scholarly journals Exposure misclassification in propensity score-based time-to-event data analysis

2021 ◽  
pp. 096228022199841
Author(s):  
Yingrui Yang ◽  
Molin Wang
Keyword(s):  
Propensity Score ◽  
Cox Regression ◽  
Asymptotic Properties ◽  
Practical Importance ◽  
Lung Cancer Mortality ◽  
Research Area ◽  
Health Study ◽  
Exposure Misclassification ◽  
Time To Event ◽  
Cox Models

In epidemiology, identifying the effect of exposure variables in relation to a time-to-event outcome is a classical research area of practical importance. Incorporating propensity score in the Cox regression model, as a measure to control for confounding, has certain advantages when outcome is rare. However, in situations involving exposure measured with moderate to substantial error, identifying the exposure effect using propensity score in Cox models remains a challenging yet unresolved problem. In this paper, we propose an estimating equation method to correct for the exposure misclassification-caused bias in the estimation of exposure-outcome associations. We also discuss the asymptotic properties and derive the asymptotic variances of the proposed estimators. We conduct a simulation study to evaluate the performance of the proposed estimators in various settings. As an illustration, we apply our method to correct for the misclassification-caused bias in estimating the association of PM2.5 level with lung cancer mortality using a nationwide prospective cohort, the Nurses’ Health Study. The proposed methodology can be applied using our user-friendly R program published online.

Abstract 160: Impact of Medication Adherence on Risk of Stroke, Major Bleeding and Other Outcomes in Atrial Fibrillation Patients Using Novel Oral Anticoagulants (Dabigatran and Rivaroxaban)

2017 ◽  
Vol 10 (suppl_3) ◽  
Author(s):  
Chinmay G Deshpande ◽  
Cynthia Willey Temkin ◽  
Robert Laforge ◽  
Stephen Kogut
Keyword(s):  
Atrial Fibrillation ◽  
Ischemic Stroke ◽  
Drug Use ◽  
Propensity Score ◽  
Major Bleeding ◽  
Cox Regression ◽  
Oral Anticoagulants ◽  
Cox Models ◽  
Adherence Assessment ◽  
The Impact

Introduction: Dabigatran and Rivaroxaban have shown better or similar efficacy to lower stroke risk compared to warfarin in clinical trials. Evidence suggests adherence to cardiac drugs tend to reduce outcomes and cost. Our study is the first to examine the impact of atleast 6 to 12 month adherence to NOACs on ischemic stroke, major bleeding, Deep Vein Thrombosis and Pulmonary Embolism (DVTPE) risk in a propensity score based matched sample. Methods: A retrospective cohort study utilized de-identified data from Optum® Clinformatics™ Data Mart database (OptumInsight, Eden Prairie, MN) (Jan 1, 2010 and Dec 31, 2012). Adult patients with ≥ 1 diagnosis of atrial fibrillation or flutter (ICD9 427.31/32), >1 prescription of NOACs, 6 months pre-index continuous enrollment and CHA2DS2VASC score >1 were included. Adherence was calculated using Proportion of Days Covered (PDC ≥80%) for atleast 6 and 12 months of NOAC use and cohorts (adherent vs. non adherent) were matched on propensity score using Inverse Probability Treatment Weighting (IPTW) controlling for demographic and clinical characteristics at baseline. The risk of ischemic stroke, major bleeding (primary outcomes) and DVTPE (exploratory outcome) was evaluated for the matched cohorts post adherence assessment using Cox regression. Results: Out of 25,150 NOAC patients, a total of 3,629 and 1,946 patients with atleast 6 and 12 months of NOAC use were included. Across 2 cohorts, the mean age of the sample was 65 years, 65% were males and >60% had a moderate-high risk of stroke (CHA2DS2VASC>2). Adherence (PDC ≥80%) was 77% and 76% for patients with 6 and 12 month drug use. Post 12 months of drug use, the overall incidence of bleeding, stroke, and DVTPE in the follow-up period was 4.42%, 1.80%, and 0.82% respectively. Each outcome was analyzed separately to avoid calculation of competing risks. Using Cox models with IPTW balanced cohorts, non-adherence was significantly (p ≤0.05) associated with an increase in stroke (≥ 1.5 fold) and DVTPE (≥ 2 fold) risk in both 6 and 12 month users. The risk of bleeding was not significantly different across adherent vs. non adherent users (Table). Conclusion: In our sample, adherence to NOACs was associated with a reduction in stroke and DVTPE risk but did not substantially increase bleeding risk. Further studies with newer NOACs are warranted.

