A simulation study of the number of events per variable in logistic regression analysis

Peduzzi, Peter; Concato, John; Kemper, Elizabeth; Holford, Theodore R.; Feinstein, Alvan R.

doi:10.1016/s0895-4356(96)00236-3

Cited by 6,658 publications

(4,664 citation statements)

References 14 publications

Supporting

Mentioning

4,405

Contrasting

Unclassified

Order By: Relevance

“…The large sample size and the use of multiple imputations for handling missing data confer reasonable statistical power to reliably evaluate the relationship of interest. Even for outcomes with low event rates, like maternal mortality, we had enough data to meet the 10 events per variable rule to avoid over‐fitting the models 10, 11. The use of multiple imputations for handling missing data, and the control that we were able to exert on the influence of many potential confounding factors, adds strength to the validity of our observations; however, as many factors are unknown, unmeasured, or poorly measured, the adjustment for confounding had some deficiencies.…”

Section: Discussionmentioning

confidence: 99%

Interpregnancy interval and perinatal outcomes across Latin America from 1990 to 2009: a large multi‐country study

Mignini

Carroli

Betrán

et al. 2015

BJOG

View full text Add to dashboard Cite

ObjectiveTo determine the relationship of interpregnancy interval with maternal and offspring outcomes.DesignRetrospective study with data from the Perinatal Information System database of the Latin American Centre for Perinatology and Human Development, Uruguay.SettingLatin America, 1990–2009.PopulationA cohort of 894 476 women delivering singleton infants.MethodsDuring 1990–2009 the Perinatal Information System database of the Latin American Centre for Perinatology identified 894 476 women with defined interpregnancy intervals: i.e. the time elapsed between the date of the previous delivery and the first day of the last normal menstrual period for the index pregnancy. Using the interval 12–23 months as the reference category, multiple logistic regression estimated adjusted odds ratios (aORs) with 95% confidence intervals (95% CIs) of the association between various interval lengths and maternal and offspring outcomes.Main outcome measuresMaternal death, pre‐eclampsia, eclampsia, puerperal infection, fetal death, neonatal death, preterm birth, and low birthweight.ResultsIn the reference interval there was 0.05% maternal death, 1.00% postpartum haemorrhage, 2.80% pre‐eclampsia, 0.15% eclampsia, 0.28% puerperal infection, 3.45% fetal death, 0.68% neonatal death, 12.33% preterm birth, and 9.73% low birthweight. Longer intervals had increased odds of pre‐eclampsia (>72 months), fetal death (>108–119 months), and low birthweight (96–107 months). Short intervals of <12 months had increased odds of pre‐eclampsia (aOR 0.80; 95% CI 0.76–0.85), neonatal death (aOR 1.18; 95% CI 1.08–1.28), and preterm birth (aOR 1.16; 95% CI 1.11–1.21). Statistically, the interval had no relationship with maternal death, eclampsia, and puerperal infection.ConclusionsA short interpregnancy interval of <12 months is associated with pre‐eclampsia, neonatal mortality, and preterm birth, but not with other maternal or offspring outcomes. Longer intervals of >72 months are associated with pre‐eclampsia, fetal death, and low birthweight, but not with other maternal or offspring outcomes.Tweetable abstractA short interpregnancy interval of <12 months is associated with neonatal mortality and preterm birth.

show abstract

Section: Discussionmentioning

confidence: 99%

Interpregnancy interval and perinatal outcomes across Latin America from 1990 to 2009: a large multi‐country study

Mignini

Carroli

Betrán

et al. 2015

BJOG

View full text Add to dashboard Cite

show abstract

“…Using the same method, we also imputed values for women with missing information on pre-pregnancy body mass index or their infant’s birth weight (8.6%) in our Swedish validation cohort. On the basis of an estimated 300 venous thromboembolism events during the first six weeks postpartum and 22 candidate predictors in our derivation cohort, we had an effective sample size of 14 venous thromboembolism events per predictor, above the minimum requirement suggested by Peduzzi et al22…”

