Conditional versus unconditional logistic regression in the medical literature

Rahman, Mahbubur; Sakamoto, Junichi; Fukui, Tsuguya

doi:10.1016/s0895-4356(02)00507-3

Cited by 17 publications

(8 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To statistically compare cases and controls, we used McNemar's chi-squared (χ 2 ) tests when the frequency in all of the cells of the cross-tabulation was ≥5 and Fisher's exact test otherwise. Univariate conditional logistic regression models [38] were fitted to determine the strength of the association between SLT use and CHD, with the effect of SLT use expressed as a matched odds ratios (ORs) with 95% confidence intervals (CIs). Then multivariate conditional logistic regression models were fitted to adjust for potential confounding variables.…”

Section: Methodsmentioning

confidence: 99%

Is There Any Association between Use of Smokeless Tobacco Products and Coronary Heart Disease in Bangladesh?

et al. 2012

View full text Add to dashboard Cite

BackgroundMost epidemiological studies exploring the association between smokeless tobacco (SLT) use and coronary heart disease (CHD) have been in Western populations, and have focused on SLT products used in those countries. Few studies come from South Asian countries. Our objective was to determine the association between SLT use and CHD among non-smoking adults in Bangladesh.MethodsA matched case-control study of non-smoking Bangladeshi adults aged 40–75 years was conducted in 2010. Incident cases of CHD were selected from two cardiac hospitals. Community controls, matched to CHD cases, were selected from neighbourhoods, and hospital controls were selected from outpatient departments of the same hospitals. The Rose Angina Questionnaire (RAQ) was also used to re-classify cases and controls.ResultsThe study enrolled 302 cases, 1,208 community controls and 302 hospital controls. Current use was higher among community controls (38%) compared to cases (33%) and hospital controls (32%). Current use of SLT was not significantly associated with an increased risk of CHD when community controls were used (adjusted OR 0.87, 95% CI 0.63–1.19), or when hospital controls were used (adjusted OR 1.00, 95% CI 0.63–1.60), or when both control groups were combined (adjusted OR 1.00, 95% CI 0.74–1.34). Risk of CHD did not increase with use of individual types except gul, frequency, duration, past use of SLT products, or using the RAQ to re-classify cases and controls. There was a significant association between gul use and CHD when both controls were combined (adjusted OR 2.93, 95% CI 1.28–6.70).ConclusionsThere was no statistically significant association between SLT use in general and CHD among non-smoking adults in Bangladesh. Further research on the association between gul use and CHD in Bangladesh along with SLT use and CHD in other parts of the subcontinent will guide public health policy and interventions that focus on SLT-related diseases.

show abstract

Section: Methodsmentioning

confidence: 99%

Is There Any Association between Use of Smokeless Tobacco Products and Coronary Heart Disease in Bangladesh?

et al. 2012

View full text Add to dashboard Cite

show abstract

“…Within our nested case-control study, we determined whether each variable was associated with AKI and performed a conditional logistic regression [34] to determine independent risk factors of AKI using variables with a p value <0.20 by univariate analysis. We used a conditional forward method of variable entry.…”

Section: Methodsmentioning

confidence: 99%

Acute Kidney Injury and Renal Recovery with the Use of Aminoglycosides: A Large Retrospective Study

Paquette

Bernier-Jean

Brunette³

et al. 2015

Nephron

View full text Add to dashboard Cite

Background: Recent acute kidney injury (AKI) guidelines, based on studies performed a decade ago, recommend avoiding aminoglycosides (AGs) in patients at risk of AKI. Whether present patient characteristics and management have changed this risk is uncertain. We determined the current incidence, risk factors and outcomes of AG-AKI. Methods: We retrospectively identified adult patients who received gentamicin or tobramycin for ≥5 days in 2 large university-affiliated centers, excluding critically ill and dialysis patients. We assessed the incidence of Risk, Injury, Failure, Loss and End-stage kidney disease criteria of AKI risk and then matched each AKI to 2 controls of same age and gender to determine factors associated with AG-AKI and its recovery, defined by a creatinine within 150% of baseline by 21 days. Results: Since 2001, the frequency of AG administration and dosing declined, but the incidence of AG-AKI remained constant. Of the 562 patients who received AG for ≥5 days, 65 (12%) developed AG-AKI after 11 (IQR 8-15) days, with 56, 29 and 15% having stages 1, 2 and 3 AKI, respectively. We matched these to 130 controls. In this nested case-control study, independent AKI risk factors were vancomycin coadministration, high AG trough levels and heart failure. AG-AKI compared to AG exposure without AKI was associated with greater mortality. Renal recovery occurred in 51% of the AKI patients and was less likely with heart failure and higher AKI severity. Conclusion: AG administration has recently decreased but the risk of AKI remained unchanged and half of the patients did not recover. Vancomycin coadministration, high AG trough levels and heart failure independently predicted AKI.

show abstract

“…We assume each case is matched to a variable number of controls based on certain confounding variables. The matched case-control data can be analysed using conditional logistic regression, especially, when matching variables are not entered in the analysis [12]. Comparisons of unconditional and conditional likelihood functions for matched case-control studies have been discussed extensively in the literature, e.g.…”

Section: Score Tests For Matched Datamentioning

confidence: 99%

Robust trend tests for genetic association using matched case‐control design

Zheng

Tian

Weinberger

et al. 2005

Statistics in Medicine

View full text Add to dashboard Cite

Trend tests for genetic association using a matched case-control design are studied, which allows for a variable number of controls per case. However, the tests depend on the scores based on the underlying genetic model, thus it may result in substantial loss of power when the model is misspecified. Since the mode of inheritance may be unknown for complex diseases, robust trend tests in matched case-control studies are developed. Simulation is conducted to compare the trend tests and the robust trend tests under various genetic models. The results are applied to detect candidate-gene association using an example from a case-control aetiologic study of sarcoidosis.

show abstract

Conditional versus unconditional logistic regression in the medical literature

Cited by 17 publications

References 1 publication

Is There Any Association between Use of Smokeless Tobacco Products and Coronary Heart Disease in Bangladesh?

Is There Any Association between Use of Smokeless Tobacco Products and Coronary Heart Disease in Bangladesh?

Acute Kidney Injury and Renal Recovery with the Use of Aminoglycosides: A Large Retrospective Study

Robust trend tests for genetic association using matched case‐control design

Contact Info

Product

Resources

About