A REVIEW OF GOODNESS OF FIT STATISTICS FOR USE IN THE DEVELOPMENT OF LOGISTIC REGRESSION MODELS1

1982 ◽  
Vol 115 (1) ◽  
pp. 92-106 ◽  
Author(s):  
STANLEY LEMESHOW ◽  
DAVID W. HOSMER
Keyword(s):  
Logistic Regression ◽  
Goodness Of Fit ◽  
Fit Statistics

Cardiovascular Risk Prediction using JBS3 Tool: A Kerala based Study

2020 ◽  
Vol 16 ◽  
Author(s):  
Paulin Paul ◽  
Noel George ◽  
B. Priestly Shan
Keyword(s):  
Logistic Regression ◽  
Cardiovascular Risk ◽  
Risk Prediction ◽  
Goodness Of Fit ◽  
Significant Risk ◽  
Lifetime Risk ◽  
British Society ◽  
Cvd Risk ◽  
South Indian ◽  
Cardiovascular Risk Prediction

Background: The accuracy of Joint British Society calculator3 (JBS3) cardiovascular risk prediction may vary within Indian population, and is not yet studied using south Indian Kerala based population data. Objectives: To evaluate the cardiovascular disease (CV) risk estimation using the traditional CVD risk factors (TRF) in Kerala based population. Methods: This cross sectional study has 977 subjects aged between 30 and 80 years. The traditional CVD risk markers are recorded from the medical archives of clinical locations at Ernakulum district, in Kerala The 10 year risk categories used are low (<7.5%), intermediate (≥7.5% and <20%), and high (≥20%). The lifetime classifications low lifetime (≤39%) and high lifetime (≥40%) are used. The study was evaluated using statistical analysis. Chi-square test was done for dependent and categorical CVD risk variable comparison. Multivariate ordinal logistic regression for 10-year risk model and odds logistic regression analysis for lifetime model was used to identify significant risk variables. Results: The mean age of the study population is 52.56±11.43 years. The risk predictions has 39.1% in low, 25.0% in intermediate, and 35.9% had high 10-year risk. The low lifetime risk had 41.1% and 58.9% is high lifetime risk. Reclassifications to high lifetime are higher from intermediate 10-year risk category. The Hosmer-Lemeshow goodness-of-fit statistics indicates a good model fit. Conclusion: The risk prediction and timely intervention with appropriate therapeutic and lifestyle modification is useful in primary prevention. Avoiding short-term incidences and reclassifications to high lifetime can reduce the CVD mortality rates.

scholarly journals Using Ultrasound and Inflammation to Improve Prediction of Ischemic Stroke: A Secondary Analysis of the Multi-Ethnic Study of Atherosclerosis

2021 ◽  
pp. 37-43
Author(s):  
Hediyeh Baradaran ◽  
Alen Delic ◽  
Ka-Ho Wong ◽  
Nazanin Sheibani ◽  
Matthew Alexander ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Goodness Of Fit ◽  
Primary Outcome ◽  
Stroke Risk ◽  
Secondary Analysis ◽  
Predictor Variables ◽  
Baseline Model ◽  
Fit Statistics ◽  
Ischemic Stroke Risk

Introduction: Current ischemic stroke risk prediction is primarily based on clinical factors, rather than imaging or laboratory markers. We examined the relationship between baseline ultrasound and inflammation measurements and subsequent primary ischemic stroke risk. Methods: In this secondary analysis of the Multi-Ethnic Study of Atherosclerosis (MESA), the primary outcome is the incident ischemic stroke during follow-up. The predictor variables are 9 carotid ultrasound-derived measurements and 6 serum inflammation measurements from the baseline study visit. We fit Cox regression models to the outcome of ischemic stroke. The baseline model included patient age, hypertension, diabetes, total cholesterol, smoking, and systolic blood pressure. Goodness-of-fit statistics were assessed to compare the baseline model to a model with ultrasound and inflammation predictor variables that remained significant when added to the baseline model. Results: We included 5,918 participants. The primary outcome of ischemic stroke was seen in 105 patients with a mean follow-up time of 7.7 years. In the Cox models, we found that carotid distensibility (CD), carotid stenosis (CS), and serum interleukin-6 (IL-6) were associated with incident stroke. Adding tertiles of CD, IL-6, and categories of CS to a baseline model that included traditional clinical vascular risk factors resulted in a better model fit than traditional risk factors alone as indicated by goodness-of-fit statistics. Conclusions: In a multiethnic cohort of patients without cerebrovascular disease at baseline, we found that CD, CS, and IL-6 helped predict the occurrence of primary ischemic stroke. Future research could evaluate if these basic ultrasound and serum measurements have implications for primary prevention efforts or clinical trial inclusion criteria.

