Search

Henrik Toft Sørensen,1 Tina Christensen,1 Hans Erik Bøtker,2 Christian Fynbo Christiansen,1 Cecilia H Fuglsang,1 Sigrid B Gribsholt,1 Frederik Pagh Bredahl Kristensen,1 Kristina Laugesen,1 Anne Sofie D Laursen,1 Mette Nørgaard,1 Morten Schmidt,1 Nils Skajaa,1 Frederikke S Troelsen,1 Lars Pedersen1 1Department of Clinical Epidemiology, Aarhus University Hospital and Aarhus University, Aarhus, Denmark; 2Department of Clinical Medicine, Aarhus University, Aarhus, DenmarkCorrespondence: Henrik Toft Sørensen, Department of Clinical Epidemiology, Aarhus University Hospital and Aarhus University, Olof Palmes Allé 43-45, Aarhus, 8200, Denmark, Tel +45 87 16 82 15, Email hts@clin.au.dkPurpose: Humans are living longer and may develop multiple chronic diseases in later life. The Better Health in Late Life cohort study aims to improve our understanding of the risks and outcomes of multimorbidity in the Danish population.Methods: A randomly-selected sample of Danish residents who were 50– 65 years of age received a questionnaire and an invitation to participate in this study. Respondents completed an online survey between October 2021 and January 2022 which addressed topics that included self-assessed health, mental health, sleep, specific medical conditions, use of painkillers, diet, alcohol consumption, smoking, physical activity, and body composition. This information was linked to the Danish health and social registries (some established in 1943 and onwards) that maintain data on filled prescriptions, hospital records, socioeconomic status, and health care utilization.Results: Responses were received from 115,431 of the 301,244 residents invited to participate (38%). We excluded respondents who answered none of the questions as well as those who provided no information on sex or indicated an age other than 50– 65 years. Of the 114,283 eligible respondents, 54.8% were female, 30.3% were overweight, and 16.7% were obese. Most participants reported a weekly alcohol consumption of less than seven units and 13.3% were current smokers; 5.2% had a history of hospitalization for solid cancer, and 3.0%, 2.3%, 2.0%, and 0.9% reported chronic pulmonary disease, diabetes, stroke, and myocardial infarction, respectively. The most frequently filled prescriptions were for medications used to treat the nervous system and cardiovascular diseases (38.1% and 37.4%, respectively).Keywords: aging, epidemiology, health registries, life course epidemiology, multimorbidity, prospective cohort

Frederik Cosedis Enevoldsen,1 Christian Fynbo Christiansen,1,2 Simon Kok Jensen1,2 1Department of Clinical Medicine, Aarhus University, Aarhus, Denmark; 2Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, DenmarkCorrespondence: Frederik Cosedis Enevoldsen, Department of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes Allé 43-45, 8200 Aarhus N, Central Denmark Region, Denmark, Tel +45 87 16 72 12, Email freene@rm.dkBackground: The occurrence of acute and chronic kidney diseases has been rising in the last decades. Although drug use is a common risk factor for impaired kidney function, changes in utilization of potential nephrotoxic drugs have received little attention.Purpose: To describe temporal trends in the utilization of potentially nephrotoxic drugs in Denmark between 1999 and 2021.Methods: Specific drugs known or suspected to be nephrotoxic were identified in the literature. Data on the sold defined daily doses (DDDs) of potentially nephrotoxic drugs between 1999 and 2021 were retrieved using the Danish Register of Medical Product Statistics. Trends in sales of DDDs per 1000 inhabitants per day were tabulated and illustrated graphically.Results: From 1999 to 2021, the total sale of all selected drugs increased from 286 to 457 DDDs per 1000 inhabitants per day. The overall sale reached a preliminary peak in 2012 with 449 DDDs per 1000 inhabitants per day and remained relatively stable thereafter until reaching an all-time high in 2021 with 457 DDDs per 1000 inhabitants per day. Contributing with the majority in volume, sales of drugs inhibiting the renin-angiotensin-aldosterone system (RAAS) increased dramatically throughout the period. The same was observed for acetaminophen, methotrexate, tacrolimus, and iodinated contrast dye. In contrast, the sales of diuretics, acetylsalicylic acid, and ciclosporin decreased during the last decade of the study period.Conclusion: From 1999– 2021 considerable changes in sales of potentially nephrotoxic drugs were observed. In general, the sales increased, in volume predominated by RAAS inhibiting drugs. This increase in sales of potential nephrotoxins could contribute to an increasing occurrence of kidney diseases.Keywords: adverse events, nephrotoxins, drug utilization, acute kidney injury, kidney disease

Simon Kok Jensen,1,2 Uffe Heide-Jørgensen,1,2 Søren Viborg Vestergaard,1,2 Henrik Toft Sørensen,1,2 Christian Fynbo Christiansen1,2 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Department of Clinical Medicine, Aarhus University, Aarhus, DenmarkCorrespondence: Simon Kok Jensen, Department of Clinical Epidemiology, Department of Clinical Medicine, Aarhus University and Aarhus University Hospital, Olof Palmes Allé 43-45, Aarhus, 8200, Denmark, Tel +45 8716 8481, Email skj@clin.au.dkPurpose: Creatinine data are key in identifying acute and chronic kidney disease. In Denmark, routine clinical care creatinine data have been collected regionally in the Clinical Laboratory Information System Research Database (LABKA) since the 1990s and nationwide in the Register of Laboratory Results for Research (RLRR) since 2013. Here we describe the geographical coverage of the databases and characteristics of Danish individuals with creatinine tests. This information is pivotal for the design and interpretation of studies using these data to examine kidney disease epidemiology.Patients and Methods: We included all creatinine tests in LABKA and RLRR from 1990 through 2018. The daily number of creatinine tests by municipality and region of residence were plotted and geographical coverage was ascertained. In addition, we characterized a contemporary cohort of creatinine-tested individuals in 2016– 2018.Results: During 1990– 2018, 61,011,941 creatinine tests were available for 4,647,966 unique Danish residents. The North Denmark Region was the first region to achieve complete reporting in November 2004, and nationwide reporting was complete starting in October 2015. In each year from 2016 to 2018, more than a third of Danish residents had a recorded creatinine measurement, with the highest proportion of tested individuals aged 77– 87 years and the lowest proportion aged 3– 5 years. During 2016– 2018, the creatinine-tested cohort had a median age of 53 years (IQR, 35– 67 years) and included 54.3% women. The most common comorbidity was a hospital-based diagnosis of hypertension (12.0%), and the most common prescription drug was angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers (15.8%).Conclusion: In combination, the population-based LABKA and RLRR databases provide regional creatinine data with long follow-up and nationwide data for the Danish population. There was considerable variation in the time of complete geographical coverage by region, which needs to be considered when using these data for studies on kidney disease epidemiology.Keywords: cohort, creatinine, epidemiology, glomerular filtration rate, kidney disease, registries

