Breaking the matching in nested case–control data offered several advantages for risk estimation

Delcoigne, Bénédicte; Colzani, Edoardo; Prochazka, Michaela; Gagliardi, Giovanna; Hall, Per; Abrahamowicz, Michał; Czene, Kamila; Reilly, Marie

doi:10.1016/j.jclinepi.2016.11.014

Cited by 8 publications

(4 citation statements)

References 114 publications

(208 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All analyses to estimate the association of the biomarkers with breast cancer risk were conducted on the population of the placebo group rst with conditional logistic models and then with Cox proportional hazards models, in which the dependent variable is the couple ("breast cancer event"; "time to event"). To break the matching between the cases and their controls and e ciently include the time dimension in the investigation of the association between exposures (baseline BMI and adiponectin increase) and outcome (breast cancer event), a weighted Cox regression analysis was employed [19]. To obtain inferences for the full placebo cohort, the women included in the analysis were up-weighted using the Borgan II weights approach [20].…”

Section: Discussionmentioning

confidence: 99%

A mediation analysis of obesity and adiponectin association with postmenopausal breast cancer risk: a nested cohort study in the IBIS-II Prevention Trial

Macis

Bellerba

Aristarco

et al. 2023

Preprint

View full text Add to dashboard Cite

Purpose Obesity is a risk factor for postmenopausal breast cancer and evidence supports the pivotal role of adiponectin in the association between obesity and breast cancer. We aimed to investigate whether the effect of baseline body mass index (BMI) on postmenopausal breast cancer risk was mediated by adiponectin or other circulating biomarkers. Methods We measured adiponectin, leptin, IGF-I, IGFBP-1, high-sensitivity C-reactive protein, glycemia, insulin, HOMA-IR index, and SHBG in baseline and 12-month serum samples from 123 cases and 302 matched controls in the placebo arm of the IBIS-II Prevention trial. We conducted the main mediation analysis considering BMI at baseline as an exposure and the increase in adiponectin at 12 months as a mediator after adjustment for the Tyrer-Cuzick score and the use of lipid-lowering medications and supplements. Results The 12-month increase in adiponectin levels was significantly associated with a 40% decrease in the risk of breast cancer [hazard ratio (HR), 0.60; 95% confidence interval (95% CI), 0.36-1.00]. The estimate of the association of BMI was low, indicating only a 5% increase in breast cancer risk (HR, 1.05; 95% CI, 1.00-1.09). The increase in adiponectin levels at 12 months did not mediate the effect of BMI on breast cancer risk (HR, 1.00; 95% CI, 0.98–1.02). Conclusion Mediation analysis indicated an independent role of adiponectin in postmenopausal breast cancer risk. Raising adiponectin levels might be an attractive target for postmenopausal breast cancer prevention.

show abstract

Section: Discussionmentioning

confidence: 99%

A mediation analysis of obesity and adiponectin association with postmenopausal breast cancer risk: a nested cohort study in the IBIS-II Prevention Trial

Macis

Bellerba

Aristarco

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Specifically, this study design offers the benefit of eliminating the problem of referral or partial verification bias, commonly seen with typical case-control designs. Moreover, this study design also ensured that all the patients who were selected for the study signify a true representation of the actual disease status and act as a standard reference of interest for the study [15,16]. Furthermore, a nested case-control design allows a simple way for obtaining all the measures of diagnostic accuracy for the target sample [16].…”

Section: Study Design and Study Populationmentioning

confidence: 99%

Stroke risk associated with NSAIDs uses in women with dysmenorrhea: A population-based cohort study

2021

View full text Add to dashboard Cite

Objective Dysmenorrhea is among the most common type of gynecological problem, affecting young women across the globe. This study assessed the comparative risk of stroke associated with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) and non-NSAIDs in women with dysmenorrhea while taking into account the following factors such as age, history of pregnancy, NSAIDs uses and its duration of use, and selected comorbidities. Methodology We used a quantitative research approach based on a comparative case-control study design. The study data was selected from the Longitudinal Health Insurance Database (LHID) 2000, of the Taiwan National Health Research Institutes. Among the estimated 23.4 million insured Taiwanese, who were covered by the Taiwan health insurance system, in the 2000 registry of beneficiaries, one million individuals were randomly selected from the database. A total of 24,955 females suffering from dysmenorrhea were selected for the study. Out of those 3238 (13%) participated in the study group and 21,717 (87%) were randomly distributed into the controls group. Women in the age range, 15–49 years, who did not have any history of stroke, hysterectomy, and/or ovariectomy, were included in the study. A comparative proportional distribution analysis was used for data analysis. Results Age and use of NSAIDs and its duration of usage were factors associated with an increased incidence of stroke. The stroke incidence rate was 12.77 per 10,000 person-years, and 1.83-fold higher in NSAIDs use cohort than in comparisons with adjusted hazard ratio (aHR) of 1.47 (95% CI = 0.93–2.32). Among women with dysmenorrhea use of NSAIDs, the stroke incidence increased to 2.29-fold (aHR 95% CI = 1.36–3.84) in those use for ≧24 days per month and to 0.51-fold (aHR 95% CI = 0.13–2.10) in those use for 6–12 days per month. Conclusions Women with dysmenorrhea who use NSAIDs have a higher risk of stroke. Especially young women, the risk of stroke is further increased, and the longer the medication, the higher the risk of stroke. Every woman with symptoms of dysmenorrhea deserves specialized outpatient treatment and care.

show abstract

“…24,25 Extensive simulations confirmed that IPW estimators, which break the matching for two-phase designs, are efficient. 26,27…”

Section: Likelihood and Estimationmentioning

confidence: 99%

Semiparametric isotonic regression analysis for risk assessment under nested case-control and case-cohort designs

Feng

et al. 2019

Stat Methods Med Res

View full text Add to dashboard Cite

Two-phase sampling designs, including nested case-control and case-cohort designs, are frequently utilized in large cohort studies involving expensive biomarkers. To analyze data from two-phase designs with a binary outcome, parametric models such as logistic regression are often adopted. However, when the model assumptions are not valid, parametric models may lead to biased estimation and risk evaluation. In this paper, we propose a robust semiparametric regression model for binary outcomes and an easy-to-implement computational procedure that combines the pool-adjacent violators algorithm with inverse probability weighting. The asymptotic properties are established, including consistency and the convergence rate. Simulation studies show that the proposed method performs well and is more robust than logistic regression methods. We demonstrate the application of the proposed method to real data from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial.

show abstract

Breaking the matching in nested case–control data offered several advantages for risk estimation

Cited by 8 publications

References 114 publications

A mediation analysis of obesity and adiponectin association with postmenopausal breast cancer risk: a nested cohort study in the IBIS-II Prevention Trial

A mediation analysis of obesity and adiponectin association with postmenopausal breast cancer risk: a nested cohort study in the IBIS-II Prevention Trial

Stroke risk associated with NSAIDs uses in women with dysmenorrhea: A population-based cohort study

Semiparametric isotonic regression analysis for risk assessment under nested case-control and case-cohort designs

Contact Info

Product

Resources

About