Validation of algorithms in studies based on routinely collected health data: general principles.

Clinicians, researchers, regulators, and other decision-makers increasingly rely on evidence from real-world data (RWD), including data routinely accumulating in health and administrative databases. RWD studies often rely on algorithms to operationalize variable definitions. An algorithm is a combination of codes or concepts used to identify persons with a specific health condition or characteristic. Establishing the validity of algorithms is a prerequisite for generating valid study findings that can ultimately inform evidence-based health care. This paper aims to systematize terminology, methods, and practical considerations relevant to the conduct of validation studies of RWD-based algorithms. We discuss measures of algorithm accuracy; gold/reference standard; study size; prioritizing accuracy measures; algorithm portability; and implication for interpretation. Information bias is common in epidemiologic studies, underscoring the importance of transparency in decisions regarding choice and prioritizing measures of algorithm validity. The validity of an algorithm should be judged in the context of a data source, and one size does not fit all. Prioritizing validity measures within a given data source depends on the role of a given variable in the analysis (eligibility criterion, exposure, outcome or covariate). Validation work should be part of routine maintenance of RWD sources.

Long-term outcomes after coronary intervention with biodegradable polymer stents in patients with acute coronary syndromes.

BACKGROUND: Patients with acute coronary syndromes (ACS) may have worse outcomes after percutaneous coronary intervention compared to patients without ACS. AIMS: To compare 5-year efficacy and safety outcomes in patients with and without ACS treated with biodegradable polymers, the ultrathin strut sirolimus-eluting Orsiro stent (O-SES) or the biolimus-eluting Nobori stent (N-BES). METHODS: The Scandinavian Organisation for Randomized Trials with Clinical Outcome VII is a randomized trial comparing O-SES and N-BES in an all-comer setting. Of 2525 patients, 1329 (53%) patients had ACS and 1196 (47%) patients were without ACS. Endpoints were target lesion failure (TLF) (a composite of cardiac death, target lesion myocardial infarction, or target lesion revascularization) and definite stent thrombosis within 5 years. RESULTS: At 5-year follow-up, TLF did not differ significantly between patients with and without ACS (12.3% vs. 13.2%; rate ratio (RR) 1.00; 95% confidence interval (CI): 0.70-1.44), whereas the risk of definite stent thrombosis was increased in patients with ACS (2.3% vs. 1.3; RR: 2.01 [95% CI: 1.01-3.98]). In patients with ACS, the rate of TLF was similar between O-SES and N-BES (12.4% vs. 12.3%; RR: 1.02; 95% CI: 0.74-1.40). The reduced risk of definite stent thrombosis in O-SES treated ACS patients within the first year (0.2% vs. 1.6%; RR: 0.12; 95% CI: 0.02-0.93) was not maintained after 5 years (1.8% vs. 2.7%; RR: 0.77; 95% CI: 0.37-1.63). CONCLUSION: Patients with ACS had an increased risk of stent thrombosis regardless of the stent type used. Long-term outcomes were similar for ACS patients treated with O-SES or N-BES at 5 years.

Risk of progression of cervical intraepithelial neoplasia grade 2 in human papillomavirus-vaccinated and unvaccinated women: a population-based cohort study.

