Adjuvant Radiation in Older Patients With Glioblastoma: A Retrospective Single Institution Analysis

ObjectivesStandard 6-week and hypofractionated 3-week courses of adjuvant radiation therapy (RT) are both options for older patients with glioblastoma (GBM), but deciding the optimal regimen can be challenging. This analysis explores clinical factors associated with selection of RT course, completion of RT, and outcomes following RT.Materials and MethodsThis IRB-approved retrospective analysis identified patients ≥70 years old with GBM who initiated adjuvant RT at our institution between 2004 and 2016. We identified factors associated with standard or hypofractionated RT using the Cochran-Armitage trend test, estimated time-to-event endpoints using the Kaplan-Meier method, and found predictors of overall survival (OS) using Cox proportional hazards models.ResultsSixty-two patients with a median age of 74 (range 70–90) initiated adjuvant RT, with 43 (69%) receiving standard RT and 19 (31%) receiving hypofractionated RT. Selection of short-course RT was associated with older age (p = 0.04) and poor KPS (p = 0.03). Eight (13%) patients did not complete RT, primarily for hospice care due to worsening symptoms. After a median follow-up of 37 months, median OS was 12.3 months (95% CI 9.0–15.1). Increased age (p < 0.05), poor KPS (p < 0.0001), lack of MGMT methylation (p < 0.05), and lack of RT completion (p < 0.0001) were associated with worse OS on multivariate analysis. In this small cohort, GTV size and receipt of standard or hypofractionated RT were not associated with OS.ConclusionsIn this cohort of older patients with GBM, age and KPS was associated with selection of short-course or standard RT. These regimens had similar OS, though a subset of patients experienced worsening symptoms during RT and discontinued treatment. Further investigation into predictors of RT completion and survival may help guide adjuvant therapies and supportive care for older patients.

Download Full-text

Effect of Sedative-Hypnotic Medicines on Mortality in Peritoneal Dialysis Patients with Sleep Disorders: A Retrospective Cohort Study

Blood Purification ◽

10.1159/000484922 ◽

2017 ◽

Vol 45 (1-3) ◽

pp. 95-101

Author(s):

Ying Zhou ◽

Peiran Yin ◽

Bin Li ◽

Jianying Li ◽

Naya Huang ◽

...

Keyword(s):

Peritoneal Dialysis ◽

Sleep Disorders ◽

Older Patients ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Dialysis Patients ◽

Factors Associated ◽

Cox Proportional Hazards Models ◽

Hypnotic Medication ◽

All Cause Mortality

Background: Sedative-hypnotic medication is widely used among continuous ambulatory peritoneal dialysis (CAPD) patients with sleep disorders; however, its effect on mortality has rarely been investigated. Methods: Logistic regression was employed to identify factors associated with sedative-hypnotic medication, whose effect on mortality was evaluated by Cox proportional hazards models. Results: A total of 146 CAPD patients with sleep disorders were recruited, of which 46 patients (31.5%) used either benzodiazepines or zolpidem. Sedative-hypnotic medication was more frequently used by older patients and those with longer duration of CAPD therapy and there was no significant association between sedative-hypnotic medicines and all-cause mortality after adjusting for age, gender, diabetes, cardiovascular disease, and duration of CAPD. Conclusion: Sedative-hypnotic medication was more often used by older patients and patients with a longer duration of CAPD. There was no association between these agents and all-cause mortality in CAPD patients with sleep disorders.

Download Full-text

Adverse Effect of Older Age on the Recurrence of Soft Tissue Sarcoma of the Extremities and Trunk

Journal of Clinical Oncology ◽

10.1200/jco.2010.34.0711 ◽

2011 ◽

Vol 29 (30) ◽

pp. 4029-4035 ◽

Cited By ~ 32

Author(s):

David J. Biau ◽

Peter C. Ferguson ◽

Robert E. Turcotte ◽

Peter Chung ◽

Marc H. Isler ◽

...

