Introduction

The majority of patients with chronic obstructive pulmonary disease (COPD) have at least one other coexisting chronic disease.1 Some comorbid disorders have shared pathophysiology like systemic inflammation, and others have shared risk factors like older age and smoking.2 Comorbidity in COPD is associated with worsened health status and an increased risk of exacerbations and mortality.3,4

A recent Swedish cohort study from primary care reported a higher prevalence of several comorbid conditions in COPD patients two years prior to the COPD diagnosis and two years subsequently, compared to an age and sex matched reference population without COPD.5 The prevalence of diseases two years post-diagnosis increased in both groups, but the increase was significantly higher in COPD patients, especially regarding cardiovascular diseases, diabetes, and depression.

However, the extent to which this increased risk of morbidity is present years earlier than the first hospital COPD diagnosis and possibly before the typical ages of clinical COPD onset, is not clear. Information about this and the long-term development of comorbidity in COPD patients could improve our understanding of multimorbidity in COPD and potentially help clinicians provide these patients with optimal holistic care.

This study aims to describe the life-course burden of morbidity in patients with COPD compared to those without COPD in a general population cohort using Swedish national registers.

Materials and Methods Study Population and Design

The initial data included all residents in Sweden 25 years of age or older and identified in the Swedish Total Population Register from 1990 to 2019. The register includes information on birth, death, and migration of all residents in Sweden. The study population was limited to all individuals who were alive and between 50 and 90 years in 1997. Individuals immigrating to Sweden after 1987 were not included due to incomplete health data.

We used two study designs. First, we used a nested case–control study design to investigate the risk of various disease diagnoses over time for future COPD patients and their matched comparators (controls). Patients with COPD were identified using hospital inpatient or outpatient diagnosis from 1997 to the end of 2018. For both COPD patients and their controls, we identified disease diagnoses from 1st January 1987 to the day of COPD diagnosis (or the date of matching for controls). The date of first COPD diagnosis was used as the index date, and for each case of COPD, we sampled five individuals (controls) who did not have COPD by the index date from the study population. The controls were matched by birthyear, sex, and county of residence at the index date. A control could only serve as a control once, and controls who later became a case only served as case. The majority had five controls per case, but 33% had fewer.

Second, we used a cohort study design to investigate the risk of morbidity subsequent to the COPD diagnosis. Follow-up time started on 1st January 1997 and ended with death, emigration, age 90 years, date of first diagnosis of the outcome, or on 31st December 2018, whichever occurred first. COPD diagnosis was modelled as a time-varying variable.

In both studies, individuals with missing data on covariates were excluded from the analysis.

Variables in the Study COPD Diagnosis

We identified patients with COPD from the National Patient Register using the ICD-10 code J44 from 1997 and onwards. The register contains information on diagnoses from inpatient care with complete national coverage since 1987, and of hospital-based outpatient care from 2001.6 We used both primary and secondary hospital diagnoses for the identification of COPD. We identified COPD from age 50 years. Median and mean age at diagnosis have been found to be 68 years in previous studies.7,8

Other Morbidity

We used the National Patient Register and the Cancer Register between 1987 and 2018 (see Supplemental Table 1 for full list) to obtain data on other morbidity and defined the first recorded date of diagnosis as the incidence date.9 The diseases were selected based on previous literature and with consideration to the Patient register not containing data from primary care.1,3–6 Data on underlying and contributing causes of death were obtained from the National Cause of Death Register and we considered them together in three main categories not mutually exclusive: cardiovascular, respiratory and cancers (see Supplemental Table 1 for ICD-codes).10

Covariates

Highest attained level of education was used as a marker of socioeconomic characteristics and was obtained from the Longitudinal Integrated Database for Health Insurance and Labour Market Studies in 1997.11 Data on county of residence and date of birth, death, and emigration were extracted from the Total Population Register. The county of residence was defined at the year of matching in the nested case–control study and in 1997 in the cohort study.

Statistical Analysis

For descriptive data, categorical variables were summarised using frequencies and percentages. Continuous variables were summarised by median and interquartile range (IQR) given their distributions. The Chi-square test was used to compare differences between the COPD and non-COPD group in attained level of education.