BACKGROUND: Many countries have implemented active surveillance (ie, leaving the lesion untreated) as an option among younger women with cervical intraepithelial neoplasia grade 2 because regression rates are high and excisional treatment increases the risk for preterm birth in subsequent pregnancies. However, early identification of women at increased risk for progression to cervical intraepithelial neoplasia grade 3 or worse is important to ensure timely treatment. Because women who have received a human papillomavirus vaccine have a lower risk for cervical cancer, they may have a lower risk for progression of untreated cervical intraepithelial neoplasia grade 2 to cervical intraepithelial neoplasia grade 3 or worse. OBJECTIVE: This study aimed to investigate if women who received a human papillomavirus vaccine and who are undergoing active surveillance for cervical intraepithelial neoplasia grade 2 are less likely to progress to cervical intraepithelial neoplasia grade 3 or worse when compared with women who did not receive the vaccine. STUDY DESIGN: We conducted a population-based cohort study in Denmark using data from national health registers. We identified all women aged 18 to 40 years who were undergoing active surveillance for cervical intraepithelial neoplasia grade 2 from January 1, 2007, to December 31, 2020. Women with a previous record of cervical intraepithelial neoplasia grade 2 or worse, hysterectomy, or a loop electrosurgical excision procedure were excluded. Exposure was defined as having received ≥1 dose of a human papillomavirus vaccine at least 1 year before the cervical intraepithelial neoplasia grade 2 diagnosis. We used cumulative incidence functions to estimate the risk for progression to cervical intraepithelial neoplasia grade 3 or worse within 28 months using hysterectomy, emigration, and death as competing events. We used modified Poisson regression to calculate crude and adjusted relative risks of progression during the 28-month surveillance period. Results were stratified by age at vaccination and adjusted for index cytology, disposable income, and educational level. RESULTS: The study population consisted of 7904 women of whom 3867 (48.9%) were vaccinated at least 1 year before a diagnosis of cervical intraepithelial neoplasia grade 2. At the time of cervical intraepithelial neoplasia grade 2 diagnosis, women who were vaccinated were younger (median age, 25 years; interquartile range, 23-27 years) than those who were not (median age, 29 years; interquartile range, 25-33 years). The 28-month cumulative risk for cervical intraepithelial neoplasia grade 3 or worse was significantly lower among women who were vaccinated before the age of 15 years (22.9%; 95% confidence interval, 19.8-26.1) and between the ages of 15 and 20 years (31.5%; 95% confidence interval, 28.8-34.3) when compared with women who were not vaccinated (37.6%; 95% confidence interval, 36.1-39.1). Thus, when compared with women who were not vaccinated, those who were vaccinated before the age of 15 years had a 35% lower risk for progression to cervical intraepithelial neoplasia grade 3 or worse (adjusted relative risk, 0.65; 95% confidence interval, 0.57-0.75), whereas women who were vaccinated between the ages of 15 and 20 years had a 14% lower risk (adjusted relative risk, 0.86; 95% confidence interval, 0.79-0.95). For women who were vaccinated after the age of 20 years, the risk was comparable with that among women who were not vaccinated (adjusted relative risk, 1.02; 95% confidence interval, 0.96-1.09). CONCLUSION: Women who were vaccinated and who were undergoing active surveillance for cervical intraepithelial neoplasia grade 2 had a lower risk for progression to cervical intraepithelial neoplasia grade 3 or worse during 28 months of follow-up when compared with women who were not vaccinated but only if the vaccine was administered by the age of 20 years. These findings may suggest that the human papillomavirus vaccination status can be used for risk stratification in clinical management of cervical intraepithelial neoplasia grade 2.

Human papillomavirus genotypes and risk of persistence and progression in women undergoing active surveillance for cervical intraepithelial neoplasia grade 2.

BACKGROUND: In recent years, active surveillance has been introduced as an alternative to excisional treatment in younger women with cervical intraepithelial neoplasia grade 2 because regression rates are high and excisional treatment is associated with increased risk of preterm birth. However, early identification of women at increased risk of persistence/progression is important to ensure timely treatment. Evidence is limited on biomarkers that may be used to identify women at increased risk of persistence/progression. OBJECTIVE: This study aimed to describe human papillomavirus HPV type-specific persistence/progression in women undergoing active surveillance for cervical intraepithelial neoplasia grade 2. STUDY DESIGN: We conducted a historical cohort study of women aged 23 to 40 years diagnosed with cervical intraepithelial neoplasia grade 2 at Aarhus University Hospital from 2000 to 2010. Women were identified through the Danish Pathology Data Bank (DPDB) and were considered as undergoing active surveillance if they had a first record of a cervical biopsy within 2 years after index diagnosis and no loop electrosurgical excision procedure before this. Human papillomavirus genotyping was performed on archived tissue samples using the HPV SPF10-DEIA-LiPA25 system (DNA ELISA [enzyme-linked immunosorbent assay] HPV SPF10 kit and RHA HPV SPF10-LiPA25 kit). Persistence/progression was defined as having a record of cervical intraepithelial neoplasia grade ≥2 in the DPDB determined on the last and worst diagnosis on a biopsy or loop electrosurgical excision procedure specimen during follow-up. We estimated the relative risk (95% confidence interval) of persistence/progression using a modified Poisson model. RESULTS: A total of 455 women were included. Two-thirds were aged ≤30 years (73.8%) at index diagnosis, and nearly half had a high-grade index cytology (48.8%). Overall, 52.2% of all women had cervical intraepithelial neoplasia grade ≥2 during follow-up; 70.5% were human papillomavirus-16-positive and 29.5% were positive for other human papillomavirus types. Human papillomavirus-16 was associated with a significantly higher risk of persistence/progression (relative risk, 1.64; 95% confidence interval, 1.37-1.95) compared with non-human papillomavirus-16. The risk of persistence/progression was highest in human papillomavirus-16-positive women with a high-grade index cytology compared with human papillomavirus-16-positive women with a low-grade cytology (relative risk, 1.29; 95% confidence interval, 1.03-1.61), whereas no differences were observed across age groups. CONCLUSION: The highest risk of persistence/progression was observed among human papillomavirus-16-positive women, particularly those with associated high-grade cytology. These findings suggest that early excisional treatment should be considered in this group of women.

Triglycerides and risk of cardiovascular events in statin-treated patients with newly diagnosed type 2 diabetes: a Danish cohort study.