scholarly journals Risk factors for peripheral venous catheter failure: A prospective cohort study of 5345 patients

2021 ◽  
pp. 112972982110150
Author(s):  
Ya-mei Chen ◽  
Xiao-wen Fan ◽  
Ming-hong Liu ◽  
Jie Wang ◽  
Yi-qun Yang ◽  
...  
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Cohort Study ◽  
Goodness Of Fit ◽  
Insertion Site ◽  
Healthcare Providers ◽  
Venous Catheter ◽  
Goodness Of Fit Test ◽  
Peripheral Venous Catheter ◽  
Independent Risk Factors

Purpose: The objective of this study was to determine the independent risk factors associated with peripheral venous catheter (PVC) failure and develop a model that can predict PVC failure. Methods: This prospective, multicenter cohort study was carried out in nine tertiary hospitals in Suzhou, China between December 2017 and February 2018. Adult patients undergoing first-time insertion of a PVC were observed from catheter insertion to removal. Logistic regression was used to identify the independent risk factors predicting PVC failure. Results: This study included 5345 patients. The PVC failure rate was 54.05% ( n = 2889/5345), and the most common causes of PVC failure were phlebitis (16.3%) and infiltration/extravasation (13.8%). On multivariate analysis, age (45–59 years: OR, 1.295; 95% CI, 1.074–1.561; 60–74 years: OR, 1.375; 95% CI, 1.143–1.654; ⩾75 years: OR, 1.676; 95% CI, 1.355–2.073); department (surgery OR, 1.229; 95% CI, 1.062–1.423; emergency internal/surgical ward OR, 1.451; 95% CI, 1.082–1.945); history of venous puncture in the last week (OR, 1.298, 95% CI 1.130–1.491); insertion site, number of puncture attempts, irritant fluid infusion, daily infusion time, daily infusion volume, and type of sealing liquid were independent predictors of PVC failure. Receiver operating characteristic curve analysis indicated that a logistic regression model constructed using these variables had moderate accuracy for the prediction of PVC failure (area under the curve, 0.781). The Hosmer-Lemeshow goodness of fit test demonstrated that the model was correctly specified (χ2 = 2.514, p = 0.961). Conclusion: This study should raise awareness among healthcare providers of the risk factors for PVC failure. We recommend that healthcare providers use vascular access device selection tools to select a clinically appropriate device and for the timely detection of complications, and have a list of drugs classified as irritants or vesicants so they can monitor patients receiving fluid infusions containing these drugs more frequently.

scholarly journals Prediction of successful de-cannulation of tracheostomised patients in medical intensive care units

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Chul Park ◽  
Ryoung-Eun Ko ◽  
Jinhee Jung ◽  
Soo Jin Na ◽  
Kyeongman Jeon
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Goodness Of Fit ◽  
Characteristic Curve ◽  
Multivariable Logistic Regression Analysis ◽  
Limited Data ◽  
Underlying Malignancy ◽  
Medical Intensive Care ◽  
Good Calibration

