Using the hazard ratio to evaluate allowable total error in predictive measurands

2016 ◽  
Vol 0 (0) ◽  
Author(s):  
Arne Åsberg ◽  
Ingrid Hov Odsæter ◽  
Gustav Mikkelsen ◽  
Gunhild Garmo Hov
Keyword(s):  
Measurement Errors ◽  
Cox Regression ◽  
State Of The Art ◽  
Total Error ◽  
Biological Variation ◽  
Hazard Ratio ◽  
Published Data ◽  
Acceptable Error ◽  
Albumin Creatinine Ratio ◽  
Urine Albumin

AbstractAllowable total error is usually derived from data on biological variation or from state-of-the-art of measuring technology. Here we present a new principle for evaluating allowable total error when the concentration of the analyte (the measurand) is used for prediction: What are the predictive consequences of allowable total errors in terms of errors in the estimate of the hazard ratio (HR)?We explored the effect of analytical measurement errors on Cox regression estimates of HR. Published data on Cox regression coefficients were used to illustrate the effect of measurement errors on predicting cardiovascular events or death based on serum concentration of cholesterol and on progression of chronic kidney disease to kidney failure based on serum concentrations of albumin, bicarbonate, calcium and phosphate, and urine albumin/creatinine-ratio.If the acceptable error in the estimate of the HR is 10%, allowable total errors in serum cholesterol, bicarbonate and phosphate are approximately the same as allowable total error based on biological variation, while allowable total error in serum albumin and calcium are a little larger than estimates based on biological variation.Evaluating allowable total error from its effect on the estimate of HR is universally applicable to measurands used for prediction.

scholarly journals State of the Art for Measurement of Urine Albumin: Comparison of Routine Measurement Procedures to Isotope Dilution Tandem Mass Spectrometry

2014 ◽  
Vol 60 (3) ◽  
pp. 471-480 ◽  
Author(s):  
Lorin M Bachmann ◽  
Goran Nilsson ◽  
David E Bruns ◽  
Matthew J McQueen ◽  
John C Lieske ◽  
...  
Keyword(s):  
Mass Spectrometry ◽  
Isotope Dilution ◽  
State Of The Art ◽  
Total Error ◽  
Isotope Dilution Mass Spectrometry ◽  
Urine Albumin ◽  
Routine Measurement ◽  
Concentration Interval ◽  
High Concentrations ◽  
Instructions For Use

Abstract BACKGROUND Urine albumin is the primary biomarker for detection and monitoring of kidney damage. Because fixed decision criteria are used to identify patients with increased values, we investigated if commonly used routine measurement procedures gave comparable results. METHODS Results from 17 commercially available urine albumin measurement procedures were investigated vs an isotope dilution mass spectrometry (IDMS) procedure. Nonfrozen aliquots of freshly collected urine from 332 patients with chronic kidney disease, diabetes, cardiovascular disease, and hypertension were distributed to manufacturers to perform urine albumin measurements according to the respective instructions for use for each procedure. Frozen aliquots were used for measurements by the IDMS procedure. An error model was used to determine imprecision and bias components. RESULTS Median differences between the largest positive and negative biases vs IDMS were 45%, 37%, and 42% in the concentration intervals of 12–30 mg/L, 31–200 mg/L, and 201–1064 mg/L, respectively. Biases varied with concentration for most procedures and exceeded ±10% over the concentration interval for 14 of 16 quantitative procedures. Mean biases ranged from −35% to 34% at 15 mg/L. Dilution of samples with high concentrations introduced bias for 4 procedures. The combined CV was >10% for 5 procedures. It was not possible to estimate total error due to dependence of bias on concentration. CVs for sample-specific influences were 0% to 15.2%. CONCLUSIONS Bias was the dominant source of disagreement among routine measurement procedures. Consequently, standardization efforts will improve agreement among results. Variation of bias with concentration needs to be addressed by manufacturers.