BACKGROUND: Elevated triglyceride levels are a clinically useful marker of remnant cholesterol. It is unknown whether triglycerides are associated with residual cardiovascular risk in CVD-naïve patients with newly diagnosed type 2 diabetes mellitus (T2DM), who are already on statin therapy. We aimed to assess the association between triglyceride levels and risk of major cardiovascular events (MACE) in statin-treated patients with newly diagnosed T2DM managed in routine clinical care. METHODS: This cohort study included newly diagnosed T2DM patients without a previous diagnosis of cardiovascular disease in Northern Denmark during 2005-2017. Individual triglyceride levels while on statin treatment were assessed within 1 year after T2DM diagnosis. The primary outcome was a composite of myocardial infarction, ischemic stroke, or cardiac death (MACE). Patients were followed from one year after T2DM diagnosis until 30 April 2021, MACE, emigration, or death. We used Cox regression to compute hazard ratios (HRs) controlling for confounding factors. RESULTS: Among 27,080 statin-treated patients with T2DM (median age 63 years; 53% males), triglyceride levels were < 1.0 mmol/L in 17%, 1.0-1.9 mmol/L in 52%, 2.0-2.9 mmol/L in 20%, and ≥ 3.0 mmol/L in 11%. During follow-up, 1,957 incident MACE events occurred (11.0 per 1000 person-years). Compared with triglyceride levels < 1.0 mmol/L, confounder-adjusted HRs for incident MACE were 1.14 (95% CI 1.00-1.29) for levels between 1.0 and 1.9 mmol/L, 1.30 (95% CI 1.12-1.51) for levels between 2.0 and 2.9 mmol/L, and 1.44 (95% CI 1.20-1.73) for levels ≥ 3.0 mmol/L. This association was primarily driven by higher rates of myocardial infarction and cardiac death and attenuated only slightly after additional adjustment for LDL cholesterol. Spline analyses confirmed a linearly increasing risk of MACE with higher triglyceride levels. Stratified analyses showed that the associations between triglyceride levels and MACE were stronger among women. CONCLUSIONS: In statin-treated patients with newly diagnosed T2DM, triglyceride levels are associated with MACE already from 1.0 mmol/L. This suggests that high triglyceride levels are a predictor of residual cardiovascular risk in early T2DM and could be used to guide allocation of additional lipid-lowering therapies for CVD prevention.

Impact of acute coronary syndrome on clinical outcomes after revascularization with the dual-therapy CD34 antibody-covered sirolimus-eluting Combo stent and the sirolimus-eluting Orsiro stent.

OBJECTIVES: To compare the efficacy and safety of the dual-therapy CD34 antibody-covered sirolimus-eluting Combo stent (DTS) and the sirolimus-eluting Orsiro stent (O-SES) in patients with and without acute coronary syndrome (ACS) included in the SORT OUT X study. BACKGROUND: The incidence of target lesion failure (TLF) after treatment with modern drug-eluting stents has been reported to be significantly higher in patients with ACS when compared to patients without ACS. Whether the results from the SORT OUT X study apply to patients with and without ACS remains unknown. METHODS: In total, 3146 patients were randomized to stent implantation with DTS (n = 1578; ACS: n = 856) or O-SES (n = 1568; ACS: n = 854). The primary end point, TLF, was a composite of cardiac death, target-lesion myocardial infarction (MI), or target lesion revascularization (TLR) within 1 year. RESULTS: At 1 year, the rate of TLF was higher in the DTS group compared to the O-SES group, both among patients with ACS (6.7% vs. 4.1%; incidence rate ratio: 1.65 [95% confidence interval, CI: 1.08-2.52]) and without ACS (6.0% vs. 3.2%; incidence rate ratio: 1.88 [95% CI: 1.13-3.14]). The differences were mainly explained by higher rates of TLR, whereas rates of cardiac death and target lesion MI did not differ significantly between the two stent groups in patients with or without ACS CONCLUSION: Compared to the O-SES, the DTS was associated with a higher risk of TLF at 12 months in patients with and without ACS. The differences were mainly explained by higher rates of TLR.

Clinical course of cervical intraepithelial neoplasia grade 2: a population-based cohort study.