Nested Case–Control Study

To examine the risk of pre-COPD morbidity, we analysed the matched case–control data using conditional logistic regression models. Odds ratios (OR) with 95% confidence intervals (CI) were calculated. The analyses were adjusted for level of attained education in 1997 and stratified by sex. For assessment of the accumulating burden of comorbidity, and to identify possible age patterns, we repeated the analysis focusing on pre-COPD morbidity occurring before 50 years of age. The analysis was further repeated to include pre-COPD morbidity occurring up to ages 65 years and 75 years.

Cohort Study

We performed Cox regression analyses to assess risk of comorbid disease and cause of death following the first COPD diagnosis. Each comorbid disease was analysed separately. Age was the underlying timescale. Hazard ratios (HR) with 95% CI were calculated. COPD diagnosis was modelled as a time-varying covariate, where COPD status changed from “no” to “yes” at date of first COPD diagnosis. The proportional hazards assumption was tested graphically using Kaplan-Meier, smooth hazard function, and log-minus-log curves. For melanoma and brain tumours, the proportional hazards assumption appeared violated towards older ages, so right truncation at age 80 years (melanoma) and 70 years (brain tumour) was implemented. The analyses were performed unadjusted and adjusted for sex, level of attained education, and county of residence in 1997. Analyses were also performed stratified by sex. We assessed the accumulation of comorbidity by age, by repeating the Cox regression analyses and ending follow-up at 65 years and 75 years.

Sensitivity Analyses

We performed two sensitivity analyses. First, to examine the effect of missing data, we compared HRs obtained from unadjusted analyses with and without excluding individuals with missing data. Second, to examine possible surveillance bias, we defined COPD only using primary diagnoses and refitted models for both nested case-control and cohort studies.

Data management was performed using SPSS version 28 and Stata version 17. Statistical analysis was performed using Stata version 17.

Ethics

Following approval by the national Swedish Ethical Review Authority for this study (2019–04755 and 2023–03585-02), the organisations responsible for the data approved and supplied the variables. The National Board of Health and Welfare provided data from the following registers: the National Patient Register, the Cancer Register, and the National Cause of Death Register.

Statistics Sweden provided data from the following registers: the Total Population Register and the Longitudinal Integrated Database for Health Insurance and Labour Market Studies. All data received from the registers are pseudonymised.

Results Participants Characteristics

In the nested case–control study, there were 223945 individuals with COPD (cases) and 1062731 individuals without COPD (controls). Cases without information on county of residence in the year of COPD diagnosis or with no available matched controls were excluded prior to analysis (n = 1214). The median age at COPD diagnosis was 76 years and 50.2% were female (see Supplemental Table 2).

In total, 2966804 individuals were alive and between ages 50 and 90 years in 1997. Some 2% (n=70,647) had immigrated to Sweden after 1987 and were not included in the analysis due to incomplete health data. After exclusion due to lack of data on attained level of education or county of residence (n=189,343), the cohort study included the remaining 2706 814 individuals with complete data. Of these, 225159 had a diagnosis of COPD sometime during follow-up (1997–2018). Median age at the start of follow-up (1997) was 64 years for both the COPD and the non-COPD group. The non-COPD group had higher attained level of education (Ρ <0.001). In the COPD group 50.1% were female, compared with 53.5% in the non-COPD group.

Diagnoses Prior to First Hospital COPD Diagnosis: Nested Case–Control Study

The results of the nested case–control study investigating the risk of comorbid diseases prior to the first COPD diagnosis are shown in Table 1. Individuals with a future diagnosis of COPD showed higher risks of having most of the studied conditions. Exceptions were dementia, Parkinson’s disease, melanoma and cancer in the brain and endometrium, where we instead found statistically significantly lower risks for future COPD patients. Among the cardiovascular diseases, highest risks were found for chronic heart failure and peripheral arterial disease. When data were stratified by sex, the ORs differed by sex for several conditions. For osteoporosis, men had OR 2.48 (95% CI 2.32–2.64) and women OR 1.92 (95% CI 1.86–1.97).

Table 1 Risk of Diseases Prior to COPD Diagnosis

The results were consistent when only primary diagnoses of COPD were used (data not shown).

