Inappropriate use of bivariable analysis to screen risk factors for use in multivariable analysis

Sun, Guo-Wen; Shook, Thomas; Kay, Gregory L.

doi:10.1016/0895-4356(96)00025-x

Cited by 789 publications

(580 citation statements)

References 57 publications

Supporting

Mentioning

545

Contrasting

Unclassified

Order By: Relevance

“…However, it is not simple to select predictors only based on statistical significance during model development, as it is crucial to retain those risk predictors known to be important from literatures, but which may not reach statistical significance in a data set. This has been shown to be inappropriate, as it can wrongly reject potentially important variables when the relationship between an outcome and a risk factor is confounded by any confounder and when this confounder is not properly controlled, thus leading to an unreliable model [30,33]. Five studies in this review reduced the initial number of candidate risk predictor prior to the final model; however, two studies failed to provide sufficient detail on how the procedure was carried out.…”

Section: Discussion Main Findingsmentioning

confidence: 98%

Developing Risk Prediction Models for Postoperative Pancreatic Fistula: a Systematic Review of Methodology and Reporting Quality

Zhang

Guo

et al. 2016

Indian J Surg

View full text Add to dashboard Cite

Postoperative pancreatic fistula is still a major complication after pancreatic surgery, despite improvements of surgical technique and perioperative management. We sought to systematically review and critically access the conduct and reporting of methods used to develop risk prediction models for predicting postoperative pancreatic fistula. We conducted a systematic search of PubMed and EMBASE databases to identify articles published before January 1, 2015, which described the development of models to predict the risk of postoperative pancreatic fistula. We extracted information of developing a prediction model including study design, sample size and number of events, definition of postoperative pancreatic fistula, risk predictor selection, missing data, modelbuilding strategies, and model performance. Seven studies of developing seven risk prediction models were included. In three studies (42 %), the number of events per variable was less than 10. The number of candidate risk predictors ranged from 9 to 32. Five studies (71 %) reported using univariate screening, which was not recommended in building a multivariate model, to reduce the number of risk predictors. Six risk prediction models (86 %) were developed by categorizing all continuous risk predictors. The treatment and handling of missing data were not mentioned in all studies. We found use of inappropriate methods that could endanger the development of model, including univariate pre-screening of variables, categorization of continuous risk predictors, and model validation. The use of inappropriate methods affects the reliability and the accuracy of the probability estimates of predicting postoperative pancreatic fistula.

show abstract

Section: Discussion Main Findingsmentioning

confidence: 98%

Developing Risk Prediction Models for Postoperative Pancreatic Fistula: a Systematic Review of Methodology and Reporting Quality

Zhang

Guo

et al. 2016

Indian J Surg

View full text Add to dashboard Cite

show abstract

“…We have introduced in the multivariate, logistic regression analysis all the variables that are plausibly important based on theory, even if the p-value was inferior to 0.20 in the univariate analysis. 32 We tested a second model (Model 2), including, in addition to the previous variables, our biologic tested markers, h-FABP and IMA, if they had revealed a clear interest as diagnostic markers of ACS in the previous univariate analysis. Odds ratios (ORs) were reported with 95% CIs.…”

Section: Discussionmentioning

confidence: 99%

Clinical Assessment of Ischemia‐modified Albumin and Heart Fatty Acid–binding Protein in the Early Diagnosis of Non‐ST‐elevation Acute Coronary Syndrome in the Emergency Department