BACKGROUND: Cervical intraepithelial neoplasia grade 2 has historically been the threshold for surgical excision, but because of high regression rates, many countries are transitioning to active surveillance. However, estimates for regression rates are based on small studies with heterogeneous definitions of regression and progression. OBJECTIVE: This study aimed to describe regression and progression rates of cervical intraepithelial neoplasia grade 2 using nationwide healthcare registry data. STUDY DESIGN: This was a nationwide population-based cohort study on women aged 18 to 40 years who had undergone active surveillance for cervical intraepithelial neoplasia grade 2 in Denmark from 1998 to 2020. This study excluded women with a previous record of cervical intraepithelial neoplasia grade 2 or worse or surgical excision. Cumulative incidence functions were used to estimate the rates of regression and progression at 6, 12, 18, and 24 months after diagnosis. In addition, a modified Poisson regression was used to estimate the crude and adjusted relative risks of progression within 24 months stratified by index cytology and age. RESULTS: During the study period, 11,056 women underwent active surveillance, 6767 of whom regressed and 3580 of whom progressed within 24 months. This corresponded to regression rates of 62.9% (95% confidence interval, 61.9-63.8) and progression rates of 33.3% (95% confidence interval, 32.4-34.2) at 24 months of follow-up. Most women regressed (90%) or progressed (90%) within the first 12 months. Women with high-grade index cytology had a higher risk of progression than women with normal index cytology (adjusted relative, 1.58; 95% confidence interval, 1.43-1.76), whereas there was no difference in the risk of progression between women aged 30 and 40 years and women aged 23 to 29 years (adjusted relative risk, 0.98; 95% confidence interval, 0.88-1.10). CONCLUSION: The observed high regression rates of cervical intraepithelial neoplasia grade 2 supported the transition in clinical management from surgical excision to active surveillance, particularly among women with low-grade or normal index cytology.

Randomized Clinical Comparison of the Dual-Therapy CD34 Antibody-Covered Sirolimus-Eluting Combo Stent With the Sirolimus-Eluting Orsiro Stent in Patients Treated With Percutaneous Coronary Intervention: The SORT OUT X Trial.

BACKGROUND: Target lesion failure remains an issue with contemporary drug-eluting stents. Thus, the dual-therapy sirolimus-eluting and CD34+ antibody-coated Combo stent (DTS) was designed to further improve early healing. This study aimed to investigate whether the DTS is noninferior to the sirolimus-eluting Orsiro stent (SES) in an all-comers patient population. METHODS: The SORT OUT X (Combo Stent Versus Orsiro Stent) trial, was a large-scale, randomized, multicenter, single-blind, 2-arm, noninferiority trial with registry-based follow-up. The primary end point target lesion failure was a composite of cardiac death, myocardial infarction, or target lesion revascularization within 12 months, analyzed using intention-to-treat. The trial was powered for assessing target lesion failure noninferiority of the DTS compared with the SES with a predetermined noninferiority margin of 0.021. RESULTS: A total of 3146 patients were randomized to treatment with the DTS (1578 patients; 2008 lesions) or SES (1568 patients; 1982 lesions). At 12 months, intention-to-treat analysis showed that 100 patients (6.3%) assigned the DTS and 58 patients (3.7%) assigned the SES met the primary end point (absolute risk difference, 2.6% [upper limit of 1-sided 95% CI, 4.1%]; P (noninferiority)=0.76). The SES was superior to the DTS (incidence rate ratios for target lesion failure, 1.74 [95% CI, 1.26-2.41]; P=0.00086). The difference was explained mainly by a higher incidence of target lesion revascularization in the DTS group compared with the SES group (53 [3.4%] vs. 24 [1.5%]; incidence rate ratio, 2.22 [95% CI, 1.37-3.61]; P=0.0012). CONCLUSIONS: The DTS did not confirm noninferiority to the SES for target lesion failure at 12 months in an all-comer population. The SES was superior to the DTS mainly because the DTS was associated with an increased risk of target lesion revascularization. However, rates of death, cardiac death, and myocardial infarction at 12 months did not differ significantly between the 2 stent groups. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT03216733.

Impact of diabetes on 1-year clinical outcome in patients undergoing revascularization with the BioFreedom stents or the Orsiro stents from the SORT OUT IX trial.

This sub-study of the SORT OUT IX trial sought to compare clinical outcomes between patients with diabetes randomized to implantation of either the polymer-free biolimus A9-coated BioFreedom stent (BF-BES) or the ultra-thin strut, biodegradable polymer sirolimus-eluting Orsiro stent (O-SES). Patients with diabetes have an increased risk of target lesion failure (TLF) after percutaneous coronary intervention (PCI). The impact of different stent types in patients with diabetes is still discussed. A total of 607 of the 3151 patients (19.3%) enrolled in the SORT OUT IX study had diabetes. Randomization was stratified by patients with/without diabetes; 304 received BF-BES and 303 O-SES. The primary endpoint was TLF, which was a composite of cardiac death, myocardial infarction (not related to other than the index lesion) and target lesion revascularization (TLR) within 1 year. After 1 year, patients with diabetes had higher TLF (7.2% vs. 3.7%, incidence rate ratio [IRR]: 1.65; 95% confidence interval [CI]: 1.08-2.50), than patients without diabetes. TLF did not differ significantly between BF-BES and O-SES in patients with diabetes (8.2% vs. 6.3%, IRR: 1.17; 95% CI: 0.63-2.20). In patients with diabetes, cardiac death occurred in 2.3% of BF-BES and in 3.6% of O-SES (IRR: 0.58; 95% CI: 0.23-1.45) and TLR occurred in 5.3% and 2.3% of BF-BES and O-SES, respectively (IRR: 2.12; 95% CI: 0.81-5.56). Definite stent thrombosis rates of 1.3% were found in both stent types. Patients with diabetes had higher 1-year TLF rate after PCI compared to patients without diabetes, whereas TLF did not differ significantly between the two stent types BF-BES and O-SES in patients with diabetes.