scholarly journals Self-reported visual impairment, physical activity and all-cause mortality: The HUNT Study

2016 ◽  
Vol 45 (1) ◽  
pp. 33-41 ◽  
Author(s):  
Audun Brunes ◽  
W. Dana Flanders ◽  
Liv Berit Augestad
Keyword(s):  
Physical Activity ◽  
Visual Impairment ◽  
Mortality Risk ◽  
Cox Regression ◽  
Age Groups ◽  
Health Study ◽  
Cox Models ◽  
Hazard Ratios ◽  
All Cause Mortality ◽  
Hunt Study

Aims: To examine the associations of self-reported visual impairment and physical activity (PA) with all-cause mortality. Methods: This prospective cohort study included 65,236 Norwegians aged ⩾20 years who had participated in the Nord-Trøndelag Health Study (HUNT2, 1995−1997). Of these participants, 11,074 (17.0%) had self-reported visual impairment (SRVI). The participants’ data were linked to Norway’s Cause of Death Registry and followed throughout 2012. Hazard ratios and 95% confidence intervals (CI) were assessed using Cox regression analyses with age as the time-scale. The Cox models were fitted for restricted age groups (<60, 60−84, ⩾85 years). Results: After a mean follow-up of 14.5 years, 13,549 deaths were identified. Compared with adults with self-reported no visual impairment, the multivariable hazard ratios among adults with SRVI were 2.47 (95% CI 1.94–3.13) in those aged <60 years, 1.22 (95% CI 1.13–1.33) in those aged 60–84 years and 1.05 (95% CI 0.96–1.15) in those aged ⩾85 years. The strength of the associations remained similar or stronger after additionally controlling for PA. When examining the joint associations, the all-cause mortality risk of SRVI was higher for those who reported no PA than for those who reported weekly hours of PA. We found a large, positive departure from additivity in adults aged <60 years, whereas the departure from additivity was small for the other age groups. Conclusions: Adults with SRVI reporting no PA were associated with an increased all-cause mortality risk. The associations attenuated with age.

scholarly journals Integrative eQTL-weighted hierarchical Cox models for SNP-set based time-to-event association studies

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Haojie Lu ◽  
Yongyue Wei ◽  
Zhou Jiang ◽  
Jinhui Zhang ◽  
Ting Wang ◽  
...  
Keyword(s):  
Genetic Variants ◽  
Cox Regression ◽  
Association Studies ◽  
Effect Sizes ◽  
Time To Event ◽  
Omnibus Test ◽  
Cox Models ◽  
Functional Annotations ◽  
Analysis Strategy ◽  
Survival Risk

Abstract Background Integrating functional annotations into SNP-set association studies has been proven a powerful analysis strategy. Statistical methods for such integration have been developed for continuous and binary phenotypes; however, the SNP-set integrative approaches for time-to-event or survival outcomes are lacking. Methods We here propose IEHC, an integrative eQTL (expression quantitative trait loci) hierarchical Cox regression, for SNP-set based survival association analysis by modeling effect sizes of genetic variants as a function of eQTL via a hierarchical manner. Three p-values combination tests are developed to examine the joint effects of eQTL and genetic variants after a novel decorrelated modification of statistics for the two components. An omnibus test (IEHC-ACAT) is further adapted to aggregate the strengths of all available tests. Results Simulations demonstrated that the IEHC joint tests were more powerful if both eQTL and genetic variants contributed to association signal, while IEHC-ACAT was robust and often outperformed other approaches across various simulation scenarios. When applying IEHC to ten TCGA cancers by incorporating eQTL from relevant tissues of GTEx, we revealed that substantial correlations existed between the two types of effect sizes of genetic variants from TCGA and GTEx, and identified 21 (9 unique) cancer-associated genes which would otherwise be missed by approaches not incorporating eQTL. Conclusion IEHC represents a flexible, robust, and powerful approach to integrate functional omics information to enhance the power of identifying association signals for the survival risk of complex human cancers.