Risk of Pre-COPD Morbidity by Age

When the analysis focused on morbidity identified by age 50 years, future COPD patients had an increased risk of having almost all studied conditions, except for most cancers (Table 2). High magnitude associations were seen for lung cancer and chronic heart failure. For most diseases, ORs decreased with increasing age, as the occurrence of the diseases increased among non-COPD individuals. Dementia had a positive association prior to age 50 years, shifting towards an inverse association by age 75 years.

Table 2 Risk of Diseases Prior to COPD Diagnosis by Age

Diagnoses Subsequent to First Hospital COPD Diagnosis: Cohort Study

After the first hospital COPD diagnosis, individuals with COPD showed a higher risk of being diagnosed with most comorbid conditions including dementia (Table 3). When looking separately at vascular dementia and Alzheimer’s, the risk of vascular dementia was increased with HR 1.53 (95% CI 1.48–1.58), while the hazard ratio for Alzheimer’s was not significantly different from the non-COPD population (HR 1.02, 95% CI 0.98–1.05). The risk of Parkinson’s disease was lower, while the risk of subsequent depression was increased. Except for melanoma, endometrial and prostate cancer, there were positive associations for all cancers. Lung cancer had the highest estimated relative risk. The results for melanoma were truncated at age 80 years, and for brain cancer at age 70 years due to violation of the proportional hazard assumption. The three examined causes of death showed positive associations, the highest magnitude HR found for respiratory causes. The sex-stratified analyses showed higher risks for women than men for all cardiovascular diseases, with statistically significant interaction terms (Ρ < 0.001). For osteoporosis, the sex-stratified analysis showed a higher risk for men than women (Ρ < 0.001).

Table 3 Risk of Comorbid Disease After COPD Diagnosis

Results were consistent when only primary COPD diagnoses were used and when sample with complete data was compared with all available sample (data not shown).

Risk of Post-COPD Morbidity by Age

When follow-up was truncated up until 65 years and 75 years (see Table 4), there was a pattern of decreasing relative risks with increasing age for most conditions.

Table 4 Risk of Disease After COPD Diagnosis by Age

Discussion

The first main finding of this nationwide study in Sweden was that individuals with a future inpatient or outpatient hospital diagnosis of COPD, already by 50 years of age, had a higher risk of chronic morbidity than individuals who remained COPD-free until the study end. Second, both prior to and following the first COPD diagnosis, patients with COPD had a higher risk of cardiovascular, endocrine, gastrointestinal, psychiatric, and respiratory diseases, than non-COPD controls.

This study aimed to describe the life-course patterns of disease accumulation, both the expected and possibly unexpected, and not causation. Smoking induces systemic inflammation and is a known risk factor for cardiovascular diseases and cancers, as well as COPD.12,13 A Danish study of never-smokers with COPD found that they had less severe disease, limited to the lungs, while smokers and ex-smokers with COPD had an increased risk of cardiovascular comorbidity.14 A previous study reported that the current smoking rate in Swedish COPD patients was 34% in 2004.15 Some 42% were reported ex-smokers. For participants without COPD, the rates were 13% and 40%, respectively. Thus, the non-COPD population in this study is likely to have a lower rate of both current smokers and ex-smokers than the population with COPD. Our findings may reflect the systemic nature of COPD, as well as shared risks, such as tobacco smoking.

Previous research has indicated that age-related chronic disorders occur at younger ages in COPD patients than others and our results are consistent with this.16 We found a pattern of decreasing relative risks with increasing age for most diagnoses, as disease prevalence increased with increasing age in the non-COPD controls. Interestingly, even prior to the COPD diagnosis, the risk of other chronic morbidity was increased. Since COPD was defined by the earliest diagnosis registered at hospital in our study, it is likely that the onset of COPD was at least some years earlier, and possibly already diagnosed in primary care. In Sweden, 59% of COPD patients were first diagnosed in primary care in 1999, which increased to 81% in 2009.8 The average age at diagnosis in primary care was 73 years in 1999, and 66 years in 2009.8 In our study, the median age at diagnosis was 76 years, so the first hospital diagnosis of COPD is suggested to be later for most patients. The accumulation of morbidity we found indicates that the disease accumulation started long before the COPD diagnosis is evident. The similarity between the results of the cohort study and the nested case-control suggests a greater susceptibility to morbidity both before and after diagnosis of COPD.