Anne Høy Seemann Vestergaard,1 Christian Fynbo Christiansen,1 Mette Asbjoern Neergaard,2 Jan Brink Valentin,3 Søren Paaske Johnsen3 1Department of Clinical Epidemiology, Department of Clinical Medicine, Aarhus University and Aarhus University Hospital, Aarhus N, 8200, Denmark; 2Palliative Care Unit & Child and Youth Palliative Care Team, Oncology Department, Aarhus University Hospital, Aarhus N, 8200, Denmark; 3Danish Center for Clinical Health Services Research, Department of Clinical Medicine, Aalborg University and Aalborg University Hospital, Aalborg Ø, 9220, DenmarkCorrespondence: Anne Høy Seemann Vestergaard, Department of Clinical Epidemiology, Department of Clinical Medicine, Aarhus University and Aarhus University Hospital, Olof Palmes Allé 43-45, Aarhus N, 8200, Denmark, Tel +45 87168434, Fax +45 87167215, Email ahsv@clin.au.dkPurpose: Socioeconomic disparities in end-of-life care have been reported across underlying diseases, but there is a paucity of information on potential time trends. Thus, we aimed to examine time trends in use of health-care services at the end of life according to socioeconomic position in patients dying from cancer and non-cancer diseases.Materials and Methods: We conducted a nationwide registry-based study among adults dying from cancer or non-cancer diseases (diabetes, dementia, heart failure, ischemic heart disease, stroke, chronic liver disease, and chronic obstructive pulmonary disease) in Denmark in 2006– 2016. We obtained data on patients’ educational level and income level and use of health-care services within three months before death. Use of health-care services according to educational level and income level was plotted by calendar year of death and compared by regression analyses adjusting for age, sex, comorbidity, cohabitation, and municipality.Results: In both cancer (n = 169,694) and non-cancer patients (n = 180,350), we found limited socioeconomic disparities and no clear temporal trends in use of hospital, intensive care, emergency room, general practice, home care nurse, and hospice. In 2006/2007, one percentage point more cancer patients with high income level compared with low income level were affiliated with hospital-based specialist palliative care (adjusted mean difference: 0.01 (95% confidence interval (CI): 0.01; 0.02)), whereas this was 12 percentage points in 2016 (adjusted mean difference: 0.12 (95% CI: 0.09; 0.14)).Conclusion: Socioeconomic disparities in specialist palliative care tended to increase over time among cancer patients but were limited and without clear time trends in use of other health-care services in both cancer and non-cancer patients.Keywords: chronic disease, death, hospitals, neoplasms, palliative care, socioeconomic factors

Søren Korsgaard,1,2 Christian Fynbo Christiansen,1,2 Morten Schmidt,1– 3 Henrik Toft Sørensen1,2 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Department of Clinical Medicine, Aarhus University, Aarhus, Denmark; 3Department of Cardiology, Aarhus University Hospital, Aarhus, DenmarkCorrespondence: Søren KorsgaardDepartment of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes Allé 43-45, Aarhus N, Aarhus, 8200, DenmarkTel +45 87167212Fax +45 87167215Email soren@clin.au.dkPurpose: Studies examining myocardial infarction (MI) often seek to include only incident MIs by excluding recurrent MIs. When based on historical data, identification of previous MI depends on the length of the look-back period. However, international registries often cover a short time period, consequently containing left-censored data, making it impossible to determine if a first MI in a period is truly an incident MI. We evaluated whether the proportion of MIs identified as recurrent MIs depends on the look-back period, and how including recurrent MIs in a planned incident MI cohort impacts survival estimates.Patients and Methods: We used the Danish National Patient Registry, covering all Danish hospitals since 1977 to identify first MIs during 2010– 2016 (index events). The hospital registry history preceding the index event was then searched for previous MIs. We plotted the proportion of index events identified as recurrent MIs as a function of the look-back period. Moreover, we calculated 5-year all-cause mortality and confidence intervals (CIs) using the 1-Kaplan–Meier method for five cohorts based on the index events and defined by look-back periods of 0, 5, 10, 20, and up to 39 years.Results: Among 63,885 index events, 3.4% were identified as recurrent MIs with 5 years of look-back, 7.9% with 10 years, 14% with 24 years, and 15% with up to 39 years. All-cause mortality risk was 36% (95% CI: 36– 37%) with 0 years of look-back, 35% (95% CI: 35– 36%) with 5 years, 35% (95% CI: 35– 36%) with 10 years, 34% (95% CI: 34– 35%) with 20 years, and 34% (95% CI: 33– 34%) with up to 39 years.Conclusion: Most recurrent MIs were identified with a look-back period of 24 years. Including recurrent MIs in a planned incident MI cohort, due to shorter look-back periods, overestimated the mortality risk.Keywords: acute myocardial infarction, bias, cardiology, cohort, epidemiology, methodology

Søren Viborg Vestergaard,1 Uffe Heide-Jørgensen,1 Heleen van Haalen,2 Glen James,3 Katarina Hedman,4 Henrik Birn,5,6 Reimar Wernich Thomsen,1 Christian Fynbo Christiansen1 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Health Economics & Payer Evidence, AstraZeneca, Gothenburg, Sweden; 3Epidemiology, AstraZeneca, Cambridge, UK; 4Biometrics, AstraZeneca, Gothenburg, Sweden; 5Department of Renal Medicine, Aarhus University Hospital, Aarhus, Denmark; 6Department of Biomedicine, Aarhus University, Aarhus, DenmarkCorrespondence: Søren Viborg VestergaardDepartment of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes Allé 43-45, Aarhus N 8200, DenmarkTel +45 87168514Email sovi@clin.au.dkPurpose: Anemia is prevalent in patients with chronic kidney disease (CKD), but the longitudinal risk of anemia in patients with newly identified CKD is unknown. We therefore examined the risks of experiencing anemia in persons with newly identified CKD.Patients and Methods: This cohort study included adult patients with newly identified CKD stages 3– 5 defined by an estimated glomerular filtration rate (eGFR) level < 60 mL/min/1.73m2 (at least two measurements ≥ 90 days apart) ascertained from a population-based registry with complete laboratory test results in Northern Denmark (population ∼ 2.2 million) during 2009– 2016. We calculated 1) cumulative incidence (risk) of anemia [hemoglobin < 12/< 13 g/dl in women/men] by CKD stage, and 2) adjusted hazard ratios (HRs) of anemia using Cox regression analyses.Results: We identified 55,940 distinct individuals with newly identified CKD stages 3– 5 and no prevalent anemia [n=41,958 patients in stage 3a, n=17,875 in stage 3b, n=5182 in stage 4, and n=931 in stage 5]. After one year, 42.3% (95%-confidence interval [CI]: 41.9– 42.7) of patients with CKD stages 3– 5 had newly measured anemia, increasing to 67.7% (95%-CI: 67.2– 68.2) after five years. The absolute and relative anemia risk increased markedly with higher CKD stages. The adjusted HR of any anemia was 5.42 (95%-CI: 5.09– 5.77) in patients with CKD stage 5 compared to patients with CKD stage 3a.Conclusion: Patients with newly identified CKD stages 3– 5 have a substantial risk of anemia, increasing with higher CKD stages. This study underlines that clinical awareness of anemia risk is important in patients with newly identified or progressed CKD.Keywords: anemia, chronic kidney disease, CKD, epidemiology, cohort study