Impact of diabetes on clinical outcomes after revascularization with the dual therapy CD34 antibody-covered sirolimus-eluting Combo stent and the sirolimus-eluting Orsiro stent.

OBJECTIVES: To compare the efficacy and safety of the dual therapy CD34 antibody-covered sirolimus-eluting Combo stent (DTS) and the sirolimus-eluting Orsiro stent (SES) in patients with and without diabetes mellitus (DM) included in the Scandinavian Organization for Randomized Trials with Clinical Outcome (SORT OUT) X study. BACKGROUND: The incidence of target lesion failure (TLF) after treatment with modern drug-eluting stents has been reported to be significantly higher in patients with DM when compared to patients without DM. Thus, whether the results from the SORT OUT X study apply to patients with and without DM remains unknown. METHODS: In total 3146 patients were randomized to stent implantation with DTS (n = 1578; DM: n = 279) or SES (n = 1568; DM: n = 271). The primary end point, TLF, was a composite of cardiac death, target-lesion myocardial infarction (MI), or target lesion revascularization (TLR) within 1 year. RESULTS: At 1 year, the rate of TLF was increased in the DTS group compared to the SES group, both among patients with DM (9.3% vs. 4.8%; risk difference: 4.5%; incidence rate ratio: 1.99, 95% confidence interval [CI]: 1.02-3.90) and without DM (5.7% vs. 3.5%; incidence rate ratio: 1.67, 95% CI: 1.15-2.42). The differences were mainly explained by higher rates of TLR. CONCLUSION: Compared to the SES, the DTS was associated with an increased risk of TLF at 12 months in patients with and without DM. The differences were mainly explained by higher rates of TLR, whereas rates of cardiac death and target lesion MI did not differ significantly between the two stent groups in patients with or without DM.

Randomized Comparison of the Polymer-Free Biolimus-Coated BioFreedom Stent With the Ultrathin Strut Biodegradable Polymer Sirolimus-Eluting Orsiro Stent in an All-Comers Population Treated With Percutaneous Coronary Intervention: The SORT OUT IX Trial.

BACKGROUND: In patients with increased bleeding risk, the biolimus A9-coated BioFreedom stent, a stainless steel drug-coated stent free from polymer, has shown superiority compared with a bare-metal stent. The aim of this study was to investigate whether the BioFreedom stent is noninferior to a modern ultrathin strut biodegradable polymer cobalt-chromium sirolimus-eluting Orsiro stent in an all-comers patient population treated with percutaneous coronary intervention. METHODS: The SORT OUT IX trial (Scandinavian Organization for Randomized Trials With Clinical Outcome IX), was a large-scale, registry-based, randomized, multicenter, single-blind, 2-arm, noninferiority trial. The primary end point, major adverse cardiovascular events, was defined as the composite of cardiac death, myocardial infarction not related to any segment other than the target lesion, or target lesion revascularization within 1 year, analyzed by intention-to-treat. The trial was powered to assess noninferiority for major adverse cardiovascular events of the BioFreedom stent compared with the Orsiro stent with a predetermined noninferiority margin of 0.021. RESULTS: Between December 14, 2015 and April 21, 2017, 3151 patients were assigned to treatment with the BioFreedom stent (1572 patients, 1966 lesions) or to the Orsiro stent (1579 patients, 1985 lesions). Five patients were lost to follow-up because of emigration (99.9% follow-up rate). Mean age was 66.3±10.9, diabetes mellitus was seen in 19.3% of patients, and 53% of the patients had acute coronary syndromes. At 1 year, intention-to-treat analysis showed that 79 (5.0%) patients, who were assigned the BioFreedom stent, and 59 (3.7%), who were assigned the Orsiro stent, met the primary end point (absolute risk difference 1.29% [upper limit of one-sided 95% CI 2.50%]; Pnoninferiority=0.14). Significantly more patients in the BioFreedom stent group had target lesion revascularization than those in the Orsiro stent group (55 [3.5%] vs 20 [1.3%], rate ratio 2.77 [95% CI, 1.66-4.62]; P<0.0001). CONCLUSIONS: The biolimus A9-coated BioFreedom polymer-free stent did not meet criteria for noninferiority for major adverse cardiovascular events at 12 months when compared with the ultrathin strut biodegradable polymer sirolimus-eluting Orsiro stent in an all-comers population Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT02623140.

Trends in hysterectomy-corrected uterine cancer mortality rates during 2002 to 2015: mortality of nonendometrioid cancer on the rise?