Falls Predict Acute Hospitalization in Parkinson’s Disease

2021 ◽  
pp. 1-20
Author(s):  
Diego Santos García ◽  
Teresa de Deus Fonticoba ◽  
Carlos Cores ◽  
Ester Suárez Castro ◽  
Jorge Hernández Vara ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Cox Regression ◽  
Independent Predictor ◽  
Hazard Ratio ◽  
Acute Hospitalization ◽  
Baseline Visit ◽  
Kaplan Meier ◽  
Non Motor Symptoms

Background: There is a need for identifying risk factors for hospitalization in Parkinson’s disease (PD) and also interventions to reduce acute hospital admission. Objective: To analyze the frequency, causes, and predictors of acute hospitalization (AH) in PD patients from a Spanish cohort. Methods: PD patients recruited from 35 centers of Spain from the COPPADIS-2015 (COhort of Patients with PArkinson’s DIsease in Spain, 2015) cohort from January 2016 to November 2017, were included in the study. In order to identify predictors of AH, Kaplan-Meier estimates of factors considered as potential predictors were obtained and Cox regression performed on time to hospital encounter 1-year after the baseline visit. Results: Thirty-five out of 605 (5.8%) PD patients (62.5±8.9 years old; 59.8% males) presented an AH during the 1-year follow-up after the baseline visit. Traumatic falls represented the most frequent cause of admission, being 23.7% of all acute hospitalizations. To suffer from motor fluctuations (HR [hazard ratio] 2.461; 95% CI, 1.065–5.678; p = 0.035), a very severe non-motor symptoms burden (HR [hazard ratio] 2.828; 95% CI, 1.319–6.063; p = 0.008), falls (HR 3.966; 95% CI 1.757–8.470; p = 0.001), and dysphagia (HR 2.356; 95% CI 1.124–4.941; p = 0.023) was associated with AH after adjustment to age, gender, disease duration, levodopa equivalent daily dose, total number of non-antiparkinsonian drugs, and UPDRS-IIIOFF. Of the previous variables, only falls (HR 2.998; 95% CI 1.080–8.322; p = 0.035) was an independent predictor of AH. Conclusion: Falls is an independent predictor of AH in PD patients.

scholarly journals Characteristics of ischemic stroke and intracranial hemorrhage in patients with nephrotic syndrome

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Wen-Yi Huang ◽  
Chun-Wei Chang ◽  
Chiung-Mei Chen ◽  
Kuan-Hsing Chen ◽  
Chien-Hung Chang ◽  
...  
Keyword(s):  
Nephrotic Syndrome ◽  
Ischemic Stroke ◽  
Intracranial Hemorrhage ◽  
Cox Regression ◽  
Hazard Ratio ◽  
Chang Gung Memorial Hospital ◽  
Cerebral Stroke ◽  
Research Database ◽  
Anterior Circulation ◽  
Cox Regression Analysis

Abstract Background The incidence of cerebral stroke, including ischemic infarction and intracranial hemorrhage (ICH), increases in patients with nephrotic syndrome (NS). However, the clinical characteristics of patients with NS and stroke remain elusive. We aimed to investigate the clinical presentation and prognosis among patients with NS and ischemic stroke (IS) or ICH. Methods We conducted a population-based retrospective cohort study of patients with NS and acute stroke using the Chang Gung Research Database of Taiwan from January 1, 2001, to December 31, 2017. The participants were recruited from the 7 branches of Chang Gung Memorial Hospital. Results A total of 233 patients with IS and 57 patients with ICH were enrolled. The median age was 60 (52–70) years. The prevalence rates of hyperlipidemia, hyperuricemia, and smoking were higher in IS than in ICH. IS demonstrated lower white blood cell count (7.80 vs. 8.92 × 109/L) and high-sensitivity C-reactive protein level (33.42 vs. 144.10 nmol/L) and higher cholesterol (5.74 vs. 4.84 mmol/L), triglyceride (1.60 vs. 1.28 mmol/L), and albumin (24 vs. 18 g/L) levels compared with ICH. The dependent functional status and 30-day mortality were higher in ICH than in IS. The risk factors for 30-day mortality for patients with NS and stroke were coronary artery disease (CAD), ICH, and total anterior circulation syndrome. The multivariate Cox regression analysis revealed that CAD was positively associated with 30-day mortality in patients with IS (hazard ratio 24.58, 95 % CI 1.48 to 408.90). In patients with ICH, CAD and subarachnoid hemorrhage were positively associated with 30-day mortality (hazard ratio 5.49, 95 % CI 1.54 to 19.56; hazard ratio 6.32, 95 % CI 1.57 to 25.53, respectively). Conclusions ICH demonstrated a higher risk of dependence and 30-day mortality compared with IS in patients with NS. Intensive monitoring and treatment should be applied particularly in patients with NS and ICH.