Finn Erland Nielsen,1,2 Johan Joakim Cordtz,1 Thomas Bøjer Rasmussen,3 Christian Fynbo Christiansen3 1Department of Emergency Medicine, Slagelse Hospital, Slagelse DK-4200, Denmark; 2Department of Emergency Medicine, Bispebjerg and Frederiksberg Hospital, Copenhagen DK-2400 NV, Denmark; 3Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus N 8200, DenmarkCorrespondence: Finn Erland NielsenDepartment of Emergency Medicine, Bispebjerg and Frederiksberg Hospital, Ebba Lunds Vej 40A, Entrance 67, Copenhagen 2400 NV, DenmarkTel +45 26822753Fax +45 38639863Email finn.erland.nielsen@regionh.dkBackground: We examined the association between creatine phosphokinase level in rhabdomyolysis patients and risk of acute kidney injury, renal replacement therapy, and death within 30 days.Methods: The cohort included patients admitted with rhabdomyolysis from November 1, 2011 to March 1, 2014. Rhabdomyolysis was defined as a creatine phosphokinase level higher than 1000 U/L. Information on laboratory variables was obtained from a laboratory database. Medical data were obtained from registries. Acute kidney injury was defined according to the Kidney Disease Improving Global Outcome (KDIGO) guidelines. The 30-day risk of outcomes was estimated using the cumulative incidence method. Spline regression applied to imputed datasets with adjustment for baseline variables was used to assess the appropriateness of the categorization chosen for creatine phosphokinase (1000– 5000 U/L, 5001– 15,000 U/L, and 15,000+ U/L).Results: The study included 1027 patients (58.2% male) with a median age of 73.5 years. The median creatine phosphokinase level at rhabdomyolysis diagnosis was 2257 U/L (interquartile range=1404– 3961 U/L). The 30-day risks of acute kidney injury according to the three creatinine phosphokinase levels were 42% (95% CI=38– 45%), 44% (95% CI=36– 52%), and 74% (95% CI=57– 85%), respectively, and the risks of renal replacement therapy for the three levels were 3% (95% CI=2– 5%), 4% (95% CI=2– 7%), and 11% (3– 23%), respectively. The 30-day risk of death was 17% (95% CI=14– 20%), 16% (95% CI=11– 22%), and 11% (95% CI=3– 23%), respectively. With increasing creatine phosphokinase levels, the spline plots supported the increasing risk of acute kidney injury and renal replacement therapy, as well as a decreasing risk of death. However, the risk estimates for renal replacement therapy and death were imprecise.Conclusion: Elevated initial creatine phosphokinase values were associated with an increased risk of acute kidney injury, while estimates of the risk of renal replacement therapy and death were imprecise.Keywords: acute kidney injury, cohort study, mortality, renal replacement therapy, rhabdomyolysis, spline plots

Anton Pottegård,1 Kasper Bruun Kristensen,1 Mette Reilev,1 Lars Christian Lund,1 Martin Thomsen Ernst,1 Jesper Hallas,1,2 Reimar Wernich Thomsen,3 Christian Fynbo Christiansen,3 Henrik Toft Sørensen,3,4 Nanna Borup Johansen,5 Henrik Støvring,1,6 Steffen Christensen,7 Marianne Kragh Thomsen,8 Anders Husby,9 Marianne Voldstedlund,10 Jesper Kjær,11 Nikolai C Brun5 1Clinical Pharmacology and Pharmacy, Department of Public Health, University of Southern Denmark, Odense, Denmark; 2Department of Clinical Biochemistry and Clinical Pharmacology, Odense University Hospital, Odense, Denmark; 3Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 4Department of Epidemiology and Population Health, Stanford University, Stanford, CA, USA; 5Department of Medical Evaluation and Biostatistics, Danish Medicines Agency, Copenhagen, Denmark; 6Department of Public Health – Biostatistics, Aarhus University, Aarhus, Denmark; 7Department of Anesthesia and Intensive Care Medicine, Aarhus University Hospital, Aarhus, Denmark; 8Department of Clinical Microbiology, Aarhus University Hospital, Aarhus, Denmark; 9Department of Epidemiology Research, Statens Serum Institut, Copenhagen, Denmark; 10Infection Preparedness, Statens Serum Institut, Copenhagen, Denmark; 11Data Analytics Center, Danish Medicines Agency, Copenhagen, DenmarkCorrespondence: Anton PottegårdClinical Pharmacology and Pharmacy, Department of Public Health, University of Southern Denmark, JB Winsløws Vej 19, 2, Odense DK-5000, DenmarkTel +45 28913340Email apottegaard@health.sdu.dkBackground: To facilitate research on severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a prospective cohort of all Danish residents tested for SARS-CoV-2 in Denmark is established.Data Structure: All Danish residents tested by reverse transcriptase polymerase chain reactions (RT-PCR) for SARS-CoV-2 in Denmark are included. The cohort is identified using the Danish Microbiology Database. Individual-level record linkage between administrative and health-care registries is facilitated by the Danish Civil Registration System. Information on outcomes related to SARS-CoV-2 infection includes hospital admission, intensive care unit admission, mechanical ventilation, and death and is retrieved from the five administrative Danish regions, the Danish National Patient Registry, and the Danish Register of Causes of Death. The Patient Registry further provides a complete hospital contact history of somatic and psychiatric conditions and procedures. Data on all prescriptions filled at community pharmacies are available from the Danish National Prescription Registry. Health-care authorization status is obtained from the Danish Register of Healthcare Professionals. Finally, selected laboratory values are obtained from the Register of Laboratory Results for Research. The cohort is governed by a steering committee with representatives from the Danish Medicines Agency, Statens Serum Institut, the Danish Health Authority, the Danish Health Data Authority, Danish Patients, the Faculties of Health Sciences at the Danish universities, and Danish regions. The steering committee welcomes suggestions for research studies and collaborations. Research proposals will be prioritized based on timeliness and potential clinical and public health implications. All research protocols assessing specific hypotheses for medicines will be made publicly available using the European Union electronic Register of Post-Authorisation Studies.Conclusion: The Danish COVID-19 cohort includes all Danish residents with an RT-PCR test for SARS-CoV-2. Through individual-level linkage with existing Danish health and administrative registries, this is a valuable data source for epidemiological research on SARS-CoV-2.Keywords: Covid-19, SARS-CoV-2, epidemiology, follow-up, database, prognosis, prospective cohort