Charpentier

Ducassé

Cournot

et al. 2010

Academic Emergency Medicine

View full text Add to dashboard Cite

Objectives: Heart fatty acid-binding protein (h-FABP) and ischemia-modified albumin (IMA) have recently been evaluated, but to the best of our knowledge, no study has reported an analysis of these two markers for the detection of early myocardial infarction and myocardial ischemia in a large cohort of consecutive patients presenting to an emergency department (ED). This study evaluates the diagnostic accuracy and the clinical utility of h-FABP and IMA for non-ST-segment elevation acute coronary syndrome (ACS) diagnosis in the first hour of management in an ED.Methods: In a prospective 11-month study, 677 patients admitted to the ED with chest pain and suspected non-ST-segment elevation ACS were enrolled. On presentation, blood samples were obtained for the measurement of the biomarkers h-FABP (immunodetection with CardioDetect) and IMA (albumin cobalt-binding test). Two physicians, blinded to the results of the markers, independently categorized patients as having or not having non-ST-segment elevation ACS.Results: Of the 677 patients who were prospectively recruited, non-ST-segment elevation ACS was diagnosed in 185 (27.3%). While IMA was not predictive of the ACS diagnosis (odds ratio [OR] = 1.23; 95% CI = 0.87 to 1.81), h-FABP was predictive of ACS diagnosis (OR = 4.65; 95% CI = 2.39 to 9.04) with specificity at 96.8% (95% CI = 95.4% to 98.1%) and sensitivity at 13.5% (95% CI = 10.9% to 16.1%). However, h-FABP did not add significant additional information to a predictive model that included the usual diagnostic tools for non-ST-elevation ACS management (p = 0.40). Conclusions:In this study on a large cohort of patients admitted to an ED for chest pain, IMA and h-FABP did not provide valuable information for ACS diagnosis.

show abstract

“…First, all covariates with P 0.10 in the univariate analyses were considered for inclusion in the multivariable model. Second, backward selection procedure with (P < 0.05) was used to choose the covariates in the final multivariable model (22). Covariates eliminated were reentered in the final multivariable model, with adjustment for the remaining significant covariates to ensure that no omitted covariate significantly reduced the log likelihood c 2 of the model (23).…”

Section: Statistical Analysesmentioning

confidence: 99%

LLPi: Liverpool Lung Project Risk Prediction Model for Lung Cancer Incidence

Marcus

Chen

Raji

et al. 2015

Cancer Prevention Research

View full text Add to dashboard Cite

Identification of high-risk individuals will facilitate early diagnosis, reduce overall costs, and also improve the current poor survival from lung cancer. The Liverpool Lung Project prospective cohort of 8,760 participants ages 45 to 79 years, recruited between 1998 and 2008, was followed annually through the hospital episode statistics until January 31, 2013. Cox proportional hazards models were used to identify risk predictors of lung cancer incidence. C-statistic was used to assess the discriminatory accuracy of the models. Models were internally validated using the bootstrap method. During mean follow-up of 8.7 years, 237 participants developed lung cancer. Age [hazard ratio (HR), 1.04; 95% confidence interval (CI), 1.02-1.06], male gender (HR, 1.48; 95% CI, 1.10-1.98), smoking duration (HR, 1.04; 95% CI, 1.03-1.05), chronic obstructive pulmonary disease (HR, 2.43; 95% CI, 1.79-3.30), prior diagnosis of malignant tumor (HR, 2.84; 95% CI, 2.08-3.89), and early onset of family history of lung cancer (HR, 1.68; 95% CI, 1.04-2.72) were associated with the incidence of lung cancer. The LLPi risk model had a good calibration (goodness-of-fit c 2 7.58, P ¼ 0.371). The apparent C-statistic was 0.852 (95% CI, 0.831-0.873) and the optimism-corrected bootstrap resampling C-statistic was 0.849 (95% CI, 0.829-0.873). The LLPi risk model may assist in identifying individuals at high risk of developing lung cancer in population-based screening programs. Cancer Prev Res; 8(6); 570-5. Ó2015 AACR.

show abstract

Inappropriate use of bivariable analysis to screen risk factors for use in multivariable analysis

Cited by 789 publications

References 57 publications

Developing Risk Prediction Models for Postoperative Pancreatic Fistula: a Systematic Review of Methodology and Reporting Quality

Developing Risk Prediction Models for Postoperative Pancreatic Fistula: a Systematic Review of Methodology and Reporting Quality

Clinical Assessment of Ischemia‐modified Albumin and Heart Fatty Acid–binding Protein in the Early Diagnosis of Non‐ST‐elevation Acute Coronary Syndrome in the Emergency Department

LLPi: Liverpool Lung Project Risk Prediction Model for Lung Cancer Incidence

Contact Info

Product

Resources

About