Patterns of Care and Survival of Cutaneous Angiosarcoma of the Head and Neck

2020 ◽  
Vol 162 (6) ◽  
pp. 881-887 ◽  
Author(s):  
Clifford Chang ◽  
S. Peter Wu ◽  
Kenneth Hu ◽  
Zujun Li ◽  
David Schreiber ◽  
...  
Keyword(s):  
Head And Neck ◽  
Cox Regression ◽  
Multivariable Analysis ◽  
Multimodality Treatment ◽  
The Elderly ◽  
Hazard Ratio ◽  
Patterns Of Care ◽  
Definitive Treatment ◽  
National Cancer Database ◽  
Curative Intent

Objective To analyze the patterns of care and survival of cutaneous angiosarcomas of the head and neck. Study Design Retrospective cohort study. Setting National Cancer Database. Methods The National Cancer Database was queried to select patients with cutaneous angiosarcoma of the head and neck between 2004 and 2015. For survival analysis, patients were included only if they received definitive treatment and complete data. Prognostic factors were analyzed by univariate and multivariable Cox regression. Results We identified 693 patients diagnosed with head and neck angiosarcomas during the study period. The majority were male (n = 489, 70.6%) and elderly (median, 77 years). A total of 421 patients (60.8%) met the criteria for survival analyses. These patients were treated with surgery and radiation (n = 178, 42.3%), surgery alone (n = 138, 32.8%), triple-modality therapy (n = 48, 11.4%), surgery and chemotherapy (n = 29, 6.9%), and chemoradiation (n = 28, 6.7%). With a median follow-up of 29 months, the 3-year survival was 50.1%. Patients undergoing surgery had better median survival than those who did not (38.1 vs 21.0 months, P = .04). Age, comorbidity, tumor size, and surgical margins were significant factors in univariate analyses. On multivariable analysis, age ≥75 years (hazard ratio, 2.65; 95% CI, 1.80-3.88; P < .001) and positive margins (hazard ratio, 1.91; 95% CI, 1.44-2.51; P < .001) predicted worse overall survival. Conclusion Angiosarcoma of head and neck is a rare malignancy that affects the elderly. Surgical treatment with negative margins is associated with improved survival. Even with curative-intent multimodality treatment, the survival of patients aged ≥75 years is limited.

scholarly journals Performance Criteria for Testosterone Measurements Based on Biological Variation in Adult Males: Recommendations from the Partnership for the Accurate Testing of Hormones

2012 ◽  
Vol 58 (12) ◽  
pp. 1703-1710 ◽  
Author(s):  
Yeo-Min Yun ◽  
Julianne Cook Botelho ◽  
Donald W Chandler ◽  
Alex Katayev ◽  
William L Roberts ◽  
...  
Keyword(s):  
Total Error ◽  
Full Range ◽  
Biological Variation ◽  
Analytical Performance ◽  
Performance Criteria ◽  
Performance Goals ◽  
Adult Males ◽  
Large Variability ◽  
Wide Range ◽  
Variation Data

