupation. Risk factors for current asthma were analyzed by multivariable logistic regression.

Results: A family history of asthma yielded the highest odds ratios (OR) all three years with ORs 3.19-3.66. Data on occupational exposure to gas, dust or fumes (GDF) was not available in 1996, but associated with current asthma in both 2006 (1.86, 1.51-2.30) and 2016 (1.70, 1.37-2.10). Ex-smoking was a risk factor in 1996 (1.39, 1-12-1.73) and 2006 (1.38, 1.11-1.71) but not in 2016 (1.15, 0.92-1.44). Manual workers in service and industry, non-manual employees and self-employed had increased risk for current asthma in 2016, but not in 1996 or 2006, compared to professionals and executives.

Conclusion: The risk factor pattern for asthma among adults has changed in Sweden from 1996 to 2016. While occupational exposure to GDF was a prominent and constant risk factor for asthma, ex-smoking lost significance in 2016, and socioeconomic differences with regard to occupation emerged.

Background: Skeletal muscle dysfunction is a known systemic- or extra pulmonary effect in COPD. The relationship between the severity of the disease according to GOLD’s spirometric criteria and muscle dysfunction is however not clearly explored.

Aim: To investigate hand grip strength (HGS) in subjects with COPD compared to subjects without airflow obstruction.

Method: Lung function and HGS was measured according to validated methods in the years of 2009-2010 within the Obstructive Lung Disease in Northern Sweden (OLIN) COPD-study. HGS, was measured with a hand-held dynamometer (Jamar®) in kilogram-strength (kgf), which has been shown to correlate strongly with both upper- and lower extremity strength. The sample included 441 subjects with COPD according to GOLD criteria (FEV1/FVC<0.70) and 571 non-COPD subjects in the same ages.

Results: Among men, HGS was lower in GOLD III-IV, mean=31.4 kgf, compared to GOLD II, mean=37.2 kgf (p=0.009), GOLD I, mean=36.9 kgf (p=0.024), and subjects without COPD, mean=37.6 kgf (p=0.024). Among women no such correlation was found. In the whole sample, HGS, did not differ significantly between subjects with and without COPD, or COPD stages according to GOLD. No correlation was found between percent predicted FEV1 but a strong and positive relationship was found between the crude value of FEV1 and HGS (pearson's r=0.67).

Conclusion: Among men, GOLD III-IV was associated with a lower HGS, however not among women.