Guillaume Moulis, Christian Fynbo Christiansen, Bianka Darvalics, Ina Trolle Andersen, Mette Nørgaard Department of Clinical Epidemiology, Institute of Clinical Medicine, Aarhus University Hospital, Aarhus, DenmarkCorrespondence: Guillaume MoulisDepartment of Clinical Epidemiology, Institute of Clinical Medicine, Aarhus University Hospital, Olof Palmes Allé 43-45, Aarhus N 8200, DenmarkTel +45 871 68205Fax +45 871 67215Email gmoulis@hotmail.comBackground: Apart from their hemostatic role, platelets are immune cells that play a role in fighting infections. The presence of thrombocytopenia and thrombocytosis at hospital admission are predictors of mortality in community-acquired pneumonia patients. We hypothesized that variations in platelet counts within the normal range also may be associated with mortality in these patients.Methods: The study included all adults in the North and Central Denmark Regions with a first acute hospital admission for community-acquired pneumonia during 2006– 2012. We assessed the association between the first platelet count within ± 24 hours of admission (within the normal range of 150 to 400 x 109/L) and 30-day mortality using Cox models. Analyses were adjusted for age, sex, Charlson Comorbidity Index score, hemoglobin level, leukocyte count, and creatinine level at admission.Results: Among the 12,905 study patients, 30-day mortality was 12.4%. The mean platelet count upon admission was 250 × 109/L. Compared with the 250– 275 × 109/L category, platelet counts of 151– 175 were associated with a lower 30-day mortality (adjusted hazard ratio [aHR]: 0.79, 95% confidence interval [CI]: 0.63– 0.99), while higher platelet counts were associated with a higher 30-day mortality (351– 375 × 109/L, aHR: 1.34, 95% CI: 1.07– 1.68; 376– 400× 109/L, aHR: 1.21, 95% CI: 0.94– 1.56).Conclusion: Platelet counts, even within the normal range, are associated with mortality in adult patients hospitalized for community-acquired pneumonia.Keywords: platelet count, thrombocytopenia, thrombocytosis, community-acquired pneumonia, mortality

Sissel Ravn,1 Christian F Christiansen,2 Rikke H Hagemann-Madsen,3 Victor J Verwaal,1 Lene H Iversen1,4 1Department of Surgery, Aarhus University Hospital, Aarhus, Denmark; 2Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 3Department of Pathology, Lillebaelt Hospital, Vejle, Denmark; 4Danish Colorectal Cancer Group (DCCG), Copenhagen, DenmarkCorrespondence: Sissel RavnDepartment of Surgery, Aarhus University Hospital, Aarhus N DK-8200, DenmarkTel +45 51627366Email sissel.ravn@rm.dkIntroduction: Treatment options for peritoneal metastases (PM) from colorectal cancer (CRC) have increased, their efficiency should be monitored. For this purpose, register-based data on PM can be used, if valid.Purpose: We aimed to evaluate the completeness and positive predictive value (PPV) of synchronous peritoneal metastases (S-PM) registered among CRC patients in the Danish National Patient Register (DNPR) and/or the Danish National Pathology Register (the DNPatR) using the Danish Colorectal Cancer Group database (DCCG) as a reference.Patients and Methods: We identified Danish patients with newly diagnosed primary CRC in the DCCG during 2014– 2015. S-PM were routinely registered in the DCCG. We excluded patients with non-CRC cancers and identified S-PM using all three registries. We estimated the completeness and the PPV of registered S-PM in the DNPR, the DNPatR and the DNPR and/or the DNPatR (DNPR/DNPatR) in combination using the DCCG as the reference. We stratified by age, gender, WHO performance status, tumour location and distant metastases to liver and/or lungs.Results: We identified 9142 patients with CRC in DCCG. In DCCG, 366 patients were registered with S-PM, among whom 213 in DCCG only, whereas 153 in DCCG and in at least one of DNPR and/or DNPatR. In DNPR/DNPatR, S-PM was registered with a completeness of 42% [95% CI: 37– 47] and a PPV of 60% [95% CI: 54– 66]. In the DNPR only, the completeness was 32% [95% CI: 27– 37] and the PPV 57% [95% CI: 50– 64]. The completeness in the DNPatR was 19% [95% CI: 15– 23] and the PPV was 76% [95% CI: 68– 85]. In the DNPR/DNPatR patients aged < 60 years (57% [95% CI: 46– 69]), patients with WHO performance status 0 (46% [95% CI: 37– 54]) and patients with no distant metastases (58% [95% CI: 50– 65]) were registered with a higher completeness.Conclusion: Our algorithm demonstrates that the DNPR/DNPatR captures less than half of CRC patients with S-PM. Potential candidates for curative treatment options are registered with a higher completeness. Clinicians should be encouraged to register the presence of S-PM to increase the validity of register-based S-PM data.Keywords: validity, synchronous peritoneal metastases, registries, colorectal cancer, epidemiology, completeness

Henrik Frederiksen,1–3 Szimonetta Szépligeti,1 Marie Bak,4 Waleed Ghanima,5,6 Hans Carl Hasselbalch,4 Christian Fynbo Christiansen1 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Department of Haematology, Odense University Hospital, Odense, Denmark; 3Department of Clinical Research, University of Southern Denmark, Odense, Denmark; 4Department of Haematology, Zealand University Hospital, Roskilde, Denmark; 5Departments of Oncology, Medicine and Research, Østfold Hospital Trust, Kalnes, Norway; 6Department of Haematology, Institute of Clinical Medicine, University of Oslo, Oslo, NorwayCorrespondence: Henrik FrederiksenDepartment of Haematology, Odense University Hospital, Kloevervaenget 6, Entrance 93, 12th Floor, Odense C DK-5000, DenmarkEmail henrik.frederiksen@rsyd.dkBackground: Patients with chronic myeloproliferative neoplasms (MPNs), including essential thrombocythemia (ET), polycythemia vera (PV), and primary myelofibrosis (PMF), are at high risk of vascular complications. However, the magnitude of this is risk not well known and the possible effect of comorbidity is poorly understood.Aim: Our aim was to compare the risk of vascular diseases in patients with MPNs and matched comparisons from the general population and to study the effect modification of comorbidity.Methods: We followed 3087 patients with ET, 6076 with PV, 3719 with PMF or unspecified MPN, and age- and sex-matched general population comparisons to estimate the risks of cardiovascular diseases such as myocardial infarction and stroke. We computed 5-year cumulative incidences (risks) for vascular disease in patients with MPNs and comparisons as well as 1-year and 5-year risks, risk differences, and hazard ratios (HRs) for vascular diseases comparing rates in each group of patients with their comparison cohort by level of comorbidity based on the Charlson Comorbidity Index (CCI) [score of 0 (low comorbidity), of 1–2 (moderate comorbidity), and of >2 (severe comorbidity)], as well as other comorbid conditions.Results: The overall 5-year risk of vascular disease ranged from 0.5% to 7.7% in patients with MPNs, which was higher than the risk in the general population. In the same period, the adjusted HRs for vascular disease were 1.3 to 3.7 folds higher in patients with MPNs compared to the general population. An increase in CCI score was associated with an equally increased rate of most types of vascular diseases during the first 5 years of follow-up in both MPN and comparisons.Conclusion: Patients with MPNs have a higher risk of vascular diseases during the first 5 years than that of the general population; however, comorbidity modifies the rates similarly in MPN and in the general population.Keywords: myeloproliferative neoplasms, thrombosis, comorbidity, stroke, epidemiology