BACKGROUND Testosterone measurements that are accurate, reliable, and comparable across methodologies are crucial to improving public health. Current US Food and Drug Administration–cleared testosterone assays have important limitations. We sought to develop assay performance requirements on the basis of biological variation that allow physiologic changes to be distinguished from assay analytical errors. METHODS From literature review, the technical advisory subcommittee of the Partnership for the Accurate Testing of Hormones compiled a database of articles regarding analytical and biological variability of testosterone. These data, mostly from direct immunoassay-based methodologies, were used to specify analytical performance goals derived from within- and between-person variability of testosterone. RESULTS The allowable limits of desirable imprecision and bias on the basis of currently available biological variation data were 5.3% and 6.4%, respectively. The total error goal was 16.7%. From recent College of American Pathologists proficiency survey data, most currently available testosterone assays missed these analytical performance goals by wide margins. Data from the recently established CDC Hormone Standardization program showed that although the overall mean bias of selected certified assays was within 6.4%, individual sample measurements could show large variability in terms of precision, bias, and total error. CONCLUSIONS Because accurate measurement of testosterone across a wide range of concentrations [approximately 2–2000 ng/dL (0.069–69.4 nmol/L)] is important, we recommend using available data on biological variation to calculate performance criteria across the full range of expected values. Additional studies should be conducted to obtain biological variation data on testosterone from women and children, and revisions should be made to the analytical goals for these patient populations.

scholarly journals FMRI-based prediction of naltrexone response in alcohol use disorder: a replication study

2021 ◽  
Author(s):  
Patrick Bach ◽  
Georg Weil ◽  
Enrico Pompili ◽  
Sabine Hoffmann ◽  
Derik Hermann ◽  
...  
Keyword(s):  
Alcohol Use ◽  
Pharmacological Treatment ◽  
Alcohol Use Disorder ◽  
Cox Regression ◽  
Small Sample Size ◽  
Cue Reactivity ◽  
Small Sample ◽  
Hazard Ratio ◽  
Number Needed To Treat ◽  
Significant Interaction Effect

AbstractPharmacological treatment in alcohol use disorder suffers from modest effect sizes. Efforts have been undertaken to identify patient characteristics that help to select individuals that benefit from pharmacological treatment. Previous studies indicated that neural alcohol cue-reactivity (CR) might provide a marker that identifies patients, which benefit from naltrexone treatment.We investigated the reproducibility of the association between ventral striatum (VS) activation and naltrexone (NTX) treatment response by analyzing data from a recent longitudinal clinical trial in N = 44 abstinent treatment-seeking alcohol-dependent patients. A follow-up was conducted over 3 months. We computed the percentage of significant voxels in VS and tested main effects and interactions with NTX treatment on relapse risk using Cox Regression models.We found a significant interaction effect between pre-treatment cue reactivity in the VS and NTX treatment on time to first heavy relapse (Hazard Ratio = 7.406, 95% CI 1.17–46.56, p = 0.033), such that the patient group with high VS activation (defined by a mean split) showed a significant medication effect (Hazard Ratio = 0.140, 95% CI 0.02–0.75, p = 0.022) with a number needed to treat of 3.4 [95% CI 2.413.5], while there was no significant effect in the group with low VS activation (Hazard Ratio = 0.726, p = 0.454).Thus, using an independent sample we replicated the previously described positive association between VS activation and NTX efficacy. Although our results should be considered cautiously in light of the small sample size, our results support the potential of neural alcohol CR as a tool for precision medicine approaches in alcohol dependence.

Abstract P142: The CKD273 Urinary Peptidomic Biomarker is Associated with Mortality in People at Early Stages of Diabetic Nephropathy Independent of Traditional Risk Factors

Hypertension ◽  
2017 ◽  
Vol 70 (suppl_1) ◽  
Author(s):  
Gemma E Currie ◽  
Sheon Mary ◽  
Bernt J von Scholten ◽  
Morten Lindhardt ◽  
Harald Mischak ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Cardiovascular Disease ◽  
Diabetic Nephropathy ◽  
Cox Regression ◽  
Outcome Data ◽  
Diagnostic Purpose ◽  
Urine Albumin ◽  
Geometrical Mean ◽  
Traditional Risk Factors