Disparities in the treatment and outcomes of lung cancer among HIV-infected people in Texas.

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. 6070-6070
Author(s):  
Gita Suneja ◽  
Meredith S. Shiels ◽  
Sharon K. Melville ◽  
Melanie A. Williams ◽  
Ramesh Rengan ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Cancer Treatment ◽  
Cox Regression ◽  
Lung Cancer Mortality ◽  
Cox Models ◽  
Histologic Subtype ◽  
Infected People ◽  
Specific Mortality ◽  
Cancer Specific Mortality ◽  
To Receive

6070 Background: HIV-infected (HIV+) people are at elevated risk for lung cancer and have higher mortality following lung cancer diagnosis than uninfected (HIV-) individuals. The disparity in survival is partly due to advanced stage at diagnosis, but it is unclear whether HIV+ people with lung cancer are less likely to receive cancer treatment, which could worsen survival. Methods: We included adults ≥ 18 years of age with lung cancer reported to the Texas cancer registry (N=156,930). HIV status was determined by linkage with the enhanced Texas HIV/AIDS Reporting System. We compared HIV+ and HIV- lung cancer cases with respect to demographic and clinical characteristics. For non-small cell lung cancer (NSCLC) cases, we identified predictors of cancer treatment (surgery, radiation, and chemotherapy) using logistic regression. We used Cox regression to evaluate the effects of HIV and treatment on lung cancer-specific mortality. Results: Compared with HIV- lung cancer cases (N=156,593), HIV+ lung cancer cases (N=337) were more likely to be young, non-Hispanic black, male, and to have distant stage disease (53.7% vs. 44.4%). HIV+ cases were less likely to receive cancer treatment than HIV- cases (60.3% vs. 77.5%; odds ratio 0.39, 95%CI 0.30-0.52 after adjustment for diagnosis year, age, sex, race, stage, and histologic subtype). In Cox models adjusted for these variables, both HIV infection (hazard ratio [HR] 1.34, 95%CI 1.15-1.56) and lack of cancer treatment (HR 1.69, 95%CI 1.66-1.72) were associated with higher lung cancer-specific mortality. After adjustment for cancer treatment, the association between HIV and lung cancer mortality was attenuated (HR 1.25, 95%CI 1.06-1.47). The association between HIV and lung cancer-specific mortality was stronger among untreated lung cancer cases (HR 1.32, 95%CI 1.01-1.72) than treated cases (adjusted HR 1.16, 95%CI 0.94-1.43; p-interaction=0.34). Conclusions: In this population-based study, HIV+ people with NSCLC were less likely to be treated for lung cancer than their HIV- counterparts. This lack of treatment may be partly responsible for higher cancer-related mortality in HIV+ cases. Further investigation is needed to understand disparities in cancer treatment for HIV+ people.

scholarly journals Coffee consumption and risk of renal cell carcinoma in the NIH-AARP Diet and Health Study

2021 ◽  
Author(s):  
Jongeun Rhee ◽  
Erikka Loftfield ◽  
Neal D Freedman ◽  
Linda M Liao ◽  
Rashmi Sinha ◽  
...  
Keyword(s):  
Renal Cell Carcinoma ◽  
Cell Carcinoma ◽  
Renal Cell ◽  
Cox Regression ◽  
Coffee Consumption ◽  
Health Study ◽  
Coffee Intake ◽  
Decaffeinated Coffee ◽  
Incident Cases ◽  
Reduced Risk