Keyword(s):

Soft Tissue ◽

Local Recurrence ◽

Soft Tissue Sarcoma ◽

Older Patients ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Cox Proportional Hazards Models ◽

Hazard Ratios ◽

Age At Surgery ◽

Effect Of Age

Purpose To examine the effect of age on the recurrence of soft tissue sarcoma in the extremities and trunk. Patients and Methods This was a multicenter study that included 2,385 patients with median age at surgery of 57 years. The end points considered were local recurrence and metastasis. Cox proportional hazards models were used to estimate hazard ratios across the age ranges with and without adjustment for known confounding factors. Results Older patients presented with tumors that were larger (P < .001) and of higher grade (P < .001). The proportion of positive margins increased significantly as patients age (P < .001), but radiation therapy was relatively underused in patients older than age 60 years. The 5-year cumulative incidences of local recurrence were 7.2% (95% CI, 4% to 11.7%) for patients age 30 years or younger and 12.9% (95% CI, 9.1% to 17.5%) for patients age 75 years or older. The corresponding 5-year cumulative incidences of metastasis were 17.5% (95% CI, 12.1% to 23.7%) and 33.9% (95% CI, 28.1% to 39.8%) for the same groups. Regression models showed that age was significantly associated with local recurrence (P < .001) and metastasis (P < .001) in nonadjusted models. After adjusting for imbalance in presentation and treatment variables, age remained significantly associated with local recurrence (P = .031) and metastasis (P = .019). Conclusion Older patients have worse outcomes because they tend to present with worse tumors and are treated less aggressively. However, there remained a significant increase in the risk of both local and systemic recurrence associated with increasing age that could not be explained by tumor or treatment characteristics.

Download Full-text

Barriers to Hospice Care Among Older Patients Dying With Lung and Colorectal Cancer

Journal of Clinical Oncology ◽

10.1200/jco.2003.06.142 ◽

2003 ◽

Vol 21 (4) ◽

pp. 728-735 ◽

Cited By ~ 65

Author(s):

Ellen P. McCarthy ◽

Risa B. Burns ◽

Roger B. Davis ◽

Russell S. Phillips

Keyword(s):

Colorectal Cancer ◽

Lung Cancer ◽

Rural Communities ◽

Length Of Stay ◽

Proportional Hazards ◽

Hospice Care ◽

Cox Proportional Hazards ◽

Fee For Service ◽

Specific Patient ◽

Factors Associated

Purpose: To identify factors associated with hospice enrollment and length of stay in hospice among patients dying with lung or colorectal cancer. Methods: We used the Linked Medicare-Tumor Registry Database to conduct a retrospective analysis of the last year of life among Medicare beneficiaries diagnosed with lung or colorectal cancer at age ≥ 66 years between January 1, 1973, and December 31, 1996, in the Surveillance, Epidemiology, and End Results Program who died between January 1, 1988, and December 31, 1998. Our outcomes of interest were time from cancer diagnosis to hospice enrollment and length of stay in hospice care. We used Cox proportional hazards regression to adjust for demographic and clinical information. Results: We studied elderly patients dying with lung cancer (n = 62,117) or colorectal cancer (n = 57,260). Overall, 27% of patients (n = 16,750) with lung cancer and 20% of patients (n = 11,332) with colorectal cancer received hospice care before death. Median length of stay for hospice patients with lung and colorectal cancer was 25 and 28 days, respectively. Overall, 20% of patients entered hospice within 1 week of death, whereas 6% entered more than 6 months before death. Factors associated with later hospice enrollment include being male; being of nonwhite, nonblack race; having fee-for-service insurance; and residing in a rural community. Many of these factors also were associated with shorter stays in hospice. Conclusion: Although use of hospice care has increased dramatically over time, specific patient groups, including men, patients residing in rural communities, and patients with fee-for-service insurance continue to experience delays in hospice enrollment.

Download Full-text

Predictors and Complications of Post Kidney Transplant Leukopenia

Progress in Transplantation ◽

10.1177/15269248211024614 ◽

2021 ◽

pp. 152692482110246

Author(s):

Amanda Vinson ◽

Alyne Teixeira ◽

Bryce Kiberd ◽

Karthik Tennankore

Keyword(s):