As is well known and detailed by a recent meta-analysis, patients with COPD had a higher risk of cardiovascular diseases.17 In our study, we found higher risks of all cardiovascular diseases for COPD patients than non-COPD patients, both prior to and after the COPD diagnosis. The relative risks were most notably increased at younger ages, and sex-stratified analyses showed higher risks for women than men. The high frequency of cardiovascular diseases and the often-shared symptoms with COPD makes it a comorbidity to be expected in COPD patients. We found an increased risk of cardiovascular mortality in COPD patients, as well as for respiratory causes and cancer. Hospital admission for an acute exacerbation has been associated with increased mortality in cardiovascular events.18

Osteoporosis is a common comorbidity of COPD, sharing risk factors like age, smoking, and systemic inflammation.19 We found an increased risk of osteoporosis both before and after a COPD diagnosis and especially for men. The age-defined analyses showed higher relative risk especially at younger ages and prior to the first COPD diagnosis. COPD has been reported to be the most frequent cause of secondary osteoporosis in men.20 In clinical practice, COPD may be a pointer to look for osteoporosis, which may more often be overlooked in men.

The risk of psychiatric disorders was higher for individuals with COPD, both before and after the COPD diagnosis, but especially for subsequent depression. Individuals diagnosed with psychiatric disorders have been found to be more likely to smoke than those without such diagnoses,21 thus potentially increasing the risk of COPD. Depression is a common comorbidity of COPD, associated with increased risk of exacerbations and mortality.3 Still, depression is often unrecognised and untreated in COPD patients.22 Our findings suggest that a more active approach towards diagnosis and treatment could be of value.

Smoking has been suggested having a protective effect for development of Parkinson's disease.23 Also in our study, COPD patients had a lower risk of having Parkinson’s disease both before and after the COPD diagnosis. Observational studies have reported an increased risk of cognitive decline and dementia in patients with COPD.24 A recent Mendelian randomisation study did not find any evidence of a causal association between COPD and Alzheimer’s disease, and suggested unmeasured confounding, or other types of dementia as an explanation.25 We found a significantly increased risk of vascular dementia in COPD patients, while the risk of Alzheimer’s was not significantly different from the non-COPD population. Prior to the COPD diagnosis, we found a lower risk of dementia in future COPD patients compared to those who would not develop COPD. Thus, clinicians should be aware of subsequent vascular dementia, which is consistent with the generally elevated risk of cardiovascular comorbidities in COPD.

A Danish study reported increased incidences of both tobacco-related and other cancers in COPD patients compared with the general population.26 We found increased risks for most cancers, both before and after the diagnosis of COPD, exceptions being melanoma and endometrial cancer. Previous observational studies have reported inverse associations for endometrial cancer and melanoma with cigarette smoking.27,28

Our findings of increased morbidity among COPD patients are consistent with recent studies from other European countries. In Denmark, a population-based study found the highest increased risks among those with COPD for mood disorders, osteoporosis, heart failure, smoking-related cancers, and peripheral arterial disease, compared with the general population.29 In Germany, patients with COPD had higher relative risks for cardiovascular diseases, diabetes, chronic kidney disease, osteoporosis, and psychiatric diseases, among others.30

Early diagnosis of COPD has been advocated as a possible way to improve management of the disease.31 Further research should quantify the value of clinicians being vigilant and considering early diagnosis of COPD in individuals with other chronic morbidity and identifying likely comorbid diseases at an early stage through monitoring, providing the opportunity to plan management strategies that consider COPD in the context of other conditions.

Strengths and Limitations

Strengths of our study are the uniquely large sample size and the use of longitudinal national registers with high coverage and prospectively recorded data.6,9,10 The Patient Register has high validity for some diagnoses but less likely to identify some conditions commonly treated in primary care, such as hypertension.6 The Swedish healthcare system provides tax-supported health care to all residents, why inability to pay healthcare fees should not be a source of bias in the study. The Cause of Death Register has high completeness data, but the quality of the diagnoses is dependent on the clinicians, and agreement between the register and systematically reviewed medical records has been found to vary between age groups and diagnosis.10

This study has some potential limitations that are typical for large register-based studies. We defined the date of diagnosis of COPD and other morbidity as the earliest date of a record in the Patient Register, thus the first admission or referral to hospital. The patient may have received the diagnosis in primary care prior to this. Hence, the index date of COPD is partly arbitrary but also reflecting the clinical reality: COPD patients often have had lower respiratory symptoms a long time before finally being diagnosed with COPD.7 We identified COPD from age 50 years, as it is usually diagnosed later in life.8 In our study, the median age of COPD diagnosis was 76 years. However, to have undiagnosed COPD cases among the controls or non-COPD reference group would attenuate rather than inflate the associations.

