scholarly journals Nested Case-Control Studies With Multiple Measurements: Estimating Marginal Relationships

2021 ◽  
Author(s):  
Joshua N. Sampson ◽  
Paul S. Albert ◽  
Mark P. Purdue
Keyword(s):  
Conditional Logistic Regression ◽  
Case Control ◽  
Case Control Studies ◽  
Robust Variance ◽  
Non Hodgkin Lymphoma ◽  
Multiple Measurements ◽  
Matched Case ◽  
Nested Case Control ◽  
Standard Software ◽  
Single Time Point

Abstract Background: We consider the analysis of nested, matched, case-control studies that have multiple biomarker measurements per individual. We propose a simple approach for estimating the marginal relationship between a biomarker measured at a single time point and the risk of an event. We know of no other standard software package that can perform such analyses while explicitly accounting for the matching. Results: We propose an application of conditional logistic regression (CLR) that can include all measurements and uses a robust variance estimator. We compare our approach to other methods such as performing CLR with only the first measurement, CLR with an average of all measurements, and Generalized Estimating Equations. In simulations, our approach is significantly more powerful than CLR with one measurement or an average of all measurements, and has similar to power to GEE but correctly accounts for the matching. We then apply our approach to the CLUE cohort to show that an increased level of the immune marker sCD27 is associated with non‐Hodgkin lymphoma (NHL) and, by evaluating the strength of the association as a function of time until diagnosis, that the an increased level is likely an effect of the disease as opposed to a cause of the disease. The approach can be implemented by the R function clogitRV available at https://github.com/sampsonj74/clogitRV.Conclusion: We offered an approach and software for analyzing matched case-control studies with multiple measurements. We demonstrated that these methods are accurate, precise, and statistically powerful.

scholarly journals A systematic assessment of the association between frequently prescribed medicines and the risk of common cancers: a series of nested case-control studies

BMC Medicine ◽  
2021 ◽  
Vol 19 (1) ◽  
Author(s):  
R. D. McDowell ◽  
C. Hughes ◽  
P. Murchie ◽  
C. Cardwell
Keyword(s):  
Cancer Risk ◽  
Conditional Logistic Regression ◽  
Exposure Dose ◽  
Case Control ◽  
Case Control Studies ◽  
Systematic Assessment ◽  
Exposure Response ◽  
Prescribed Medicines ◽  
Nested Case Control ◽  
Novel Associations

Abstract Background Studies systematically screening medications have successfully identified prescription medicines associated with cancer risk. However, adjustment for confounding factors in these studies has been limited. We therefore investigated the association between frequently prescribed medicines and the risk of common cancers adjusting for a range of confounders. Methods A series of nested case-control studies were undertaken using the Primary Care Clinical Informatics Unit Research (PCCIUR) database containing general practice (GP) records from Scotland. Cancer cases at 22 cancer sites, diagnosed between 1999 and 2011, were identified from GP records and matched with up to five controls (based on age, gender, GP practice and date of registration). Odds ratios (OR) and 95% confidence intervals (CI) comparing any versus no prescriptions for each of the most commonly prescribed medicines, identified from prescription records, were calculated using conditional logistic regression, adjusting for comorbidities. Additional analyses adjusted for smoking use. An association was considered a signal based upon the magnitude of its adjusted OR, p-value and evidence of an exposure-response relationship. Supplementary analyses were undertaken comparing 6 or more prescriptions versus less than 6 for each medicine. Results Overall, 62,109 cases and 276,580 controls were included in the analyses and a total of 5622 medication-cancer associations were studied across the 22 cancer sites. After adjusting for comorbidities 2060 medicine-cancer associations for any prescription had adjusted ORs greater than 1.25 (or less than 0.8), 214 had a corresponding p-value less than or equal to 0.01 and 118 had evidence of an exposure-dose relationship hence meeting the criteria for a signal. Seventy-seven signals were identified after additionally adjusting for smoking. Based upon an exposure of 6 or more prescriptions, there were 118 signals after adjusting for comorbidities and 82 after additionally adjusting for smoking. Conclusions In this study a number of novel associations between medicine and cancer were identified which require further clinical and epidemiological investigation. The majority of medicines were not associated with an altered cancer risk and many identified signals reflected known associations between medicine and cancer.