Kidney Transplantation ◽

Kidney Transplant ◽

Proportional Hazards ◽

Cox Regression ◽

Selection Process ◽

Cox Proportional Hazards ◽

Post Transplant ◽

Factors Associated ◽

Infectious Risk ◽

Cox Proportional Hazards Models

Background: Leukopenia occurs frequently following kidney transplantation and is associated with adverse clinical outcomes including increased infectious risk. In this study we sought to characterize the causes and complications of leukopenia following kidney transplantation. Methods: In a cohort of adult patients (≥18 years) who underwent kidney transplant from Jan 2006-Dec 2017, we used univariable Cox proportional Hazards models to identify predictors of post-transplant leukopenia (WBC < 3500 mm3). Factors associated with post-transplant leukopenia were then included in a multivariable backwards stepwise selection process to create a prediction model for the outcome of interest. Cox regression analyses were subsequently used to determine if post-transplant leukopenia was associated with complications. Results: Of 388 recipients, 152 (39%) developed posttransplant leukopenia. Factors associated with leukopenia included antithymocyte globulin as induction therapy (HR 3.32, 95% CI 2.25-4.91), valganciclovir (HR 1.84, 95% CI 1.25-2.70), tacrolimus (HR 3.05, 95% CI 1.08-8.55), prior blood transfusion (HR 1.17 per unit, 95% CI 1.09- 1.25), and donor age (HR 1.02 per year, 95% CI 1.00-1.03). Cytomegalovirus infection occurred in 26 patients with leukopenia (17.1%). Other than cytomegalovirus, leukopenia was not associated with posttransplant complications. Conclusion: Leukopenia commonly occurred posttransplant and was associated with modifiable and non-modifiable pretransplant factors.

Download Full-text

Pathophysiology of Patients Over the Age of 65 with Acute Myeloid Leukemia (AML).

Blood ◽

10.1182/blood.v114.22.992.992 ◽

2009 ◽

Vol 114 (22) ◽

pp. 992-992

Author(s):

Henna Malik ◽

Lucas Wong ◽

Sarju Waghela ◽

Lisa Go ◽

William Koss ◽

...

Keyword(s):

Elderly Patients ◽

Older Patients ◽

Myeloid Leukemia ◽

Proportional Hazards ◽

Leukocyte Count ◽

Cox Proportional Hazards ◽

Multivariate Model ◽

Biological Features ◽

Kaplan Meier ◽

Cox Proportional Hazards Models

Abstract Abstract 992 Poster Board I-14 Background: Acute myeloid leukemia (AML) is a common disease in individuals ≥ 65 years old. Overall survival (OS) is significantly shorter in older patients compared with younger patients. Many patients do not receive chemotherapy due to age or co-morbidities. The aim of our study is to investigate the biologic characteristics of AML in the elderly using variables on survival. Methods: For this single-center, retrospective study, authors analyzed the following variables for the outcome patients ≥ 65 years old: age at diagnosis, gender, WBC, HGB, LDH, % blasts, risk factors, chemotherapy, co-morbidities, cytogenetics, and documented MDS/cancer. Statistical Analysis: All variables were summarized using descriptive statistics: mean (SD) for continuous variables and frequency (percent) for categorical variables. Kaplan-Meier survival curves were obtained, and univariate Cox proportional hazards models and multivariate Cox proportional hazards models were applied. A p-value of less than 0.05 indicated a statistical significance. SAS 9.1.3 (SAS Institute INC, Cary, NC) was used for data management and statistical analysis. Results: Seventy-four patients 65 or older were included for the analysis by Kaplan – Meier survival. The median survival time was 3.8 months. Seventy patients have died and 4 have survived until 1/2009. Patients over 80 years old had the worst survival, 0.7 month, compare to age 65 – 70 group which was 4 months and 71 – 80 group which was 4.6 months. Results with univariate Cox proportional hazards model shows WBC group (p=0.0390), LDH group (p=0.0153), and chemotherapy (p<0.0001) were significant variables on survival. LDH group and cytogenetics were not included in the multivariate model due to many missing measurements (43%; 32 out of 74) and (27%; 17 out of 74), respectively. Final multivariate model including all significant variables revealed significant effect of WBC group (p=0.0074) and chemotherapy (p<0.0001) on survival. Discussion and Conclusions: Prior results from clinical trials and single-center studies evaluating the prognostic factors in older patients are conflicting. In our study, patients who received chemotherapy (standard or intensive chemotherapy) had better survival (median 5.2 months) compare to untreated patients with median survival 0.4 months (p<0.0001). The tendency is to exclude elderly patients for the treatment because of poor performance status (PS), organ dysfunction, and co-morbid conditions. The approach to withhold chemotherapy in elderly patients is not supported by our results. To the contrary, it appears that chemotherapy should be pursued and may offer longer survival except for elderly patients over the age of 80. High WBC ' 10 × 103 /μL at presentation had an adverse impact on survival rates (p=0.034). Other studies have shown mixed results in regards to survival. LDH > 300 U/L was an adverse prognostic factor for survival. A higher leukocyte count probably is representative of high tumor burden and more aggressive disease biology. Cytogenetics (with MDS and without MDS) at diagnosis was not predictive of survival but our cytogenetic evaluation was incomplete due to missing data. Co-morbidities such as cardiovascular disease, diabetes, hepatic disease, pulmonary disease, and cancer did not impact on the survival. We observed adverse impact of increasing age on survival only in patients over 80. Some investigators reported no effects, and others showed increasing age as a poor prognostic factor for both CR and survival or survival alone. The cause of this discrepancy is not clear. Patients > 80 years comprise 28% of our study group and exhibited the worst survival time; they may represent a different patient population with distinct biological features. We conclude that age, biological features, chemotherapy and leukocyte count are the most important determinants of survival. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Analysis of factors associated with changing general practice in the first 14 years of life in Wales using linked cohort and primary care records: implications for using primary care databanks for life-course research