Abstract Background Limited data are available on practical predictors of successful de-cannulation among the patients who undergo tracheostomies. We evaluated factors associated with failed de-cannulations to develop a prediction model that could be easily be used at the time of weaning from MV. Methods In a retrospective cohort of 346 tracheostomised patients managed by a standardized de-cannulation program, multivariable logistic regression analysis identified variables that were independently associated with failed de-cannulation. Based on the logistic regression analysis, the new predictive scoring system for successful de-cannulation, referred to as the DECAN score, was developed and then internally validated. Results The model included age > 67 years, body mass index < 22 kg/m2, underlying malignancy, non-respiratory causes of mechanical ventilation (MV), presence of neurologic disease, vasopressor requirement, and presence of post-tracheostomy pneumonia, presence of delirium. The DECAN score was associated with good calibration (goodness-of-fit, 0.6477) and discrimination outcomes (area under the receiver operating characteristic curve 0.890, 95% CI 0.853–0.921). The optimal cut-off point for the DECAN score for the prediction of the successful de-cannulation was ≤ 5 points, and was associated with the specificities of 84.6% (95% CI 77.7–90.0) and sensitivities of 80.2% (95% CI 73.9–85.5). Conclusions The DECAN score for tracheostomised patients who are successfully weaned from prolonged MV can be computed at the time of weaning to assess the probability of de-cannulation based on readily available variables.

scholarly journals A Comparison Study of Goodness of Fit Tests of Logistic Regression in R: Simulation and Application to Breast Cancer Data

2020 ◽  
pp. 50-59
Author(s):  
El-Housainy A. Rady ◽  
Mohamed R. Abonazel ◽  
Mariam H. Metawe’e
Keyword(s):  
Breast Cancer ◽  
Logistic Regression ◽  
Sample Size ◽  
Null Hypothesis ◽  
Goodness Of Fit ◽  
Quadratic Term ◽  
Breast Cancer Dataset ◽  
Cancer Data ◽  
Interaction Term ◽  
Test Package

Goodness of fit (GOF) tests of logistic regression attempt to find out the suitability of the model to the data. The null hypothesis of all GOF tests is the model fit. R as a free software package has many GOF tests in different packages. A Monte Carlo simulation has been conducted to study two situations; the first, studying the ability of each test, under its default settings, to accept the null hypothesis when the model truly fitted. The second, studying the power of these tests when assumptions of sufficient linear combination of the explanatory variables are violated (by omitting linear covariate term, quadratic term, or interaction term). Moreover, checking whether the same test in different R packages had the same results or not. As the sample size supposed to affect simulation results, so the pattern of change of GOF tests results under different sample sizes as well as different model settings was estimated. All tests accept the null hypothesis (more than 95% of simulation trials) when the model truly fitted except modified Hosmer-Lemeshow test in "LogisticDx" package under all different model settings and Osius and Rojek’s (OsRo) test when the true model had an interaction term between binary and categorical covariates. In addition, le Cessie-van Houwelingen-Copas-Hosmer unweighted sum of squares (CHCH) test gave unexpected different results under different packages. Concerning the power study, all tests had a very low power when a departure of missing covariate existed. Generally, stukel’s test (package ’LogisticDX) and CHCH test (package "RMS") reached a power in detecting a missing quadratic term greater than 80% under lower sample size while OsRo test (package ’LogisticDX’) was better in detecting missing interaction term. Beside the simulation study, we evaluated the performance of GOF tests using the breast cancer dataset.

Two subjunctives or three?

2021 ◽  
Author(s):  
Gustavo Guajardo
Keyword(s):  
Logistic Regression ◽  
Prediction Accuracy ◽  
Goodness Of Fit ◽  
Past Tense ◽  
Regression Analyses ◽  
Binary Opposition ◽  
The Past ◽  
Complement Clauses ◽  
Embedded Clauses ◽  
Peninsular Spanish

Abstract This paper examines the use of the three non-periphrastic subjunctives in Spanish in embedded clauses under obligatory subjunctive predicates in the past tense in three Spanish varieties: Argentinean, Mexican and Peninsular Spanish. By means of random forest and logistic regression analyses, I demonstrate that a grammar where the two “past” subjunctives make up one group, such that the variation can be modeled on a binary opposition between (morphologically) past vs. (morphologically) present, achieves better prediction accuracy and goodness-of-fit parameters than a grammar with a three-way split. The results suggest that, at least in complement clauses of obligatory subjunctive predicates, there appear to be no semantic differences between the two past subjunctives but there are still relatively large differences in how the three subjunctive forms are used across the three Spanish varieties studied.1

Sign in / Sign up

Export Citation Format

Share Document