scholarly journals Risk of osteoporosis and fragility fractures in asthma due to oral and inhaled corticosteroids: two population-based nested case-control studies

Thorax ◽  
2020 ◽  
Vol 76 (1) ◽  
pp. 21-28 ◽  
Author(s):  
Christos V Chalitsios ◽  
Dominick E Shaw ◽  
Tricia M McKeever
Keyword(s):  
Cumulative Dose ◽  
Inhaled Corticosteroids ◽  
Conditional Logistic Regression ◽  
Fragility Fractures ◽  
Case Control ◽  
Practice Research ◽  
Clinical Practice Research Datalink ◽  
Case Control Studies ◽  
Increased Risk ◽  
Nested Case Control

BackgroundInhaled (ICS) and oral (OCS) corticosteroids are used widely in asthma; however, the risk of osteoporosis and fragility fracture (FF) due to corticosteroids in asthma is not well-established.MethodsWe conducted two nested case-control studies using linked data from the Clinical Practice Research Datalink (CPRD) and Hospital Episode Statistics (HES) databases. Using an asthma cohort, we separately identified patients with osteoporosis or FF and gender-, age- and practice-matched controls. Conditional logistic regression was used to determine the association between ICS and OCS exposure, and the risk of osteoporosis or FF. The prevalence of patients receiving at least one bisphosphonate was also calculated.ResultsThere was a dose–response relationship between both cumulative dose and number of OCS/ICS prescriptions within the previous year, and risk of osteoporosis or FF. After adjusting for confounders, people receiving more OCS prescriptions (≥9 vs 0) had a 4.50 (95% CI 3.21 to 6.11) and 2.16 (95% CI 1.56 to 3.32) increased risk of osteoporosis and FF, respectively. For ICS (≥11 vs 0) the ORs were 1.60 (95% CI 1.22 to 2.10) and 1.31 (95% CI 1.02 to 1.68). The cumulative dose had a similar impact, with those receiving more OCS or ICS being at greater risk. The prevalence of patients taking ≥9 OCS and at least one bisphosphonate prescription was just 50.6% and 48.4% for osteoporosis and FF, respectively.ConclusionsThe findings suggest that exposure to OCS or ICS is an independent risk factors for bone health in patients with asthma. Steroid administration at the lowest possible level to maintain asthma control is recommended.

scholarly journals Types of Carbohydrates Intake during Pregnancy and Frequency of a Small for Gestational Age Newborn: A Case-Control Study

Nutrients ◽  
2019 ◽  
Vol 11 (3) ◽  
pp. 523 ◽  
Author(s):  
Carmen Amezcua-Prieto ◽  
Juan Martínez-Galiano ◽  
Naomi Cano-Ibáñez ◽  
Rocío Olmedo-Requena ◽  
Aurora Bueno-Cavanillas ◽  
...  
Keyword(s):  
Gestational Age ◽  
Small For Gestational Age ◽  
Conditional Logistic Regression ◽  
Case Control ◽  
Inverse Association ◽  
Odds Ratios ◽  
Case Control Studies ◽  
Logistic Regression Models ◽  
Matched Case ◽  
Brown Bread