Astrid Blicher Schelde,1 Morten Schmidt,1,2 Morten Madsen,1 Søren Steen Nielsen,3 Jørgen Frøkiær,4,5 Christian Fynbo Christiansen1 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Department of Cardiology, Regional Hospital West Jutland, Herning, Denmark; 3Department of Clinical Physiology and Nuclear Medicine, Aalborg University Hospital, Aalborg, Denmark; 4Department of Nuclear Medicine, Aarhus University Hospital, Skejby, Denmark; 5Department of Clinical Medicine, Aarhus University, Aarhus, DenmarkCorrespondence: Astrid Blicher ScheldeDepartment of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes Allé 43-45, 8200 Aarhus N, DenmarkTel +45 8716 7212Email astridblicher@hotmail.comBackground: Comorbidity is common among patients with myocardial infarction (MI). We examined whether comorbidity level modified the single-photon emission computed tomography myocardial perfusion imaging (SPECT MPI)-based prediction of 5-year risk of MI and all-cause death in patients with MI.Methods: This cohort study included patients with prior MI having a SPECT MPI at Aarhus University Hospital, Denmark, 1999–2011. Using nationwide registries, we obtained information on comorbidity levels (low, moderate, and severe) and outcomes. We computed risk and hazard ratios (HRs) with 95% confidence intervals (CIs) for MI and all-cause death, comparing normal (no defects) versus abnormal scan (reversible and/or fixed defects) using Cox regression adjusting for sex, age, and comorbidity level.Results: We identified 1,192 patients with MI before SPECT MPI. The 5-year risk for patients with normal versus abnormal scans were 11.7% versus 18.3% for MI, and 8.0% versus 13.2% for all-cause death, respectively. The overall 5-year adjusted HR (aHR) of MI was 1.56 (95% CI: 1.09–2.21), 1.33 (95% CI: 0.82–2.15) with low comorbidity, 1.39 (95% CI: 0.68–2.83) with moderate comorbidity, and 2.53 (95% CI: 1.14–5.62) with severe comorbidity. Similarly, the 5-year aHR for all-cause death was 1.39 (95% CI: 0.90–2.14) overall; 2.33 (95% CI: 0.79–6.84) with low comorbidity, 2.05 (95% CI: 0.69–6.06) with moderate comorbidity, and 1.07 (95% CI: 0.64–1.80) with severe comorbidity.Conclusion: We conclude that comorbidity level may modify the 5-year risk prediction associated with an abnormal SPECT MPI scan in patients with previous MI.Keywords: comorbidity, epidemiology, myocardial infarction, myocardial perfusion imaging

Erika Frischknecht Christensen,1–3 Peter Anthony Berlac,4 Henrik Nielsen,5 Christian Fynbo Christiansen5 1Clinical Institute, Aalborg University, 2Department of Anesthesiology and Intensive Care, Aalborg University Hospital, Aalborg, 3Prehospital Emergency Medical Services, North Denmark Region, 4Copenhagen Emergency Medical Services, Copenhagen, 5Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark Aim of database: The aim of the Danish quality database for prehospital emergency medical services (QEMS) is to assess, monitor, and improve the quality of prehospital emergency medical service care in the entire prehospital patient pathway. The aim of this review is to describe the design and the implementation of QEMS. Study population: The study population consists of all “112 patient contacts” defined as emergency patients, where the entrance to health care is a 112 call forwarded to one of the five regional emergency medical coordination centers in Denmark since January 1, 2014. Estimated annual number of included “112 patients” is 300,000–350,000. Main variables: We defined nine quality indicators and the following variables: time stamps for emergency calls received at one of the five regional emergency medical coordination centers, dispatch of prehospital unit(s), arrival of first prehospital unit, arrival of first supplemental prehospital unit, and mission completion. Finally, professional level and type of the prehospital resource dispatched to an incident and end-of-mission status (mission completed by phone, on scene, or admission to hospital) are registered. Descriptive data: Descriptive data included age, region, and Danish Index for Emergency Care including urgency level. Conclusion: QEMS is a new database under establishment and is expected to provide the basis for quality improvement in the prehospital setting and in the entire patient care pathway, for example, by providing prehospital data for research and other quality databases. Keywords: emergency medical services, ambulance response times, emergency medical dispatch, quality improvement

Annmarie T Lassen,1 Henrik Jørgensen,2 Hanne Blæhr Jørsboe,3,4 Annette Odby,5 Mikkel Brabrand,6 Jacob Steinmetz,7 Julie Mackenhauer,8 Hans Kirkegaard,8 Christian Fynbo Christiansen9 1Department of Emergency Medicine, Odense University Hospital, Odense, 2Department of Surgery, Hospital of Northern Sjaelland, Hilleroed, 3Department of Emergency Medicine, Nykobing F Hospital, 4Department of Hospital Administration, Nykobing F Hospital, Nykøbing Falster, 5The Danish Clinical Registers, Registry Support Centre for Health Quality and Informatics, Aarhus, 6Department of Emergency Medicine, Hospital of South West Jutland, Esbjerg, 7Department of Anaesthesia, Rigshospitalet, Copenhagen University Hospital, Copenhagen, 8Research Center for Emergency Medicine, Aarhus University, 9Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark Aim for database: Aim of the Danish database for acute and emergency hospital contacts (DDAEHC) is to monitor the quality of care for all unplanned hospital contacts in Denmark (acute and emergency contacts).Study population: The DDAEHC is a nationwide registry that completely covers all acute and emergency somatic hospital visits at individual level regardless of presentation site, presenting complaint, and department designation since January 1, 2013.Main variables: The DDAEHC includes ten quality indicators – of which two are outcome indicators and eight are process indicators. Variables used to compute these indicators include among others day and time of hospital contact, vital status, ST-elevation myocardial infarction diagnosis, date and time of relevant procedure (percutaneous coronary intervention, coronary angiography, X-ray of wrist, and gastrointestinal surgery) as well as time for triage and physician judgment. Data are currently gathered from The Danish National Patient Registry, two existing databases (Danish Stroke Register and Danish Database for Emergency Surgery), and will eventually include data from the local and regional clinical logistic systems.Descriptive data: The DDAEHC also includes age, sex, Charlson Comorbidity Index conditions, civil status, residency, and discharge diagnoses. The DDAEHC expects to include 1.7 million acute and emergency contacts per year.Conclusion: The DDAEHC is a new database established by the Danish Regions including all acute and emergency hospital contacts in Denmark. The database includes specific outcome and process health care quality indicators as well as demographic and other basic information with the purpose to be used for enhancement of quality of acute care. Keywords: emergency care, emergency department, mortality, rehospitalization, quality of care