Section: Methodsmentioning

confidence: 99%

Development and validation of risk prediction model for venous thromboembolism in postpartum women: multinational cohort study

Sultan

West

Grainge

et al. 2016

BMJ

109

132

View full text Add to dashboard Cite

Objective To develop and validate a risk prediction model for venous thromboembolism in the first six weeks after delivery (early postpartum).Design Cohort study.Setting Records from England based Clinical Practice Research Datalink (CPRD) linked to Hospital Episode Statistics (HES) and data from Sweden based registry.Participants All pregnant women registered with CPRD-HES linked data between 1997 and 2014 and Swedish medical birth registry between 2005 and 2011 with postpartum follow-up.Main outcome measure Multivariable logistic regression analysis was used to develop a risk prediction model for postpartum venous thromboembolism based on the English data, which was externally validated in the Swedish data.Results 433 353 deliveries were identified in the English cohort and 662 387 in the Swedish cohort. The absolute rate of venous thromboembolism was 7.2 per 10 000 deliveries in the English cohort and 7.9 per 10 000 in the Swedish cohort. Emergency caesarean delivery, stillbirth, varicose veins, pre-eclampsia/eclampsia, postpartum infection, and comorbidities were the strongest predictors of venous thromboembolism in the final multivariable model. Discrimination of the model was similar in both cohorts, with a C statistic above 0.70, with excellent calibration of observed and predicted risks. The model identified more venous thromboembolism events than the existing national English (sensitivity 68% v 63%) and Swedish guidelines (30% v 21%) at similar thresholds.Conclusion A new prediction model that quantifies absolute risk of postpartum venous thromboembolism has been developed and externally validated. It is based on clinical variables that are available in many developed countries at the point of delivery and could serve as the basis for real time decisions on obstetric thromboprophylaxis.

show abstract

“…The total number of covariates was constrained so that, at most, it is equal to one-tenth of the number of patients with delirium. 24 A simple backward model selection procedure was performed where non-significant covariates (Wald test, P [ 0.3) were removed from the model one by one. The overall performance and fitting of the model was evaluated at every step.…”

Section: Multivariable Analysismentioning

confidence: 99%

Delirium following vascular surgery: increased incidence with preoperative β-blocker administration

Katznelson

Djaiani

Mitsakakis

et al. 2009

Can J Anesth/J Can Anesth

View full text Add to dashboard Cite

Purpose To determine if there is an association between perioperative administration of beta-blockers and postoperative delirium in patients undergoing vascular surgery. Methods After Institutional Review Board approval, data were retrospectively collected on patients who underwent vascular surgery in an academic hospital during the period January 2006 to January 2007. Patients with preoperative altered level of consciousness, carotid endarterectomy, or discharge within 24 h of surgery were excluded from the study. Identification of delirium was based on evaluation of the level of consciousness with the NEECHAM Confusion Scale and/or a chart-based instrument for delirium. Multivariable logistic regression analysis was used to identify independent perioperative predictors of postoperative delirium. Beta-blockers were tested for a potential effect. Conclusions Preoperative administration of beta-blockers is associated with an increased risk of postoperative delirium after vascular surgery. Conversely, preoperative statin administration is associated with a lower risk of postoperative delirium. A randomized prospective controlled trial is required to validate these findings. RésuméObjectif De´terminer s'il existe un lien entre l'administration pe´riope´ratoire de beˆta-bloquants et le delirium postope´ratoire chez les patients subissant une chirurgie vasculaire.

show abstract

A simulation study of the number of events per variable in logistic regression analysis

Cited by 6,658 publications

References 14 publications

Interpregnancy interval and perinatal outcomes across Latin America from 1990 to 2009: a large multi‐country study

Interpregnancy interval and perinatal outcomes across Latin America from 1990 to 2009: a large multi‐country study

Development and validation of risk prediction model for venous thromboembolism in postpartum women: multinational cohort study

Delirium following vascular surgery: increased incidence with preoperative β-blocker administration

Contact Info

Product

Resources

About