Conditional Logistic Regression with Sandwich Estimators: Application to a Meta-Analysis

Biometrics ◽  
1998 ◽  
Vol 54 (1) ◽  
pp. 195 ◽  
Author(s):  
Michael P. Fay ◽  
Barry I. Graubard ◽  
Laurence S. Freedman ◽  
Douglas N. Midthune
Keyword(s):  
Logistic Regression ◽  
Meta Analysis ◽  
Conditional Logistic Regression ◽  
Sandwich Estimators

scholarly journals 455Association of interpregnancy interval and preterm births: what does a sibling-matched study indicate?

2021 ◽  
Vol 50 (Supplement_1) ◽  
Author(s):  
Gizachew Tessema ◽  
M Luke Marinovich ◽  
Siri E Håberg ◽  
Mika Gissler ◽  
Jonathan A Mayo ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Birth Outcomes ◽  
Meta Analysis ◽  
Conditional Logistic Regression ◽  
Preterm Births ◽  
Adverse Birth Outcomes ◽  
Interpregnancy Interval ◽  
Insufficient Evidence ◽  
Increased Risk ◽  
Matched Study

Abstract Background Most evidence for interpregnancy interval (IPI) and adverse birth outcomes come from between-women (unmatched) studies that are prone to incomplete control for confounders that vary between women. Comparing pregnancies to the same women (sibling-matched) can address this issue. Methods We conducted an international longitudinal study of births from Australia, Finland, Norway, and United States (California) covering over three decades (1980-2017). We included 5,523,914 births to 3,849,737 women, and within each country, we first investigated the associations between IPI and preterm birth (PTB), spontaneous PTB using logistic regression (unmatched analyses). Second, we used conditional logistic regression comparing IPIs in the same women (sibling-matched analyses), with 2,908,907 births to 1,234,730 women having at least two IPIs. Random effects meta-analysis was used to calculate pooled effect estimates. Results Compared to an IPI of 18-23 months, there was insufficient evidence of association between IPI <6 months and overall PTB (adjusted odds ratio (aOR)=1.08, 95%CI 0.99-1.18) but increased odds of spontaneous PTB (aOR=1.38, 95%CI 1.21-1.57). We observed elevated odds of both birth outcomes associated with IPI ≥60 months. In comparison, between-women analyses showed elevated odds of adverse birth outcomes for <12 month and >24 month IPIs. Conclusions We found consistently elevated odds of PTB following long IPIs. IPI shorter than 6 months were associated with elevated risk of spontaneous PTB, but there was insufficient evidence for increased risk of overall PTB. Key message Current recommendations of waiting at least 24 months to conceive after a previous pregnancy, may be unnecessarily long in high-income countries.

scholarly journals Interpregnancy intervals and adverse birth outcomes in high-income countries: An international cohort study

PLoS ONE ◽  
2021 ◽  
Vol 16 (7) ◽  
pp. e0255000
Author(s):  
Gizachew A. Tessema ◽  
M. Luke Marinovich ◽  
Siri E. Håberg ◽  
Mika Gissler ◽  
Jonathan A. Mayo ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Cohort Study ◽  
Birth Outcomes ◽  
Meta Analysis ◽  
Conditional Logistic Regression ◽  
The United States ◽  
High Income ◽  
Adverse Birth Outcomes ◽  
Insufficient Evidence ◽  
Increased Risk

Background Most evidence for interpregnancy interval (IPI) and adverse birth outcomes come from studies that are prone to incomplete control for confounders that vary between women. Comparing pregnancies to the same women can address this issue. Methods We conducted an international longitudinal cohort study of 5,521,211 births to 3,849,193 women from Australia (1980–2016), Finland (1987–2017), Norway (1980–2016) and the United States (California) (1991–2012). IPI was calculated based on the time difference between two dates—the date of birth of the first pregnancy and the date of conception of the next (index) pregnancy. We estimated associations between IPI and preterm birth (PTB), spontaneous PTB, and small-for-gestational age births (SGA) using logistic regression (between-women analyses). We also used conditional logistic regression comparing IPIs and birth outcomes in the same women (within-women analyses). Random effects meta-analysis was used to calculate pooled adjusted odds ratios (aOR). Results Compared to an IPI of 18–23 months, there was insufficient evidence for an association between IPI <6 months and overall PTB (aOR 1.08, 95% CI 0.99–1.18) and SGA (aOR 0.99, 95% CI 0.81–1.19), but increased odds of spontaneous PTB (aOR 1.38, 95% CI 1.21–1.57) in the within-women analysis. We observed elevated odds of all birth outcomes associated with IPI ≥60 months. In comparison, between-women analyses showed elevated odds of adverse birth outcomes for <12 month and >24 month IPIs. Conclusions We found consistently elevated odds of adverse birth outcomes following long IPIs. IPI shorter than 6 months were associated with elevated risk of spontaneous PTB, but there was insufficient evidence for increased risk of other adverse birth outcomes. Current recommendations of waiting at least 24 months to conceive after a previous pregnancy, may be unnecessarily long in high-income countries.