Christian Fynbo Christiansen,1 Morten Hylander Møller,2 Henrik Nielsen,1 Steffen Christensen3 1Department of Clinical Epidemiology, Institute of Clinical Medicine, Aarhus University Hospital, Aarhus, 2Department of Intensive Care 4131, Copenhagen University Hospital Rigshospitalet, Copenhagen, 3Department of Intensive Care, Aarhus University Hospital, Aarhus, Denmark Aim of database: The aim of this database is to improve the quality of care in Danish intensive care units (ICUs) by monitoring key domains of intensive care and to compare these with predefined standards. Study population: The Danish Intensive Care Database (DID) was established in 2007 and includes virtually all ICU admissions in Denmark since 2005. The DID obtains data from the Danish National Registry of Patients, with complete follow-up through the Danish Civil Registration System. Main variables: For each ICU admission, the DID includes data on the date and time of ICU admission, type of admission, organ supportive treatments, date and time of discharge, status at discharge, and mortality up to 90 days after admission. Descriptive variables include age, sex, Charlson comorbidity index score, and, since 2010, the Simplified Acute Physiology Score (SAPS) II. The variables are recorded with 90%–100% completeness in the recent years, except for SAPS II score, which is 73%–76% complete. The DID currently includes five quality indicators. Process indicators include out-of-hour discharge and transfer to other ICUs for capacity reasons. Outcome indicators include ICU readmission within 48 hours and standardized mortality ratios for death within 30 days after admission using case-mix adjustment (initially using age, sex, and comorbidity level, and, since 2013, using SAPS II) for all patients and for patients with septic shock. Descriptive data: The DID currently includes 335,564 ICU admissions during 2005–2015 (average 31,958 ICU admissions per year). Conclusion: The DID provides a valuable data source for quality monitoring and improvement, as well as for research. Keywords: critical care, databases, health care quality, intensive care

Diana Hedevang Christensen,1,2 Erzsébet Horváth-Puhó,1 Morten Schmidt,1 Christian Fynbo Christiansen,1 Lars Pedersen,1 Bente Lomholt Langdahl,2 Reimar Wernich Thomsen1 1Department of Clinical Epidemiology, 2Department of Endocrinology and Internal Medicine, Aarhus University Hospital, Aarhus, Denmark Purpose: Bisphosphonate use has been associated with increased risk of fatal stroke. We examined the association between preadmission use of oral bisphosphonates and 30-day mortality following hospitalization for stroke. Patients and methods: We conducted a nationwide population-based cohort study using medical databases and identified all patients in Denmark with a first-time hospitalization for stroke between 1 July 2004 and 31 December 2012 (N=100,043). Cox regression was used to compute adjusted hazard ratios as a measure of 30-day mortality rate ratios (MRRs) associated with bisphosphonate current use (prescription filled within 90 days prior to the stroke) or recent use (prescription filled in the 90–180 days prior to the stroke). Current use was further classified as new or long-term use. Results: We found 51,982 patients with acute ischemic stroke (AIS), 11,779 with intracerebral hemorrhage (ICH), 4,528 with subarachnoid hemorrhage (SAH), and 31,754 with unspecified stroke. Absolute 30-day mortality risks were increased among current vs nonusers of bisphosphonates for AIS (11.9% vs 8.5%), ICH (43.2% vs 34.5%), SAH (40.3% vs 23.2%), and unspecified strokes (18.8% vs 14.0%). However, in adjusted analyses, current bisphosphonate use did not increase 30-day mortality from AIS (MRR, 0.87; 95% confidence interval [CI]: 0.75, 1.01); ICH (MRR, 1.05; 95% CI: 0.90, 1.23); SAH (MRR, 1.15; 95% CI: 0.83, 1.61); or unspecified stroke (MRR, 0.94; 95% CI: 0.81, 1.09). Likewise, no association with mortality was found for recent use. Adjusted analyses by type of bisphosphonate showed increased mortality following stroke among new users of etidronate (MRR, 1.40; 95% CI: 1.01, 1.93) and reduced mortality after AIS among current users of alendronate (MRR, 0.87; 95% CI: 0.74, 1.02). Conclusion: We found no overall evidence that preadmission bisphosphonate use increases 30-day mortality following stroke. Keywords: prognosis, oral bisphosphonates, stroke, mortality, cardiovascular disease, osteoporosis treatment

Karoline Skov Dalgaard,1 Henrik Gammelager,1 Claus Sværke,1 Tamás Kurics,1 Karynsa Cetin,2 Christian Fynbo Christiansen1 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Center for Observational Research, Amgen Inc., Thousand Oaks, CA, USA Objective: Skeletal-related events (SREs) among patients with bone metastases from lung cancer may be associated with considerable use of health care resources. We analyzed in- and outpatient hospital contacts in relation to SREs among all Danish lung cancer patients with bone metastases. Methods: For this cohort study, we used the Danish Cancer Registry and the Danish National Registry of Patients to identify all persons diagnosed with first-time lung cancer and bone metastases from 2003 through 2009 in Denmark. We followed these patients until December 31, 2010, for the development of SREs (spinal cord compression; pathological or osteoporotic fracture, surgery to bone; or conventional external radiation therapy). We examined the number of inpatient hospitalizations, inpatient bed-days, hospital outpatient clinic visits, and emergency room visits within three time periods: a pre-SRE period (90-day period prior to the diagnostic period), a SRE diagnostic period (14-day period prior to the SRE), and a post-SRE period (90-day period after the SRE). Results: We identified 1,146 patients with lung cancer, bone metastases, and ≥1 subsequent SRE among 28,443 patients with incident lung cancer. Over 75% of patients with SREs (n=852) had more than one SRE. The number of hospital bed-days was high in the post-SRE period compared to the pre-SRE period, as illustrated by patients with multiple SREs who had 10.7 (95% confidence interval, 10.4–10.9) hospital bed-days per 100 person-days in the pre-SRE period and 28.2 (95% confidence interval, 27.8–28.6) bed-days per 100 person-days in the post-SRE period. Conclusion: SREs secondary to bone metastases in lung cancer patients are associated with a substantial number of hospital contacts and hospital bed-days. Keywords: hospital services, lung neoplasm, utilization