Abstract Background Coffee consumption has been associated with a reduced risk of some cancers, but the evidence for renal cell carcinoma (RCC) is inconclusive. We investigated the relationship between coffee and RCC within a large cohort. Methods Coffee intake was assessed at baseline in the National Institutes of Health–American Association of Retired Persons Diet and Health Study. Among 420 118 participants eligible for analysis, 2674 incident cases were identified. We fitted Cox-regression models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for coffee consumption vs non-drinkers. Results We observed HRs of 0.94 (95% CI 0.81, 1.09), 0.94 (0.81, 1.09), 0.80 (0.70, 0.92) and 0.77 (0.66, 0.90) for usual coffee intake of &lt;1, 1, 2–3 and ≥4 cups/day, respectively (Ptrend = 0.00003). This relationship was observed among never-smokers (≥4 cups/day: HR 0.62, 95% CI 0.46, 0.83; Ptrend = 0.000003) but not ever-smokers (HR 0.85, 95% CI 0.70, 1.05; Ptrend = 0.35; Pinteraction = 0.0009) and remained in analyses restricted to cases diagnosed &gt;10 years after baseline (HR 0.65, 95% CI 0.51, 0.82; Ptrend = 0.0005). Associations were similar between subgroups who drank predominately caffeinated or decaffeinated coffee (Pinteraction = 0.74). Conclusion In this investigation of coffee and RCC, to our knowledge the largest to date, we observed a 20% reduced risk for intake of ≥2 cups/day vs not drinking. Our findings add RCC to the growing list of cancers for which coffee consumption may be protective.

scholarly journals Identification of novel cell glycolysis related gene signature predicting survival in patients with breast cancer

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Feng Jiang ◽  
Chuyan Wu ◽  
Ming Wang ◽  
Ke Wei ◽  
Jimei Wang
Keyword(s):  
Breast Cancer ◽  
Overall Survival ◽  
Cox Regression ◽  
Single Gene ◽  
Prediction Method ◽  
Gene Signature ◽  
The Cancer Genome Atlas ◽  
Cox Models ◽  
Normal Tissues ◽  
Predictive Risk

AbstractOne of the most frequently identified tumors and a contributing cause of death in women is breast cancer (BC). Many biomarkers associated with survival and prognosis were identified in previous studies through database mining. Nevertheless, the predictive capabilities of single-gene biomarkers are not accurate enough. Genetic signatures can be an enhanced prediction method. This research analyzed data from The Cancer Genome Atlas (TCGA) for the detection of a new genetic signature to predict BC prognosis. Profiling of mRNA expression was carried out in samples of patients with TCGA BC (n = 1222). Gene set enrichment research has been undertaken to classify gene sets that vary greatly between BC tissues and normal tissues. Cox models for additive hazards regression were used to classify genes that were strongly linked to overall survival. A subsequent Cox regression multivariate analysis was used to construct a predictive risk parameter model. Kaplan–Meier survival predictions and log-rank validation have been used to verify the value of risk prediction parameters. Seven genes (PGK1, CACNA1H, IL13RA1, SDC1, AK3, NUP43, SDC3) correlated with glycolysis were shown to be strongly linked to overall survival. Depending on the 7-gene-signature, 1222 BC patients were classified into subgroups of high/low-risk. Certain variables have not impaired the prognostic potential of the seven-gene signature. A seven-gene signature correlated with cellular glycolysis was developed to predict the survival of BC patients. The results include insight into cellular glycolysis mechanisms and the detection of patients with poor BC prognosis.

scholarly journals Propensity score‐matched comparison of stenting as a bridge to surgery and emergency surgery for acute malignant left‐sided colonic obstruction

BMC Surgery ◽  
2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Yuepeng Cao ◽  
Qing Chen ◽  
Zhizhan Ni ◽  
Feng Wu ◽  
Chenshen Huang ◽  
...  
Keyword(s):  
Overall Survival ◽  
Propensity Score ◽  
Emergency Surgery ◽  
Perineural Invasion ◽  
Cox Regression ◽  
Colonic Obstruction ◽  
Metal Stents ◽  
Bridge To Surgery ◽  
Cox Regression Analysis ◽  
Self Expanding Metal Stents