The objective of this study was to assess the relationship between consumption of different types of carbohydrates (CHO) during pregnancy and the risk of having a small for gestational age (SGA) newborn. A retrospective matched case–control design was carried out with a total of 518 mother-offspring pairs. A total of 137 validated items were included in the food frequency questionnaire (FFQ). Conditional logistic regression models were used to calculate crude odds ratios (cORs) and adjusted odds ratios (aORs) with 95% confidence intervals (CIs). Having more than 75 g/day of brown bread showed an inverse association with SGA (aOR = 0.64, CI 0.43–0.96). In contrast, an intake of industrial sweets more than once a day (aOR = 2.70, CI 1.42–5.13), or even 2–6 times a week (aOR = 1.84, CI 1.20–2.82), increased the odds of having a SGA newborn. During pregnancy, the higher the increase of wholegrain cereal and bread, the lower the possibility of having a SGA newborn, but the opposite occurred with refined sugar products—just consuming industrial bakery products or pastries twice a week increased the odds of having an SGA infant. Case–control studies cannot verify causality and only show associations, which may reflect residual confusion due to the presence of unknown factors. It is possible that a high consumption of sugary foods is a marker of a generally poor lifestyle.

scholarly journals Matched versus Unmatched Analysis of Matched Case-Control Studies

2021 ◽  
Author(s):  
Fei Wan ◽  
Graham A Colditz ◽  
Siobhan Sutcliffe
Keyword(s):  
Logistic Regression ◽  
Functional Form ◽  
Conditional Logistic Regression ◽  
Control Sample ◽  
Target Population ◽  
Case Control ◽  
Model Specification ◽  
Exact Matching ◽  
Case Control Studies ◽  
Matched Case

Abstract Although the need for addressing matching in the analysis of matched case-control studies is well established, debate remains as to the most appropriate analytic method when matching on at least one continuous factor. We compare the bias and efficiency of unadjusted and adjusted conditional logistic regression (CLR) and unconditional logistic regression (ULR) in the setting of both exact and non-exact matching. To demonstrate that case-control matching distorts the association between the matching variables and the outcome in the matched sample relative to the target population, we derive the logit model for the matched case-control sample under exact matching. We conduct simulations to validate our theoretical conclusions and to explore different ways of adjusting for the matching variables in CLR and ULR to reduce biases. When matching is exact, CLR is unbiased in all settings. When matching is not exact, unadjusted CLR tends to be biased and this bias increases with increasing matching caliper size. Spline smoothing of the matching variables in CLR can alleviate biases. Regardless of exact or non-exact matching, adjusted ULR is generally biased unless the functional form of the matched factors is modelled correctly. The validity of adjusted ULR is vulnerable to model specification error. CLR should remain the primary analytic approach.

scholarly journals Causal mediation analysis in nested case‐control studies using conditional logistic regression

2020 ◽  
Vol 62 (8) ◽  
pp. 1939-1959
Author(s):  
Young Min Kim ◽  
John B. Cologne ◽  
Euna Jang ◽  
Theis Lange ◽  
Yoshimi Tatsukawa ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Mediation Analysis ◽  
Conditional Logistic Regression ◽  
Case Control ◽  
Case Control Studies ◽  
Causal Mediation ◽  
Causal Mediation Analysis ◽  
Nested Case Control

Abstract P077: Plasma Homocysteine, Dietary B Vitamins, and Risk of Peripheral Artery Disease

Circulation ◽  
2013 ◽  
Vol 127 (suppl_12) ◽  
Author(s):  
Monica L Bertoia ◽  
Jennifer K Pai ◽  
Michel M Joosten ◽  
Murray A Mittleman ◽  
Eric B Rimm ◽  
...  
Keyword(s):  
Health Professionals ◽  
Conditional Logistic Regression ◽  
Ankle Brachial Index ◽  
Case Control ◽  
Plasma Homocysteine ◽  
B Vitamins ◽  
Folate Intake ◽  
Case Control Studies ◽  
Increased Risk ◽  
Nested Case Control