Jennie Maria Christin Strid,1 Henrik Gammelager,1 Martin Berg Johansen,1 Else Tønnesen,2 Christian Fynbo Christiansen,11Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus N, Denmark; 2Department of Anesthesiology and Intensive Care Medicine, Aarhus University Hospital, Aarhus C, DenmarkObjective: Current data on hospitalization and prognosis of acute asthma and status asthmaticus are inconclusive. We aim to analyze the rate of first-time hospitalizations for status asthmaticus among patients of all ages, the proportion admitted to intensive care units (ICU), and the 30-day mortality over a 16-year period.Methods: In this population-based cohort study, we used medical registries to identify all first-time status asthmaticus hospitalizations in Denmark from 1996 through 2011. Data on comorbidities were also obtained. We computed yearly hospitalization rates overall and by gender and age groups, and estimated the proportion requiring ICU admission. We estimated 30-day age- and gender-standardized mortality. We examined potential misclassification from acute exacerbation of chronic obstructive pulmonary disease (COPD) by excluding patients with preexisting or concurrent COPD.Results: Of the 5,001 patients identified with a first-time status asthmaticus hospitalization, 50.5% were male, 40.3% were ,15 years old, and 12.4% had comorbidity. The hospitalization rate increased from 48.0 per 1,000,000 person-years (PY) (95% confidence interval [CI]: 45.1–51.1 PY) during 1996–1999 to 70.1 per 1,000,000 PY (95% CI: 66.7–73.7 PY) during 2008–2011. This may be explained by an increased hospitalization rate of children. The standardized 30-day mortality risk declined from 3.3% (95% CI: 2.5%–4.1%) in 1996–1999 to 1.5% (95% CI: 0.9%–2.1%) in 2008–2011. During 2005–2011, 10.1% of status asthmaticus patients were admitted to the ICU. Hospitalization rates and mortality risk decreased by excluding 939 patients also registered with COPD, but overall temporal changes did not change.Conclusion: From 1996 to 2011, status asthmaticus hospitalization rate increased but remained below 100 hospitalizations per 1,000,000 PY. Thirty-day mortality risk was halved to less than 2%.Keywords: incidence, prognosis, cohort study, hospitalization, mortality, status asthmaticus

Sandra Kruchov Thygesen, Morten Olsen, Christian Fynbo ChristiansenDepartment of Clinical Epidemiology, Aarhus University Hospital, Aarhus, DenmarkBackground: Infant respiratory distress syndrome (IRDS) is the most common respiratory disease in preterm infants, and is associated with considerable morbidity and mortality. Valid data on IRDS are important in clinical epidemiological research.Objectives: The objective of this study was to estimate the positive predictive value (PPV) of the IRDS diagnosis registered in the population-based Danish National Patient Registry according to the International Classification of Diseases, 8th and 10th revisions.Methods: Between January 1, 1977 and December 31, 2008, we randomly selected three patients per year, 96 in total, who were registered with an IRDS diagnosis in the Danish National Patient Registry and living in the northern part of Denmark. Data on the infants included information on the presence of predefined clinical symptoms. We defined IRDS as the presence of at least two of four clinical symptoms (tachypnea, retractions or nasal flaring, grunting, and central cyanosis), which had to be present for more than 30 minutes. Using medical record review as the reference standard, we computed the positive predictive value of the registered IRDS diagnosis including 95% confidence intervals (CIs).Results: We located the medical record for 90 of the 96 patients (94%), and found an overall PPV of the IRDS diagnosis of 81% (95% CI 72%–88%). This did not vary substantially between primary and secondary diagnoses. The PPV was higher, at 89% (95% CI 80%–95%), for preterm infants born before 37 weeks of gestation.Conclusion: The PPV of the IRDS diagnosis in the Danish National Patient Registry is reasonable when compared with symptoms described in the corresponding medical records. The Danish National Patient Registry is a useful data source for studies of IRDS, particularly if restricted to preterm infants. Nonetheless, the potential impact of misclassification of the IRDS diagnosis must be considered.Keywords: epidemiology, data quality, validity, positive predictive value, hospital diagnosis, respiratory distress syndrome

Betina Vest-Hansen, Anders Hammerich Riis, Christian Fynbo Christiansen Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus N, Denmark Background: In recent years, the number of acute hospital admissions has increased and this has imposed both organizational and financial strains on the health care system. Consequently, it is of crucial importance that we have valid data on admission types in the administrative databases in order to provide data for health care planning and research. Objective: To examine the validity of registration of acute admissions among medical patients in the Danish National Patient Registry (DNPR) using medical record reviews as the reference standard. Methods: We used the nationwide DNPR to identify a sample of 160 medical patients admitted to a hospital in the North Denmark Region during 2009. Data on admission type was obtained from the DNPR and confirmed by a medical record review. We computed positive predictive values, sensitivity, and specificity including 95% confidence intervals (CI) using the medical record review as the reference standard. Results: Among the 160 medical inpatients identified in the DNPR, 128 were registered with an acute admission, and 32 were registered with a nonacute admission. Two medical records could not be located. Thus, the analyses included 158 medical patients. Among the 127 patients registered with acute admission, 124 were confirmed to be correctly classified. Correspondingly, 28 of the 31 patients with a registered nonacute admission were confirmed to be correctly classified. The overall positive predictive value of the acute admissions among medical patients was 97.6% (95% CI, 93.8%–99.3%). Sensitivity was 97.6% (95% CI, 93.8%–99.3%) and specificity was 90.3% (95% CI, 76.4%–97.2%). Conclusion: The registration of acute admission among medical patients in the DNPR has high validity. Keywords: medical patients, registries, epidemiology, positive predictive value, sensitivity, specificity