Abstract Background Bridge to elective surgery (BTS) using self-expanding metal stents (SEMSs) is a common alternative to emergency surgery (ES) for acute malignant left-sided colonic obstruction (AMLCO). However, studies regarding the long-term impact of BTS are limited and have reported unclear results. Methods A multicenter observational study was performed at three hospitals from April 2012 to December 2019. Propensity score matching (PSM) was introduced to minimize selection bias. The primary endpoint was overall survival. The secondary endpoints included surgical approaches, primary resection types, total stent-related adverse effects (AEs), surgical AEs, length of hospital stay, 30-day mortality and tumor recurrence. Results Forty-nine patients in both the BTS and ES groups were matched. Patients in the BTS group more often underwent laparoscopic resection [31 (63.3%) vs. 8 (16.3%), p < 0.001], were less likely to have a primary stoma [13 (26.5%) vs. 26 (53.1%), p = 0.007] and more often had perineural invasion [25 (51.0 %) vs. 13 (26.5 %), p = 0.013]. The median overall survival was significantly lower in patients with stent insertion (41 vs. 65 months, p = 0.041). The 3-year overall survival (53.0 vs. 77.2%, p = 0.039) and 5-year overall survival (30.6 vs. 55.0%, p = 0.025) were significantly less favorable in the BTS group. In multivariate Cox regression analysis, stenting (hazard ratio(HR) = 2.309(1.052–5.066), p = 0.037), surgical AEs (HR = 1.394 (1.053–1.845), p = 0.020) and pTNM stage (HR = 1.706 (1.116–2.607), p = 0.014) were positively correlated with overall survival in matched patients. Conclusions Self-expanding metal stents as “a bridge to surgery” are associated with more perineural invasion, a higher recurrence rate and worse overall survival in patients with acute malignant left-sided colonic obstruction compared with emergency surgery.

Impact of immunosuppressive agents on clinical manifestations and outcome of Staphylococcus aureus bloodstream infection – A propensity score matched analysis in two large, prospectively evaluated cohorts

2021 ◽  
Author(s):  
Johannes Camp ◽  
Lina Glaubitz ◽  
Tim Filla ◽  
Achim J Kaasch ◽  
Frieder Fuchs ◽  
...  
Keyword(s):  
Staphylococcus Aureus ◽  
Propensity Score ◽  
Bloodstream Infection ◽  
Cox Regression ◽  
Immunosuppressive Agents ◽  
Late Complications ◽  
Competing Risk ◽  
Immunosuppressed Patients ◽  
Staphylococcus Aureus Bloodstream Infection ◽  
Baseline Characteristics

Abstract Background Staphylococcus aureus bloodstream infection (SAB) is a common, life-threatening infection. The impact of immunosuppressive agents on the outcome of patients with SAB is incompletely understood. Methods Data from two large prospective, international, multicenter cohort studies (INSTINCT and ISAC) between 2006 and 2015 were analyzed. Patients receiving immunosuppressive agents were identified and a 1:1 propensity score (PS) matched analysis was performed to adjust for baseline characteristics of patients. Overall survival and time to SAB-related late complications (SAB relapse, infective endocarditis, osteomyelitis, or other deep-seated manifestations) were analyzed by Cox regression and competing risk analyses, respectively. This approach was then repeated for specific immunosuppressive agents (corticosteroids [CSMT] and immunosuppressive agents other than steroids [IMOTS]). Results Of 3,188 analyzed patients, 309 were receiving immunosuppressive treatment according to our definitions and were matched to 309 non-immunosuppressed patients. After PS matching, baseline characteristics were well balanced. In the Cox regression analysis, we observed no significant difference in survival between the two groups (death during follow-up: 105/309 (33.9 %) immunosuppressed patients vs. 94/309 (30.4 %) non-immunosuppressed, hazard ratio 1.20 (95% CI 0.84–1.71). Competing risk analysis showed a cause-specific hazard ratio (CSHR) of 1.81 (95% CI 0.85–3.87) for SAB-related late-complications in patients receiving immunosuppressive agents. CSHR was higher in patients taking IMOTS (3.69; 95% CI 1.41–9.68). Conclusions Immunosuppressive agents were not associated with an overall higher mortality. The risk for SAB-related late complications in patients receiving specific immunosuppressive agents such as IMOTs warrants further investigations.