Background: Clinical trials of vitamin B supplements to lower homocysteine levels have not consistently lowered risk of PAD, but these interventions were typically 1-2 years duration and limited to participants with pre-existing vascular disease. Whether long-term homocysteine levels influence the development of incident PAD remains uncertain. Objective: To examine the association between levels of plasma homocysteine and dietary B vitamins (folate, B6 and B12) and risk of PAD. Methods: We identified 276 cases of PAD in the Nurses’ Health Study (NHS) which followed 121,700 female nurses from 1980-2010, and 406 cases in the Health Professionals Follow-up Study (HPFS) which followed 51,529 male health professionals from 1986 to 2008. B vitamin intake was measured using a FFQ administered every four years. After excluding individuals with a history of CVD at baseline, plasma homocysteine was measured in baseline blood samples in two nested case-control studies with three controls matched to each case: NHS (1989-2010; 144 cases and 432 controls) and HPFS (1994-2008; 143 cases and 429 controls). PAD cases were confirmed by medical record based on at least one of: (1) amputation, bypass, or revascularization procedure, (2) angiogram/ultrasound confirming at least 50% stenosis, (3) ankle-brachial index < 0.9, or (4) documented physician’s diagnosis. We used Cox proportional hazards models to estimate hazard ratios (HR) for PAD in the cohorts and conditional logistic regression to estimate relative risks (RR) for PAD in the nested case-control studies. Results: Men in the highest tertile of plasma homocysteine had more than double the risk of PAD compared to men in the lowest tertile (Table). This association was not replicated in women. Lower folate intake was also associated with an increased risk of PAD in men but not women (Table). Total B6 and B12 intake was not consistently associated with risk of PAD in men or women. Conclusions: Higher plasma homocysteine levels were associated with an increased risk of PAD in men but not women.

scholarly journals Epigenetic drift association with cancer risk and survival, and modification by sex

2021 ◽  
Author(s):  
Chenglong Yu ◽  
Ee Ming Wong ◽  
Jihoon Eric Joo ◽  
Allison M. Hodge ◽  
Enes Makalic ◽  
...  
Keyword(s):  
Dna Methylation ◽  
Cancer Risk ◽  
Conditional Logistic Regression ◽  
B Cell Lymphoma ◽  
Case Control ◽  
Case Control Studies ◽  
Cox Models ◽  
Age Related ◽  
Matched Case ◽  
Age And Sex

AbstractTo investigate age- and sex-specific DNA methylation alterations related to cancer risk and survival, we used matched case-control studies of colorectal (N=835), gastric (N=170), kidney (N=143), lung (N=332), prostate (N=869) and urothelial (N=428) cancers, and mature B-cell lymphoma (N=438). Linear mixed-effects models were conducted to identify age-, sex- and age-by-sex-associated methylation markers using a discovery (controls) - replication (cases) strategy. Replication was further examined using summary statistics from Generation Scotland (GS). Associations between replicated markers and risk of and survival from cancer were assessed using conditional logistic regression and Cox models (hazard ratios [HR]), respectively. We found 32,659, 23,141 and 48 CpGs with replicated associations for age, sex and age-by-sex, respectively. The replication rates (GS summary data) for these CpGs were 94%, 86% and 91%, respectively. Significant signals for cancer risk and survival were identified at some individual age-related CpGs. There was a strong negative trend in the association between epigenetic drift and risk of colorectal cancer. Two CpGs overlapping TMEM49 and ARX genes were associated with survival of overall (HR=0.91, P=7.7×10−4) and colorectal (HR=1.52, P=1.8×10−4) cancer, respectively, with significant age-by-sex interaction. Our results may provide markers for cancer early detection and prognosis prediction.Simple SummaryAgeing is the strongest cancer risk factor, and men and women exhibit disparate risk profiles in terms of incidence and survival. DNA methylation is known to strongly vary by age and sex. Epigenetic drift refers to age-related DNA methylation changes and the tendency for increasing discordance between epigenomes over time, but it remains unknown to what extent the epigenetic drift might contribute to cancer risk and survival. The aims of this study were to identify age-associated, sex-associated and sexually dimorphic age-associated (age-by-sex-associated) DNA methylation markers and investigate whether age- and age-by-sex-associated markers are associated with cancer risk and survival. Our study, which used a total of 3,215 matched case-control pairs with DNA methylation in pre-diagnostic blood, is the first large study to examine the association between sex-specific epigenetic drift and cancer development and progression. The results may be useful for cancer early diagnosis and prediction of prognosis.

Sign in / Sign up

Export Citation Format

Share Document