Marie Louise Svendsen,1 Henrik Gammelager,1 Claus Sværke,1 Mellissa Yong,2 Victoria M Chia,2 Christian F Christiansen,1 Jon P Fryzek1 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Center for Observational Research, Amgen, Thousand Oaks, CA, USA Objective: Skeletal-related events (SREs) among women with breast cancer may be associated with considerable use of health-care resources. We characterized inpatient and outpatient hospital visits in a national population-based cohort of Danish women with SREs secondary to breast cancer and bone metastases. Methods: We identified first-time breast cancer patients with bone metastases from 2003 through 2009 who had a subsequent SRE (defined as pathologic fracture, spinal cord compression, radiation therapy, or surgery to bone). Hospital visits included the number of inpatient hospitalizations, length of stay, number of hospital outpatient clinic visits, and emergency room visits. The number of hospital visits was assessed for a pre-SRE period (90 days prior to the diagnostic period), a diagnostic period (14 days prior to the SRE), and a post-SRE period (90 days after the SRE). Patients who experienced more than one SRE during the 90-day post-SRE period were defined as having multiple SREs and were followed until 90 days after the last SRE. Results: We identified 569 women with SREs secondary to breast cancer with bone metastases. The majority of women had multiple SREs (73.1%). A total of 20.9% and 33.4% of women with single and multiple SREs died in the post-SRE period, respectively. SREs were associated with a large number of hospital visits in the diagnostic period, irrespective of the number and type of SREs. Women with multiple SREs generally had a higher number of visits compared to those with a single SRE in the post-SRE period, eg, median length of hospitalization was 5 days (interquartile range 0–15) for women with a single SRE and 13 days (interquartile range 4–30) for women with multiple SREs. Conclusion: SREs secondary to breast cancer and bone metastases were associated with substantial use of hospital resources. Keywords: breast neoplasms, bone metastases, skeletal-related events, hospital services, utilization

Linea Blichert-Hansen, Malene S Nielsson, Rikke B Nielsen, Christian F Christiansen, Mette NørgaardDepartment of Clinical Epidemiology, Aarhus University Hospital, DenmarkBackground: Large health care databases provide a cost-effective data source for observational research in the intensive care unit (ICU) if the coding is valid. The aim of this study was to investigate the accuracy of the recorded coding of ICU admission, mechanical ventilation, and acute dialysis in the population-based Danish National Patient Registry (DNPR).Methods: We conducted the study in the North Denmark Region, including seven ICUs. From the DNPR we selected a total of 150 patients with an ICU admission by the following criteria: (1) 50 patients randomly selected among all patients registered with an ICU admission code, (2) 50 patients with an ICU admission code and a concomitant code for mechanical ventilation, and (3) 50 patients with an ICU admission code and a concomitant code for acute dialysis. Using the medical records as gold standard we estimated the positive predictive value (PPV) for each of the three procedure codes.Results: We located 147 (98%) of the 150 medical records. Of these 147 patients, 141 (95.9%; 95% confidence interval [CI]: 91.8–98.3) had a confirmed ICU admission according to their medical records. Among patients, who were selected only on the coding for ICU admission, the PPV for ICU admission was 87.2% (95% CI: 75.6–94.5). For the mechanical ventilation code, the PPV was 100% (95% CI: 95.1–100). Forty-nine of 50 patients with the coding for acute dialysis received this treatment, corresponding to a PPV of 98.0% (95% CI: 91.0–99.8).Conclusion: We found a high PPV for the coding of ICU admission and even higher PPVs for mechanical ventilation, and acute dialysis in the DNPR. The DNPR is a valuable data source for observational studies of ICU patients.Keywords: critical care, epidemiology, intensive care unit, positive predictive values, validity

Christian Fynbo ChristiansenDepartment of Clinical Epidemiology, Aarhus University Hospital, Aarhus, DenmarkAbstract: Venous thromboembolism (VTE) is a potentially serious complication occurring in 1%–2% of hospitalized medical patients. Despite this low absolute risk, as many as 82% of medical patients are considered to be at increased risk of developing VTE and are eligible for medical thromboprophylaxis. In this commentary, The author will discuss the main findings of a recent paper published in Clinical Epidemiology that questions the large proportion of individuals who are eligible for medical thromboprophylaxis, and also discuss the potential implications for the prevention of VTE. The recent paper demonstrated that when a population is divided into high- and low-risk groups, the maximum absolute risk depends on the inverse of the proportion of patients that is considered to be high risk. Consequently, even an effective treatment will only result in a small reduction in the absolute risk when the high-risk group comprises the largest proportion of this population. For medical thromboprophylaxis, this implies that even patients considered to be at high-risk for developing VTE have a maximum absolute VTE risk of 2% when the overall risk is 1.6%. Therefore, even an effective preventive initiative will only result in a small risk reduction. This small potential benefit should be weighed against potential harms associated with prophylaxis, mainly bleeding events. Still, there may be a reasonable overall balance between prevention of pulmonary embolism and major bleeding, mainly because major bleeding events are rare. Nonetheless, this discussion underscores that future risk prediction models should aim to predict the benefits and harms in individual patients in order to provide optimal care for the right patients.Keywords: comment, epidemiology, prevention, risk factors, venous thromboembolism

Steffen Christensen1, Martin Berg Johansen1, Christian Fynbo Christiansen1, Reinhold Jensen2, Stanley Lemeshow1,31Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; 2Department of Intensive Care, Skejby Hospital, Aarhus University Hospital, Aarhus, Denmark; 3Division of Biostatistics, College of Public Health, Ohio State University, Columbus, OH, USABackground: Physiology-based severity of illness scores are often used for risk adjustment in observational studies of intensive care unit (ICU) outcome. However, the complexity and time constraints of these scoring systems may limit their use in administrative databases. Comorbidity is a main determinant of ICU outcome, and comorbidity scores can be computed based on data from most administrative databases. However, limited data exist on the performance of comorbidity scores in predicting mortality of ICU patients.Objectives: To examine the performance of the Charlson comorbidity index (CCI) alone and in combination with other readily available administrative data and three physiology-based scores (acute physiology and chronic health evaluations [APACHE] II, simplified acute physiology score [SAPS] II, and SAPS III) in predicting short- and long-term mortality following intensive care.Methods: For all adult patients (n = 469) admitted to a tertiary university–affiliated ICU in 2007, we computed APACHE II, SAPS II, and SAPS III scores based on data from medical records. Data on CCI score age and gender, surgical/medical status, social factors, mechanical ventilation and renal replacement therapy, primary diagnosis, and complete follow-up for 1-year mortality was obtained from administrative databases. We computed goodness-of-fit statistics and c-statistics (area under ROC [receiver operating characteristic] curve) as measures of model calibration (ability to predict mortality proportions over classes of risk) and discrimination (ability to discriminate among the patients who will die or survive), respectively.Results: Goodness-of-fit statistics supported model fit for in-hospital, 30-day, and 1-year mortality of all combinations of the CCI score. Combining the CCI score with other administrative data revealed c-statistics of 0.75 (95% confidence interval [CI] 0.69–0.81) for in-hospital mortality, 0.75 (95% CI 0.70–0.80) for 30-day mortality, and 0.72 (95% CI 0.68–0.77) for 1-year mortality. There were no major differences in c-statistics between physiology-based systems and the CCI combined with other administrative data.Conclusion: The CCI combined with administrative data predict short- and long-term mortality for ICU patients as well as physiology-based scores.Keywords: epidemiology, CCI, physiology-based scores, ICU