International Journal for Population Data Science ◽

10.23889/ijpds.v3i2.477 ◽

2018 ◽

Vol 3 (2) ◽

Author(s):

Carol Dezateux ◽

Lucy J Griffiths ◽

Bianca L De Stavola ◽

Ashley Akbari ◽

Amrita Bandyopadhyay ◽

...

Keyword(s):

Primary Care ◽

General Practice ◽

Life Course ◽

Rural Areas ◽

Early Life ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Factors Associated ◽

Changing Practice ◽

Cox Proportional Hazards Models

Background Databanks of primary care electronic health records (pcEHRs) are a valuable resource for life course research, however loss to follow up due to changing general practice has received little attention. ObjectiveWe investigated factors associated with changing general practice (GP) in early life and continuity of participation in the Secure Anonymised Information Linkage (SAIL) databank, to which approximately 80% of Welsh practices contribute. MethodsWe analysed linked pcEHRs for 1834 (882 girls) Millennium Cohort Study participants, resident in Wales, with consent to health record linkage. We studied time from first to next practice registration using Cox proportional hazards models, and estimated mutually-adjusted hazard ratios (aHRs) for child, household and practice factors. FindingsThere were 3065 Welsh GP registrations for 1834 children. By age 5 years, 25% changed practice at least once, with 1070 (58.3%), 477 (26.0%), 287 (15.7%) registered with 1, 2, or 3+ GPs respectively by age 14 years. Changing practice was related to maternal age (aHRs; 95% CI: 0.96; 0.95,0.98), living in rural areas (0.75;0.56,0.99), initial registration with a non-SAIL-practice (2.16;1.60,2.93), recent address change (1.62;1.21,2.16), and no maternal educational qualifications (1.40;1.15,1.71). Overall, 305 (16.6%) children had never registered with a SAIL practice. Of 403 children initially registered with a SAIL practice who then changed practice, 66.7% re-registered with a SAIL practice. ConclusionsIn a nationally representative sample of Welsh children, the majority remained registered with the same practice up to age 14 years, with change in practice varying by maternal and household factors. Continuity of participation in the Welsh SAIL databank over early life is high, reflecting the high proportion of practices contributing, and the high proportion of children registered with them. Geographically contiguous primary care databanks, such as SAIL, enable a high proportion of children to be followed over time despite changing general practice.

Download Full-text

442. Sex-Related Differences in Mortality from COVID-19: Survival Analysis of Patients from an Urban Hospital

Open Forum Infectious Diseases ◽

10.1093/ofid/ofab466.641 ◽

2021 ◽

Vol 8 (Supplement_1) ◽

pp. S323-S323

Author(s):

Mamta Sharma ◽

Susan M Szpunar ◽

Ashish Bhargava ◽

Leonard B Johnson ◽

Louis Saravolatz

Keyword(s):