Corpus uteri cancer is the most common gynecological malignancy in most developed countries. The disease is typically diagnosed at an early stage, is of endometrioid histologic subtype, and has a fairly good prognosis. Here, we describe hysterectomy-corrected mortality rates of corpus uteri cancer, overall and stratified by age, stage and histologic subtype. Using data from nationwide Danish registries, we calculated uncorrected and hysterectomy-corrected age-standardized mortality rates of corpus uteri cancer among women ≥35 years during 2002 to 2015. Individual-level hysterectomy status was obtained from national registries; hysterectomy-corrected mortality rates were calculated by subtracting posthysterectomy person-years from the denominator, unless hysterectomy was performed due to corpus uteri cancer. Correction for hysterectomy resulted in a 25.5% higher mortality rate (12.3/100000 person-years vs 9.8/100000 person-years). Mortality rates were highest in women aged 70+, irrespective of year of death, histologic subtype and stage. A significant decline was observed in overall hysterectomy-corrected mortality rates from 2002 to 2015, particularly among women aged 70+. Mortality rates of endometrioid cancer declined significantly over time (annual percent change [APC]: -2.32, 95% CI -3.9, -0.7, P = .01), whereas rates of nonendometrioid cancer increased (APC: 5.90, 95% CI: 3.0, 8.9, P < .001). With respect to stage, mortality rates increased significantly over time for FIGOI-IIa (APC: 6.18 [95% CI: 1.9, 10.7] P = .01) but remained unchanged for FIGO IIb-IV. In conclusion, increasing mortality rates of nonendometrioid cancer paralleled the previously observed rise in incidence rates of this histologic subtype. Given the poor prognosis of nonendometrioid cancer, more studies are needed to clarify the underlying reason for these findings.

Outcomes in patients with lung cancer treated with crizotinib and erlotinib in routine clinical practice: A post-authorization safety cohort study conducted in Europe and in the United States.

PURPOSE: We examined safety outcomes of interest (SOI) and overall survival (OS) among lung cancer patients initiating crizotinib and erlotinib in routine clinical practice. METHODS: This descriptive cohort study used routinely collected health data in Denmark, Finland, Sweden, the Netherlands, and the United States (US) during 2011-2017, following crizotinib commercial availability in each country. Among crizotinib or erlotinib initiators, we reported baseline characteristics and incidence rates and cumulative incidences of the SOI - hepatotoxicity, pneumonitis/interstitial lung disease, QT interval prolongation-related events, bradycardia, vision disorders, renal cysts, edema, leukopenia, neuropathy, photosensitivity, malignant melanoma, gastrointestinal perforation, cardiac failure and OS. Results from the European Union (EU) countries were combined using meta-analysis; results from the US were reported separately. RESULTS: There were 456 patients in the crizotinib cohort and 2957 patients in the erlotinib cohort. Rates of the SOI per 1000 person-years in the crizotinib cohort ranged from 0 to 65 in the EU and from 0 to 374 in the US. Rates of the SOI per 1000 person-years in the erlotinib cohort ranged from 0 to 91 in the EU and from 3 to 394 in the US. In the crizotinib cohort, 2-year OS was ~50% in both EU and US. In the erlotinib cohort, 2-year OS was 21% in the EU and 35% in the US. CONCLUSIONS: This study describes clinical outcomes among lung cancer patients initiating crizotinib or erlotinib in routine clinical practice. Differences between SOI rates in EU and US may be partially attributable to differences in the underlying databases.

Long-Term Risk of Developing Dementia After Electroconvulsive Therapy for Affective Disorders.

OBJECTIVES: Severe depression is associated with an increased risk of developing dementia, however, whether treatment with electroconvulsive therapy (ECT) modify this risk remains unknown. METHODS: In this matched cohort study, 1089 consecutive in-patients with affective disorders, receiving ECT during the period 1982 to 2000, were matched with 3011 in-patients with affective disorders not treated with ECT (non-ECT), and 108,867 individuals randomly selected from the background population. The comparison cohorts were matched on sex, age, and the non-ECT cohort was further matched according to diagnoses and admission period and hospital. Dementia diagnoses were retrieved from the national patient health registry. Analyses were adjusted for disease severity, somatic, and psychiatric comorbidities. RESULTS: The cumulative incidence of dementia was 13.45% (10.75-16.46%) in the ECT cohort after 34 years of follow-up, 10.53% (8.5-12.81%) in the non-ECT cohort, and 8.43% (8.17-8.7%) in the background cohort. Using the ECT cohort as reference and age as the underlying time scale, the adjusted hazard ratio of developing dementia was 0.73 (0.52-1.04) in the non-ECT cohort and 0.61 (0.49-0.76) in the background cohort. The stratified analysis based on age at index (<65 years; 65-80 years; >80 years) found no age-related difference in the risk of developing dementia between the ECT cohort and non-ECT cohort. CONCLUSIONS: The ECT treatment of affective disorders was not associated with an increased long-term risk of developing dementia compared with in-patients with affective disorders not treated with ECT.