We have no information about whether the diagnosis of COPD was confirmed by spirometry, and this may have implications for diagnostic accuracy. Although the use of diagnostic spirometry for COPD has increased in recent years in Sweden, its use remains incomplete.32 A previous validation study found that less than 10% of the COPD diagnoses in the Swedish Inpatient Register were misclassified or uncertain.33 Yet, there may be an overdiagnosis of COPD in patients admitted to hospital.34 However, this diagnostic heterogeneity of COPD in our study would again reflect the clinical reality.

This study was designed to examine COPD patients that are at some point treated in hospital. Therefore, patients that are managed strictly in primary care and who have never been referred to secondary care were not included in the study. A previous Swedish population-based study on inpatient care estimated that approximately 20–40% of patients with airflow limitation corresponding to COPD were never admitted to hospital.35 The admission rate was highest for male smokers, and lowest for female never-smokers and ex-smokers. Thus, the COPD patients identified in this study may represent the individuals with more severe disease than average.

Another limitation was that we were not able to identify lifestyle risk factors such as smoking, although the focus of this study was to identify patterns of disease accumulation, not the associated risk factors. The increased surveillance of people having one or more chronic disease diagnoses would likely inflate the number of new diagnoses, possibly resulting in surveillance bias if one group is more or less likely to receive a disease diagnosis, investigated in this study, than the other. However, this would be true both of people with COPD and people with other chronic disorders. During the study period, several county councils used payment systems based on the number of diagnoses, which have been found to increase the number of secondary diagnoses in the records.36 Since this may strengthen the possibility of surveillance bias, we performed sensitivity analyses, including only those with a primary diagnosis of COPD. The results were consistent with the main analyses, and surveillance bias does not seem to be the main explanation of our results. The results of our study may have limited generalisability in countries with different healthcare systems or smoking habits.

Conclusion

Accumulation of chronic morbidity may precede a COPD diagnosis. Following the COPD diagnosis, a higher-than-average risk of multimorbidity should be anticipated and addressed in clinical practice. An increased burden of cardiovascular disease and cancer is to be expected, but subsequent depression, vascular dementia, and osteoporosis – particularly among men – should also be anticipated.

Data Sharing Statement

The data underlying this study cannot be shared publicly due to regulations under the relevant Swedish laws. Researchers who are interested in Swedish register data can refer to https://www.registerforskning.se/en/. Inquiries about the data used for this study can be addressed to the corresponding author.

Funding

The data used for the study was obtained with support from the Swedish Research Council for Health, Working Life and Welfare (number: 2019-01236).

Disclosure

BS has received honoraria for educational activities and lectures from AstraZeneca, Boehringer Ingelheim, Chiesi, Meda, Novartis and Teva, and has served on advisory boards arranged by AstraZeneca, Novartis, Meda, GlaxoSmithKline, Teva and Boehringer Ingelheim. HS has received honoraria for educational activities from Boehringer Ingelheim, Novartis, AstraZeneca, Chiesi, and TEVA, an unrestricted research grant from AstraZeneca, and has served on advisory boards arranged by AstraZeneca, Novartis, Chiesi, OrionPharma, and GlaxoSmithKline. The authors report no other conflicts of interest in this work.