Risk Factors ◽

Survival Analysis ◽

Proportional Hazards ◽

Proportional Hazards Model ◽

Cox Proportional Hazards ◽

Cox Proportional Hazards Model ◽

Factors Associated ◽

Hazards Model ◽

Cox Proportional Hazards Models ◽

Failure Assessment

Abstract Background Mortality from COVID-19 is associated with male sex, older age, black race, and comorbidities including obesity. Our study identified risk factors for in-hospital mortality from COVID-19 using survival analysis at an urban center in Detroit, MI. Methods This was a single-center historical cohort study. We reviewed the electronic medical records of patients positive for severe acute respiratory syndrome coronavirus 2 (the COVID-19 virus) on qualitative polymerase-chain-reaction assay, who were admitted between 3/8-6/14/20. We assessed risk factors for mortality using Kaplan-Meier analysis and Cox proportional hazards models. Results We included 565 patients with mean age (standard deviation) 64.4 (16.2) years, 52.0% male (294) and 77.2% (436) black/African American. The overall mean body mass index (BMI) was 32.0 (9.02) kg/m2. At least one comorbidity was present in 95.2% (538) of patients. The overall case-fatality rate was 30.4% (172/565). The unadjusted mortality rate among males was 33.7% compared to 26.9% in females (p=0.08); the median time to death (range) for males was 16.8 (0.3, 33.9) compared to 14.2 (0.32, 47.7) days for females (p=0.04). Univariable survival analysis with Cox proportional hazards models revealed that age (p=< 0.0001), admission from a facility (p=0.002), public insurance (p< 0.0001), respiratory rate ≥ 22 bpm (p=0.02), lymphocytopenia (p=0.07) and serum albumin (p=0.007) were additional risk factors for mortality (Table 1). From multivariable Cox proportional hazards modeling (Table 2), after controlling for age, Charlson score and qSofa, males were 40% more likely to die than females (p=0.03). Table 1. Univariate analysis with Cox proportional hazards model on factors associated with mortality in patients with COVID-19 Abbreviations: HR: Hazard ratio, CI: Confidence interval Table 2. Multivariable analysis with Cox proportional hazards model on factors associated with mortality in patients with COVID-19 Abbreviations: HR: Hazard ratio, CI: Confidence interval, CWIC: Charlson weighted index of comorbidity, qSOFA: Quick sepsis related organ failure assessment Conclusion After controlling for risk factors for mortality including age, comorbidity and sepsis-related organ failure assessment, males continued to have a higher hazard of death. These demographic and clinical factors may help healthcare providers identify risk factors from COVID-19. Disclosures All Authors: No reported disclosures

Download Full-text

Triple class experience after initiation of combination antiretroviral treatment in Australia: survival and projections

Sexual Health ◽

10.1071/sh10008 ◽

2011 ◽

Vol 8 (3) ◽

pp. 295 ◽

Cited By ~ 2

Author(s):

Sadaf Marashi Pour ◽

Ian Woolley ◽

Peter Canavan ◽

John Chuah ◽

Darren B. Russell ◽

...

Keyword(s):

Antiretroviral Treatment ◽

Proportional Hazards ◽

Treatment Options ◽

Cox Proportional Hazards ◽

New Drugs ◽

Factors Associated ◽

Current Trends ◽

Cox Proportional Hazards Models ◽

Number Of Patients ◽

Significant Difference

Background Patients who have become triple class experienced (TCE) are at a high risk of exhausting available treatment options. This study aims to investigate factors associated with becoming TCE and to explore the effect of becoming TCE on survival. We also project the prevalence of TCE in Australia to 2012. Methods: Patients were defined as TCE when they stopped a combination antiretroviral treatment (cART) that introduced the third of the three major antiretroviral classes. Cox proportional hazards models were used to investigate factors associated with TCE and the effect of TCE on survival. To project TCE prevalence, we used predicted rates of TCE by fitting a Poisson regression model, together with the estimated number of patients who started cART in each year in Australia, assuming a mortality rate of 1.5 per 100 person-years. Results: Of the 1498 eligible patients, 526 became TCE. Independent predictors of a higher risk of TCE included current CD4 counts below 200 cells μL–1 and earlier calendar periods. No significant difference in survival was observed between those who were TCE and those who were not yet TCE. An increasing number of patients are using cART in Australia and if current trends continue, the number of patients who are TCE is estimated to increase from 2800 in 2003 to 5000 in 2012. Conclusion: Our results suggest that the prevalence of TCE in Australia is estimated to plateau after 2003. However, as an increasing number of patients are becoming TCE, it is necessary to develop new drugs that come from new classes or do not have overlapping resistance.

Download Full-text

Nomogram to Predict the Benefit of Radiation for Older Patients With Breast Cancer Treated With Conservative Surgery

Journal of Clinical Oncology ◽

10.1200/jco.2011.41.0076 ◽

2012 ◽

Vol 30 (23) ◽

pp. 2837-2843 ◽

Cited By ~ 47

Author(s):

Jeffrey M. Albert ◽

Diane D. Liu ◽

Yu Shen ◽

I-Wen Pan ◽

Ya-Chen Tina Shih ◽

...