Comparison of the polymer-free biolimus-coated BioFreedom stent with the thin-strut biodegradable polymer sirolimus-eluting Orsiro stent in an all-comers population treated with percutaneous coronary intervention: Rationale and design of the randomized SORT OUT IX trial.

BACKGROUND: In patients with increased bleeding risk during dual antiplatelet therapy, the biolimus A9-coated BioFreedom, a stainless steel drug-coated stent devoid of polymer, has shown superiority compared to a bare-metal stent. The aim of this study was to investigate whether the polymer-free biolimus A9-coated BioFreedom is noninferior to a modern thin-strut biodegradable polymer cobalt-chromium sirolimus-eluting Orsiro stent in an all-comers patient population treated with percutaneous coronary intervention. METHODS: The multicenter SORT OUT IX trial (NCT02623140) randomly assigned all-comers patients to treatment with the BioFreedom drug-coated stent or the biodegradable polymer Orsiro stent in 4 Danish University Hospitals. The primary end point target lesion failure is a composite of cardiac death, myocardial infarction (not related to other than index lesion), or target lesion revascularization within 12 months. Clinically driven event detection based on Danish registries will be used and continue through 5 years. Assuming an event rate of 4.2% in each stent group, 1,563 patients in each treatment arm will provide 90% power to detect noninferiority of the drug-coated BioFreedom stent with a noninferiority margin of 2.1%. RESULTS: A total of 3,150 patients have been randomized and enrolled in the study. CONCLUSIONS: The SORT OUT IX trial will determine whether the drug-coated BioFreedom stent is noninferior to a modern biodegradable polymer Orsiro stent.

Impact of diabetes on clinical outcomes after revascularization with sirolimus-eluting and biolimus-eluting stents with biodegradable polymer from the SORT OUT VII trial.

OBJECTIVES: In this substudy of the SORT OUT VII trial, the clinical outcomes among patient with diabetes mellitus treated with Orsiro sirolimus-eluting stent (O-SES; Biotronik, Bülach, Switzerland) or Nobori biolimus-eluting stent (N-BES; Terumo, Tokyo, Japan) were compared. BACKGROUND: Diabetes is associated with increased risk of target lesion failure (TLF) after percutaneous coronary intervention. METHODS: In total, 2525 patients were randomized to stent implantation with O-SES (n = 1261, diabetes: n = 236) or N-BES (n = 1264, diabetes: n = 235). The primary endpoint, TLF, was a composite of cardiac death, target-lesion myocardial infarction (MI), or target lesion revascularization (TLR) within 2 years. RESULTS: At 2 year, TLF did not differ between O-SES vs N-BES in diabetic (9.3% vs 9.4%; RR 0.98, 95% CI 0.54-1.78) patients. The individual components of the primary endpoint did not differ among stent type. In diabetics, cardiac death occurred in 3% of O-SES-treated and in 3.8% of N-BES-treated patients (RR 0.77, 95% CI 0.29-2.08), MI occurred in 3.0% of O-SES-treated and in 3.8% of N-BES-treated patients (RR 0.76, 95% CI 0.28-2.06) and TLR occurred in 5,5% of O-SES-treated and in 6.0% of N-BES-treated patients (RR 0.91, 95% CI 0.43-1.95). CONCLUSION: TLF did not differ between O-SES- and N-BES-treated diabetic patients.

Hysterectomy-corrected cervical cancer mortality rates in Denmark during 2002-2015: A registry-based cohort study.

INTRODUCTION: We aimed to describe the cervical cancer mortality rates after correcting for hysterectomy and to evaluate trends over calendar time and by age. MATERIAL AND METHODS: Using data from nationwide registries, we calculated uncorrected and hysterectomy-corrected cervical cancer mortality rates among women age ≥20 years in Denmark during 2002-2015. We calculated hysterectomy-corrected rates by subtracting post-hysterectomy person-years from the denominator, unless hysterectomy was performed due to cervical cancer. RESULTS: The age-standardized hysterectomy-corrected cervical cancer mortality rate of 5.8/100 000 was 18.4% higher than the corresponding uncorrected rate. The hysterectomy-corrected cervical cancer mortality rate increased significantly with age, particularly in women aged 65+ (annual percent change +4.57), peaking at 26.1/100 000 person-years in women aged ≥85. The hysterectomy-corrected cervical cancer mortality declined significantly over calendar time, from 6.3/100 000 person-years in 2002 to 4.5/100 000 person-years in 2015 (annual percent change -0.22). This was mainly due to a significant decline in women aged 60 years and older, from 20.4/100 000 person-years in 2003 to 9.3/100 000 person-years in 2015 (annual percent change -0.75). CONCLUSIONS: The mortality of cervical cancer increased significantly by age, particularly among women aged 65 years and older for whom routine cervical cancer screening is not recommended. Understanding reasons for the high mortality rate in older women is critical, as this may help identify interventions needed to ensure a continued decline in cervical cancer mortality in older Danish women.