VP26 Comparing Statistical Methods For Meta-Analysis Of Rare Event Data

2017 ◽  
Vol 33 (S1) ◽  
pp. 158-159
Author(s):  
Tarang Sharma ◽  
Peter Gøtzsche ◽  
Oliver Kuss
Keyword(s):  
Logistic Regression ◽  
Adverse Events ◽  
Odds Ratio ◽  
Meta Analysis ◽  
Conditional Logistic Regression ◽  
Rare Event ◽  
Sensitivity Analyses ◽  
Analysis Methods ◽  
Binomial Regression ◽  
Events Data

INTRODUCTION:We aimed to identify the validity and robustness of effect estimates for serious rare adverse events in clinical study reports of antidepressant trials, across different meta-analysis methods for rare binary events data (1,2).METHODS:Four serious rare adverse events (all-cause mortality, suicidality, aggressive behaviour and akathisia) were meta-analyzed using different methods (3). The Yusuf-Peto odds ratio (OR), which ignores studies with no events in the treatment arms, was compared with the alternative approaches of generalized linear mixed models (GLMM), conditional logistic regression, a Bayesian approach using Markov Chain Monte Carlo (MCMC) and a beta-binomial regression model.RESULTS:Though the estimates for the four outcomes did not change substantially across the different analysis methods, the Yusuf-Peto method underestimated the treatment harm and overstimated its precision, especially when the estimated odds ratio (OR) deviated greatly from 1. For example the OR for suicidality for children and adolescents was 2.39 (95 percent Confidence Interval, CI 1.32 to 4.33, using the Yusuf-Peto method), but increased to 2.64 (95 percent CI 1.33 to 5.26) using conditional logistic regression, to 2.69 (95 percent CI 1.19 to 6.09) using beta-binomial, to 2.73 (95 percent CI 1.37 to 5.42) using the GLMM and finally to 2.87 (95 percent CI 1.42 to 5.98) using the MCMC approach.CONCLUSIONS:The method used for meta-analysis of rare events data influences the estimates obtained and the exclusion of double zero-event studies can give misleading results. To ensure reduction of bias and erroneous inferences, sensitivity analyses should be performed using different methods and we recommend that the Yusuf-Peto approach should no longer be used. Other methods, in particular the beta-binomial method that was shown to be superior, should be considered instead.

Characteristics of Adenocarcinoma of the Gallbladder Following Cholecystectomies in the Hispanic Population

2019 ◽  
Vol 152 (Supplement_1) ◽  
pp. S64-S65
Author(s):  
David Gustafson ◽  
Osvaldo Padilla
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Incidental Finding ◽  
Conditional Logistic Regression ◽  
Gallbladder Disease ◽  
Predictor Variables ◽  
Wall Thickening ◽  
Conditional Logistic Regression Analysis

Abstract Introduction Gallbladder adenocarcinoma (GBC) is a rare malignancy. Frequency of incidental adenocarcinoma of the gallbladder in the literature is approximately 0.2% to 3%. Typically, GBC is the most common type and is discovered late, not until significant symptoms develop. Common symptoms include right upper quadrant pain, nausea, anorexia, and jaundice. A number of risk factors in the literature are noted for GBC. These risk factors are also more prevalent in Hispanic populations. This study sought to compare patients with incidental gallbladder adenocarcinomas (IGBC) to those with high preoperative suspicion for GBC. Predictor variables included age, sex, ethnicity, radiologic wall thickening, gross pathology characteristics (wall thickness, stone size, stone number, and tumor size), histologic grade, and staging. Methods Cases of GBC were retrospectively analyzed from 2009 through 2017, yielding 21 cases. Data were collected via Cerner EMR of predictor variables noted above. Statistical analysis utilized conditional logistic regression analysis. Results The majority of patients were female (n = 20) and Hispanic (n = 19). There were 14 IGBCs and 7 nonincidental GBCs. In contrast with previous research, exact conditional logistic regression analysis revealed no statistically significant findings. For every one-unit increase in AJCC TNM staging, there was a nonsignificant 73% reduction in odds (OR = 0.27) of an incidental finding of gallbladder carcinoma. Conclusion This study is important in that it attempts to expand existing literature regarding a rare type of cancer in a unique population, one particularly affected by gallbladder disease. Further studies are needed to increase predictive knowledge of this cancer. Longer studies are needed to examine how predictive power affects patient outcomes. This study reinforces the need for routine pathologic examination of cholecystectomy specimens for cholelithiasis.