Keyword(s):

Breast Cancer ◽

Older Women ◽

Older Patients ◽

Clinical Decision Making ◽

Proportional Hazards ◽

Estrogen Receptor Status ◽

Clinical Decision ◽

Conservative Surgery ◽

Cox Proportional Hazards ◽

Cox Proportional Hazards Models

Purpose The role of radiation therapy (RT) after conservative surgery (CS) remains controversial for older patients with breast cancer. Guidelines based on recent clinical trials have suggested that RT may be omitted in selected patients with favorable disease. However, it is not known whether this recommendation should extend to other older women. Accordingly, we developed a nomogram to predict the likelihood of long-term breast preservation with and without RT. Methods We used Surveillance, Epidemiology, and End Results–Medicare data to identify 16,092 women age 66 to 79 years treated with CS between 1992 and 2002, using claims to identify receipt of RT and subsequent mastectomy. Time to mastectomy was estimated using the Kaplan-Meier method. Cox proportional hazards models determined the effect of covariates on mastectomy-free survival (MFS). A nomogram was developed to predict 5- and 10-year MFS, given associated risk factors, and bootstrap validation was performed. Results With a median follow-up of 7.2 years, the overall 5- and 10-year MFS rates were 98.1% (95% CI, 97.8% to 98.3%) and 95.4% (95% CI, 94.9% to 95.8%), respectively. In multivariate analysis, age, race, tumor size, estrogen receptor status, and receipt of RT were predictive of time to mastectomy and were incorporated into the nomogram. Nodal status was also included given a significant interaction with RT. The resulting nomogram demonstrated good accuracy in predicting MFS, with a bootstrap-corrected concordance index of 0.66. Conclusion This clinically useful tool predicts 5- and 10-year MFS among older women with early breast cancer using readily available clinicopathologic factors and can aid individualized clinical decision making by estimating predicted benefit from RT.

Download Full-text

A semi-supervised approach for rapidly creating clinical biomarker phenotypes in the UK Biobank using different primary care EHR and clinical terminology systems

JAMIA Open ◽

10.1093/jamiaopen/ooaa047 ◽

2020 ◽

Author(s):

Spiros Denaxas ◽

Anoop D Shah ◽

Bilal A Mateen ◽

Valerie Kuan ◽

Jennifer K Quint ◽

...

Keyword(s):

Primary Care ◽

Proportional Hazards ◽

Biomarker Discovery ◽

Data Extraction ◽

Cox Proportional Hazards ◽

Research Quality ◽

Uk Biobank ◽

Expert Review ◽

Cox Proportional Hazards Models ◽

The Uk

Abstract Objectives The UK Biobank (UKB) is making primary care electronic health records (EHRs) for 500 000 participants available for COVID-19-related research. Data are extracted from four sources, recorded using five clinical terminologies and stored in different schemas. The aims of our research were to: (a) develop a semi-supervised approach for bootstrapping EHR phenotyping algorithms in UKB EHR, and (b) to evaluate our approach by implementing and evaluating phenotypes for 31 common biomarkers. Materials and Methods We describe an algorithmic approach to phenotyping biomarkers in primary care EHR involving (a) bootstrapping definitions using existing phenotypes, (b) excluding generic, rare, or semantically distant terms, (c) forward-mapping terminology terms, (d) expert review, and (e) data extraction. We evaluated the phenotypes by assessing the ability to reproduce known epidemiological associations with all-cause mortality using Cox proportional hazards models. Results We created and evaluated phenotyping algorithms for 31 biomarkers many of which are directly related to COVID-19 complications, for example diabetes, cardiovascular disease, respiratory disease. Our algorithm identified 1651 Read v2 and Clinical Terms Version 3 terms and automatically excluded 1228 terms. Clinical review excluded 103 terms and included 44 terms, resulting in 364 terms for data extraction (sensitivity 0.89, specificity 0.92). We extracted 38 190 682 events and identified 220 978 participants with at least one biomarker measured. Discussion and conclusion Bootstrapping phenotyping algorithms from similar EHR can potentially address pre-existing methodological concerns that undermine the outputs of biomarker discovery pipelines and provide research-quality phenotyping algorithms.

Download Full-text