Pathophysiology-based phenotyping in type 2 diabetes: A clinical classification tool.

BACKGROUND: Type 2 diabetes may be a more heterogeneous disease than previously thought. Better understanding of pathophysiological subphenotypes could lead to more individualized diabetes treatment. We examined the characteristics of different phenotypes among 5813 Danish patients with new clinically diagnosed type 2 diabetes. METHODS: We first identified all patients with rare subtypes of diabetes, latent autoimmune diabetes of adults (LADA), secondary diabetes, or glucocorticoid-associated diabetes. We then used the homeostatic assessment model to subphenotype all remaining patients into insulinopenic (high insulin sensitivity and low beta cell function), classical (low insulin sensitivity and low beta cell function), or hyperinsulinemic (low insulin sensitivity and high beta cell function) type 2 diabetes. RESULTS: Among 5813 patients diagnosed with incident type 2 diabetes in the community clinical setting, 0.4% had rare subtypes of diabetes, 2.8% had LADA, 0.7% had secondary diabetes, 2.4% had glucocorticoid-associated diabetes, and 93.7% had WHO-defined type 2 diabetes. In the latter group, 9.7% had insulinopenic, 63.1% had classical, and 27.2% had hyperinsulinemic type 2 diabetes. Classical patients were obese (median waist 105 cm), and 20.5% had cardiovascular disease (CVD) at diagnosis, while insulinopenic patients were fairly lean (waist 92 cm) and 17.5% had CVD (P = 0.14 vs classical diabetes). Hyperinsulinemic patients were severely obese (waist 112 cm), and 25.5% had CVD (P < 0.0001 vs classical diabetes). CONCLUSIONS: Patients clinically diagnosed with type 2 diabetes are a heterogeneous group. In the future, targeted treatment based on pathophysiological characteristics rather than the current "one size fits all" approach may improve patient prognosis.

The temporal and age-dependent patterns of hysterectomy-corrected cervical cancer incidence rates in Denmark: a population-based cohort study.

INTRODUCTION: Hysterectomy is a common gynecological procedure; however, the incidence of total and subtotal hysterectomy varies across countries, by age, and over time. As only women with an intact cervix are at risk of cervical cancer, failing to remove hysterectomized women from the denominator may underestimate the cervical cancer incidence. We aimed to describe the temporal and age-dependent patterns of cervical cancer incidence in Denmark before and after correction for hysterectomy. MATERIAL AND METHODS: Using data from national registries we calculated uncorrected and hysterectomy-corrected cervical cancer incidence rates among women ≥20 years during 2000-11. Hysterectomy-corrected rates were calculated by subtracting post-hysterectomy person-years from the denominator. RESULTS: The overall uncorrected cervical cancer incidence rate was 17.8/100 000 person-years (95% CI 17.3-18.3). After correction for hysterectomy, the rate increased by 8.4% to 19.3/100 000 person-years (95% CI 18.8-19.9). The highest uncorrected incidence was seen in women aged 35-39 years, peaking at 24.4/100 000 person-years, whereas the highest hysterectomy-corrected cervical cancer incidence rate was observed in women aged 75-79 years (29.4/100 000 person-years). Over time, women ≥60 years had the highest hysterectomy-corrected cervical cancer incidence. CONCLUSIONS: Correcting for hysterectomy incidence resulted in a higher cervical cancer incidence and a shift in the peak incidence from age 35-39 years to age 75-79 years. Over time, women ≥60 years were at the highest risk of cervical cancer. Given the high incidence in women >60-65 years, when women are eligible to exit screening, a revision of the screening guidelines may be warranted.

Patterns of migrating soaring migrants indicate attraction to marine wind farms.

Monitoring of bird migration at marine wind farms has a short history, and unsurprisingly most studies have focused on the potential for collisions. Risk for population impacts may exist to soaring migrants such as raptors with K-strategic life-history characteristics. Soaring migrants display strong dependence on thermals and updrafts and an affinity to land areas and islands during their migration, a behaviour that creates corridors where raptors move across narrow straits and sounds and are attracted to islands. Several migration corridors for soaring birds overlap with the development regions for marine wind farms in NW Europe. However, no empirical data have yet been available on avoidance or attraction rates and behavioural reactions of soaring migrants to marine wind farms. Based on a post-construction monitoring study, we show that all raptor species displayed a significant attraction behaviour towards a wind farm. The modified migratory behaviour was also significantly different from the behaviour at nearby reference sites. The attraction was inversely related to distance to the wind farm and was primarily recorded during periods of adverse wind conditions. The attraction behaviour suggests that migrating raptor species are far more at risk of colliding with wind turbines at sea than hitherto assessed.