Computational tools for exact conditional logistic regression

2001 ◽  
Vol 20 (17-18) ◽  
pp. 2723-2739 ◽  
Author(s):  
Chris Corcoran ◽  
Cyrus Mehta ◽  
Nitin Patel ◽  
Pralay Senchaudhuri
Keyword(s):  
Logistic Regression ◽  
Conditional Logistic Regression ◽  
Computational Tools

Predicting 30-day Hospital Readmission with Publicly Available Administrative Database

2015 ◽  
Vol 54 (06) ◽  
pp. 560-567 ◽  
Author(s):  
K. Zhu ◽  
Z. Lou ◽  
J. Zhou ◽  
N. Ballester ◽  
P. Parikh ◽  
...  
Keyword(s):  
Heart Failure ◽  
Logistic Regression ◽  
Decision Tree ◽  
Ad Hoc ◽  
Prediction Models ◽  
Conditional Logistic Regression ◽  
Hospital Readmissions ◽  
Decision Rules ◽  
Classification Models ◽  
Standard Classification

SummaryIntroduction: This article is part of the Focus Theme of Methods of Information in Medicine on “Big Data and Analytics in Healthcare”.Background: Hospital readmissions raise healthcare costs and cause significant distress to providers and patients. It is, therefore, of great interest to healthcare organizations to predict what patients are at risk to be readmitted to their hospitals. However, current logistic regression based risk prediction models have limited prediction power when applied to hospital administrative data. Meanwhile, although decision trees and random forests have been applied, they tend to be too complex to understand among the hospital practitioners.Objectives: Explore the use of conditional logistic regression to increase the prediction accuracy.Methods: We analyzed an HCUP statewide in-patient discharge record dataset, which includes patient demographics, clinical and care utilization data from California. We extracted records of heart failure Medicare beneficiaries who had inpatient experience during an 11-month period. We corrected the data imbalance issue with under-sampling. In our study, we first applied standard logistic regression and decision tree to obtain influential variables and derive practically meaning decision rules. We then stratified the original data set accordingly and applied logistic regression on each data stratum. We further explored the effect of interacting variables in the logistic regression modeling. We conducted cross validation to assess the overall prediction performance of conditional logistic regression (CLR) and compared it with standard classification models.Results: The developed CLR models outperformed several standard classification models (e.g., straightforward logistic regression, stepwise logistic regression, random forest, support vector machine). For example, the best CLR model improved the classification accuracy by nearly 20% over the straightforward logistic regression model. Furthermore, the developed CLR models tend to achieve better sensitivity of more than 10% over the standard classification models, which can be translated to correct labeling of additional 400 – 500 readmissions for heart failure patients in the state of California over a year. Lastly, several key predictor identified from the HCUP data include the disposition location from discharge, the number of chronic conditions, and the number of acute procedures.Conclusions: It would be beneficial to apply simple decision rules obtained from the decision tree in an ad-hoc manner to guide the cohort stratification. It could be potentially beneficial to explore the effect of pairwise interactions between influential predictors when building the logistic regression models for different data strata. Judicious use of the ad-hoc CLR models developed offers insights into future development of prediction models for hospital readmissions, which can lead to better intuition in identifying high-risk patients and developing effective post-discharge care strategies. Lastly, this paper is expected to raise the awareness of collecting data on additional markers and developing necessary database infrastructure for larger-scale exploratory studies on readmission risk prediction.

scholarly journals Smoking methylation marks for prediction of urothelial cancer risk