Background: Mortality in type 2 diabetes (T2D) is primarily driven by cardiovascular disease. This is amplified in diabetic nephropathy (DN), even in early ‘pre-clinical’ stages. A urinary peptidomic classifier (CKD273) has been found to predict DN development in advance of detectable microalbuminuria. Whether it is also a determinant of mortality and cardiovascular disease in patients with established albuminuria is unknown. Methods: We studied 155 subjects with T2D, albuminuria (geometrical mean [IQR]: 85 [34;194] mg/24hrs), controlled blood pressure (129±16/74±11 mmHg) and preserved renal function (eGFR 88±17 ml/min/1.73m 2 ). Blood and urine samples were collected for measurement of estimated glomerular filtration rate (eGFR), urine albumin excretion (UAE), N-terminal pro-brain natriuretic peptide (NT-proBNP; ELISA) and urinary proteomics (capillary electrophoresis coupled to mass spectrometry). Computed tomography imaging was performed to assess coronary artery calcium (CAC) score. Outcome data were collected through national disease registries over a 6 year follow up period. Results: CKD273 correlated with UAE (r=0.481, p=<0.001), age (r=0.238, p=0.003), CAC score (r=0.236, p=0.003), NT-proBNP (r=0.190, p=0.018) and eGFR (r=0.265, p=0.001). On multiple regression only UAE (β=0.402, p<0.001) and eGFR (β=-0.184, p=0.039) were statistically significant determinants. Twenty participants died during follow-up. CKD273 was a determinant of mortality (log rank [Mantel-Cox] p=0.004), and retained significance (p=0.050) after adjustment for age, sex, blood pressure, NT-proBNP and CAC score in a Cox regression model. Neither eGFR nor UAE were determinants of mortality in this cohort. Conclusions: A multidimensional biomarker can provide information on outcomes associated with its primary diagnostic purpose. Here we demonstrate that the peptidomics-based classifier CKD273 is associated with mortality in albuminuric people with T2D in even when adjusted for other established cardiovascular and renal biomarkers.

scholarly journals Sauna Bathing and Risk of Psychotic Disorders: A Prospective Cohort Study

2018 ◽  
Vol 27 (6) ◽  
pp. 562-569 ◽  
Author(s):  
Tanjaniina Laukkanen ◽  
Jari A. Laukkanen ◽  
Setor K. Kunutsor
Keyword(s):  
Psychotic Disorders ◽  
Cox Regression ◽  
Well Being ◽  
Population Based ◽  
Hazard Ratio ◽  
Male Population ◽  
Cox Regression Analysis ◽  
Sauna Bathing ◽  
General Male Population ◽  
Independent Association

Objective: Sauna bathing has been suggested to promote mental well-being and relaxation, but the evidence is uncertain with respect to mental disorders. We aimed to assess the association of frequency of sauna bathing with risk of psychosis in the Kuopio Ischemic Heart Disease prospective population-based study. Subjects and Methods: Baseline sauna bathing habits were assessed in 2,138 men aged 42–61 years who had no history of psychotic disorders. Participants were classified into three groups based on the frequency of sauna bathing (once, 2–3, and 4–7 times per week). Results: During a median follow-up of 24.9 years, 203 psychotic disorders were recorded. A total of 537, 1,417, and 184 participants reported having a sauna bath once a week, 2–3 times, and 4–7 times per week, respectively. In Cox regression analysis adjusted for age, compared to men who had 1 sauna session per week, the hazard ratio (95% confidence intervals) of psychosis for 4–7 sauna sessions per week was 0.23 (0.09–0.58). In a multivariable model adjusted for several risk factors and other potential confounders, the corresponding hazard ratio was 0.21 (0.08–0.52). The association was similar after further adjustment for total energy intake, socioeconomic status, physical activity, and C-reactive protein (0.22 [0.09–0.54]) and was unchanged on additional adjustment for duration of a sauna session and temperature of the sauna bath (0.23 [0.09–0.57]). Conclusion: Our study suggests a strong inverse and independent association between frequent sauna bathing and the future risk of psychotic disorders in a general male population.

Sign in / Sign up

Export Citation Format

Share Document