References

1. Chetty U, McLean G, Morrison D, Agur K, Guthrie B, Mercer SW. Chronic obstructive pulmonary disease and comorbidities: a large cross-sectional study in primary care. Br J Gen Pract. 2017;67(658):e321–e328. doi:10.3399/bjgp17X690605

2. Burke H, Wilkinson TMA. Unravelling the mechanisms driving multimorbidity in COPD to develop holistic approaches to patient-centred care. Eur Respir Rev. 30. 160. doi:10.1183/16000617.0041-2021

3. Smith MC, Wrobel JP. Epidemiology and clinical impact of major comorbidities in patients with COPD. Int J Chron Obstruct Pulmon Dis. 2014;9:871–888. doi:10.2147/COPD.S49621

4. Divo M, Cote C, de Torres JP, et al. Comorbidities and risk of mortality in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2012;186(2):155–161. doi:10.1164/rccm.201201-0034OC.

5. Stallberg B, Janson C, Larsson K, et al. Real-world retrospective cohort study arctic shows burden of comorbidities in Swedish COPD versus non-COPD patients. NPJ Prim Care Respir Med 2018;28(1):33. doi:10.1038/s41533-018-0101-y.

6. Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11(1):450. doi:10.1186/1471-2458-11-450.

7. Jones RC, Price D, Ryan D, et al. Opportunities to diagnose chronic obstructive pulmonary disease in routine care in the UK: a retrospective study of a clinical cohort. Lancet Respir Med. 2014;2(4):267–276. doi:10.1016/s2213-2600(14)70008-6

8. Ställberg B, Janson C, Johansson G, et al. Management, morbidity and mortality of COPD during an 11-year period: an observational retrospective epidemiological register study in Sweden (PATHOS). Prim Care Respir J. 2014;23(1):38–45. doi:10.4104/pcrj.2013.00106

9. Barlow L, Westergren K, Holmberg L, Talbäck M. The completeness of the Swedish Cancer Register – a sample survey for year 1998. Acta Oncologica. 2009;48(1):27–33. doi:10.1080/02841860802247664

10. Brooke HL, Talbäck M, Hörnblad J, et al. The Swedish cause of death register. Eur J Epidemiol. 2017;32(9):765–773. doi:10.1007/s10654-017-0316-1

11. Ludvigsson JF, Svedberg P, Olén O, Bruze G, Neovius M. The longitudinal integrated database for health insurance and labour market studies (LISA) and its use in medical research. Eur J Epidemiol Apr. 2019;34(4):423–437. doi:10.1007/s10654-019-00511-8

12. Yanbaeva DG, Dentener MA, Creutzberg EC, Wesseling G, Wouters EF. Systemic effects of smoking. Chest. 2007;131(5):1557–1566. doi:10.1378/chest.06-2179

13. U.S. Department of Health and Human Services. The health consequences of smoking: 50 years of progress: a report of the surgeon general. Atlanta, GA: U.S. Department of Health and Human Services, Centres for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. 2014.

14. Thomsen M, Nordestgaard BG, Vestbo J, Lange P. Characteristics and outcomes of chronic obstructive pulmonary disease in never smokers in Denmark: a prospective population study. Lancet Respir Med. 2013;1(7):543–550. doi:10.1016/s2213-2600(13)70137-1

15. Lindberg A, Larsson LG, Muellerova H, Rönmark E, Lundbäck B. Up-to-date on mortality in COPD - report from the OLIN COPD study. BMC Pulm Med. 2012;12(1):1. doi:10.1186/1471-2466-12-1

16. Divo MJ, Celli BR, Poblador-Plou B, et al. Chronic Obstructive Pulmonary Disease (COPD) as a disease of early aging: evidence from the epichron cohort. PLoS One. 2018;13(2):e0193143. doi:10.1371/journal.pone.0193143

17. Chen W, Thomas J, Sadatsafavi M, FitzGerald JM. Risk of cardiovascular comorbidity in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis. Lancet Respir Med. 2015;3(8):631–639. doi:10.1016/S2213-2600(15)00241-6

18. Wang M, Lin EP-Y, Huang L-C, Li C-Y, Shyr Y, Lai C-H. Mortality of Cardiovascular Events in Patients With COPD and preceding hospitalization for acute exacerbation. Chest. 2020;158(3):973–985. doi:10.1016/j.chest.2020.02.046

19. Inoue D, Watanabe R, Okazaki R. COPD and osteoporosis: links, risks, and treatment challenges. Int J Chron Obstruct Pulmon Dis. 2016;11:637–648. doi:10.2147/copd.S79638