2021 ◽  
Author(s):  
Chenglong Yu ◽  
Kristina M. Jordahl ◽  
Julie K. Bassett ◽  
Jihoon Eric Joo ◽  
Ee Ming Wong ◽  
...  
Keyword(s):  
Dna Methylation ◽  
Risk Prediction ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Conditional Logistic Regression ◽  
Area Under The Curve ◽  
Testing Dataset ◽  
Smoking Exposure ◽  
Melbourne Collaborative Cohort Study ◽  
Confounder Adjustment

AbstractBackgroundSelf-reported information may not accurately capture smoking exposure. We aimed to evaluate whether smoking-associated DNA methylation markers improve urothelial cell carcinoma (UCC) risk prediction.MethodsConditional logistic regression was used to assess associations between blood-based methylation and UCC risk using two matched case-control samples, N=404 pairs from the Melbourne Collaborative Cohort Study (MCCS) and N=440 pairs from the Women’s Health Initiative (WHI) cohort, respectively. Results were pooled using fixed-effects meta-analysis. We developed methylation-based predictors of UCC and evaluated their prediction accuracy on two replication datasets using the area under the curve (AUC).ResultsThe meta-analysis identified associations (P<4.7×10−5) for 29 of 1,061 smoking-associated methylation sites, but these were substantially attenuated after adjustment for self-reported smoking. Nominally significant associations (P<0.05) were found for 387 (36%) and 86 (8%) of smoking-associated markers without/with adjustment for self-reported smoking, respectively, with same direction of association as with smoking for 387 (100%) and 79 (92%) markers. A Lasso-based predictor was associated with UCC risk in one replication dataset in MCCS (N=134, odds ratio per SD [OR]=1.37, 95%CI=1.00-1.90) after confounder adjustment; AUC=0.66, compared with AUC=0.64 without methylation information. Limited evidence of replication was found in the second testing dataset in WHI (N=440, OR=1.09, 95%CI=0.91-1.30).ConclusionsCombination of smoking-associated methylation marks may provide some improvement to UCC risk prediction. Our findings need further evaluation using larger datasets.ImpactDNA methylation may be associated with UCC risk beyond traditional smoking assessment and could contribute to some improvements in stratification of UCC risk in the general population.

scholarly journals A novel weighting method to remove bias from within-subject exposure dependency in case-crossover studies

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Kiyoshi Kubota ◽  
Thu-Lan Kelly ◽  
Tsugumichi Sato ◽  
Nicole Pratt ◽  
Elizabeth Roughead ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Crossover Study ◽  
Odds Ratio ◽  
Time Trends ◽  
Rate Ratio ◽  
Conditional Logistic Regression ◽  
Time Varying ◽  
Weighting Method ◽  
Case Crossover ◽  
Crossover Studies

Abstract Background Case-crossover studies have been widely used in various fields including pharmacoepidemiology. Vines and Farrington indicated in 2001 that when within-subject exposure dependency exists, conditional logistic regression can be biased. However, this bias has not been well studied. Methods We have extended findings by Vines and Farrington to develop a weighting method for the case-crossover study which removes bias from within-subject exposure dependency. Our method calculates the exposure probability at the case period in the case-crossover study which is used to weight the likelihood formulae presented by Greenland in 1999. We simulated data for the population with a disease where most patients receive a cyclic treatment pattern with within-subject exposure dependency but no time trends while some patients stop and start treatment. Finally, the method was applied to real-world data from Japan to study the association between celecoxib and peripheral edema and to study the association between selective serotonin reuptake inhibitor (SSRI) and hip fracture in Australia. Results When the simulated rate ratio of the outcome was 4.0 in a case-crossover study with no time-varying confounder, the proposed weighting method and the Mantel-Haenszel odds ratio reproduced the true rate ratio. When a time-varying confounder existed, the Mantel-Haenszel method was biased but the weighting method was not. When more than one control period was used, standard conditional logistic regression was biased either with or without time-varying confounding and the bias increased (up to 8.7) when the study period was extended. In real-world analysis with a binary exposure variable in Japan and Australia, the point estimate of the odds ratio (around 2.5 for the association between celecoxib and peripheral edema and around 1.6 between SSRI and hip fracture) by our weighting method was equal to the Mantel-Haenszel odds ratio and stable compared with standard conditional logistic regression. Conclusion Case-crossover studies may be biased from within-subject exposure dependency, even without exposure time trends. This bias can be identified by comparing the odds ratio by the Mantel-Haenszel method and that by standard conditional logistic regression. We recommend using our proposed method which removes bias from within-subject exposure dependency and can account for time-varying confounders.

Sign in / Sign up

Export Citation Format

Share Document