Ttrimetyllysine and risk of new-onset atrial fibrillation in two large norwegian cohorts

2021 ◽  
Vol 28 (Supplement_1) ◽  
Author(s):  
MM Svenningsson ◽  
I Dhar ◽  
GFT Svingen ◽  
EKR Pedersen ◽  
D Nilsen ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Angina Pectoris ◽  
Coronary Angiography ◽  
Stable Angina ◽  
Cox Regression ◽  
Health Study ◽  
Stable Angina Pectoris ◽  
Age And Gender ◽  
And Gender ◽  
New Onset

Abstract Funding Acknowledgements Type of funding sources: None. Background/Aim Increased plasma trimetyllysine (TML), a methylated amino acid, has recently been linked to higher risk of acute myocardial infarction (AMI). TML is also a precursor of trimethylamine-N oxide (TMAO), which has been linked to increased cardiovascular risk, including that of  atrial fibrillation (AF). We investigated the association between TML and new-onset AF in two large Norwegian cohorts. Methods The primary cohort consisted of 6396 participants in the community-based Hordaland Health Study (HUSK). The validation cohort consited of 2027 patients who underwent coronary angiography due to suspected stable angina pectoris in the Western Norway Coronary Angiography Cohort (WECAC). Information on new-onset AF was obtained by linking patient data to Norwegian public health registries. Risk associations were explored by Cox regression. Results During median (25th-75th percentile) follow-up of 10.9 (10.6-11.3) and 7.0 (6.3-8.6) years, 560 (8.8%) patients in the HUSK and 210 (10.4%) in the WECAC was diagnosed with AF. In the HUSK, the age and gender adjusted HR (95 % CI) for the 4th vs. 1st plasma TML quartiles 1.84 (1.37-2.48) p &lt; 0.001. In multivariable models the association was only slightly attenuated. Correspondingsly, the age and gender adjusted HR (95% CI) for the 4th vs. 1st TML quartiles in the WECAC was 1.48 (0.96-2.27) p = 0.07. Testing for collinearity between TMAO and TML revealed variance inflation factors between 1.0-1.1 in HUSK and WECAC, thus ruling out collinearity. Conclusion Plasma TML was associated with new-onset AF among subjects from the general population, and the relationship was independent from established AF risk factors. A similar trend was also seen in patients with suspected stable angina pectoris, strengthening our findings, which motivate further studies to explore potential pathophysiological relationships between one-carbon metabolism and cardiac arrhythmias

scholarly journals Diabetes mellitus was not associated with lower amputation-free survival after open revascularization for chronic limb-threatening ischemia – A nationwide propensity score adjusted analysis

2021 ◽  
pp. 1358863X2110082
Author(s):  
Erika Lilja ◽  
Anders Gottsäter ◽  
Mervete Miftaraj ◽  
Jan Ekelund ◽  
Björn Eliasson ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Propensity Score ◽  
Vascular Surgery ◽  
Incidence Rate ◽  
Cox Regression ◽  
Major Amputation ◽  
Incidence Rates ◽  
Major Adverse Cardiovascular Events ◽  
Cox Regression Analysis ◽  
Number Of Patients

The risk of major amputation is higher after urgently planned endovascular therapy for chronic limb-threatening ischemia (CLTI) in patients with diabetes mellitus (DM). The aim of this nationwide cohort study was to compare outcomes between patients with and without DM following urgently planned open revascularization for CLTI from 2010 to 2014. Out of 1537 individuals registered in the Swedish Vascular Registry, 569 were registered in the National Diabetes Register. A propensity score adjusted Cox regression analysis was conducted to compare outcome between the groups with and without DM. Median follow-up was 4.3 years and 4.5 years for patients with and without DM, respectively. Patients with DM more often had foot ulcers ( p = 0.034) and had undergone more previous amputations ( p = 0.001) at baseline. No differences in mortality, cardiovascular death, major adverse cardiovascular events (MACE), or major amputation were observed between groups. The incidence rate of stroke was 70% higher (95% CI: 1.11–2.59; p = 0.0137) and the incidence rate of acute myocardial infarction (AMI) 39% higher (95% CI: 1.00–1.92; p = 0.0472) among patients with DM in comparison to those without. Open vascular surgery remains a first-line option for a substantial number of patients with CLTI, especially for limb salvage in patients with DM. The higher incidence rates of stroke and AMI among patients with DM following open vascular surgery for infrainguinal CLTI require specific consideration preoperatively with the aim of optimizing medical treatment to improve cardiovascular outcome postoperatively.

Sign in / Sign up

Export Citation Format

Share Document