20. Ryan CS, Petkov Adler RA, Adler RA. Osteoporosis in men: the value of laboratory testing. Osteoporos Int. 2011;22(6):1845–1853. doi:10.1007/s00198-010-1421-0

21. Smith PH, Mazure CM, McKee SA. Smoking and mental illness in the U.S. population. Tob Control. 2014;23(e2):e147–53. doi:10.1136/tobaccocontrol-2013-051466

22. Yohannes AM, Kaplan A, Hanania NA. Anxiety and depression in chronic obstructive pulmonary disease: recognition and management. Cleve Clin J Med. 2018;85(2 Suppl 1):S11–s18. doi:10.3949/ccjm.85.s1.03

23. Larsson SC, Burgess S. Appraising the causal role of smoking in multiple diseases: a systematic review and meta-analysis of Mendelian randomization studies. EBioMedicine. 2022;82:104154. doi:10.1016/j.ebiom.2022.104154

24. Dodd JW. Lung disease as a determinant of cognitive decline and dementia. Alzheimers Res Ther. 2015;7(1):32. doi:10.1186/s13195-015-0116-3

25. Higbee D, Granell R, Walton E, Korologou-Linden R, Davey Smith G, Dodd J. Examining the possible causal relationship between lung function, COPD and Alzheimer’s disease: a Mendelian randomisation study. BMJ Open Respir Res. 2021;8(1):1.

26. Kornum JB, Sværke C, Thomsen RW, Lange P, Sørensen HT. Chronic obstructive pulmonary disease and cancer risk: A Danish nationwide cohort study. Respir Med. 2012;106(6):845–852. doi:10.1016/j.rmed.2011.12.009

27. Raglan O, Kalliala I, Markozannes G, et al. Risk factors for endometrial cancer: An umbrella review of the literature. Int J Cancer. 2019;145(7):1719–1730. doi:10.1002/ijc.31961.

28. Arafa A, Mostafa A, Navarini AA, Dong J-Y. The association between smoking and risk of skin cancer: a meta-analysis of cohort studies. Cancer Causes Control. 2020;31(8):787–794. doi:10.1007/s10552-020-01319-8

29. Skajaa N, Laugesen K, Horváth-Puhó E, Sørensen HT. Comorbidities and mortality among patients with chronic obstructive pulmonary disease. BMJ Open Respir Res. 2023;10(1):1.

30. Greulich T, Weist BJD, Koczulla AR, et al. Prevalence of comorbidities in COPD patients by disease severity in a German population. Respir Med. 2017;132:132–138. doi:10.1016/j.rmed.2017.10.007

31. Stolz D, Mkorombindo T, Schumann DM, et al. Towards the elimination of chronic obstructive pulmonary disease: a lancet commission. Lancet. 2022;400(10356):921–972. doi:10.1016/s0140-6736(22)01273-9.

32. Athlin Å, Lisspers K, Hasselgren M, et al. Diagnostic spirometry in COPD is increasing, a comparison of two Swedish cohorts. NPJ Prim Care Respir Med. 2023;33(1):23. doi:10.1038/s41533-023-00345-8.

33. Inghammar M, Engström G, Löfdahl CG, Egesten A. Validation of a COPD diagnosis from the Swedish Inpatient Registry. Scand J Public Health. 2012;40(8):773–776. doi:10.1177/1403494812463172

34. Spero K, Bayasi G, Beaudry L, Barber KR, Khorfan F. Overdiagnosis of COPD in hospitalized patients. Int J Chron Obstruct Pulmon Dis. 2017;12:2417–2423. doi:10.2147/copd.S139919

35. Ekberg-Aronsson M, Löfdahl K, Nilsson JA, Löfdahl CG, Nilsson PM. Hospital admission rates among men and women with symptoms of chronic bronchitis and airflow limitation corresponding to the GOLD stages of chronic obstructive pulmonary disease--A population-based study. Respir Med. 2008;102(1):109–120. doi:10.1016/j.rmed.2007.07.028

36. Serdén L, Lindqvist R, Rosén M. Have DRG-based prospective payment systems influenced the number of secondary diagnoses in health care administrative data? Health Policy. 2003;65(2):101–107. doi:doi:10.1016/S0168-8510(02)00208-7