Search

To assess the association between intakes of total alcohol and individual alcoholic beverages and the incidence of exfoliation glaucoma/glaucoma suspect (XFG/XFGS) status. Prospective cohort study. A total of 195 408 participants in the Nurses' Health Study (1980–2018), the Health Professionals Follow-up Study (1986–2018), and the Nurses' Health Study II (1991–2019) were followed biennially. Eligible participants at each 2-year risk period were ≧ 40 years and free of XFG/XFGS status with available data on diet and ophthalmic examination findings. Cumulatively averaged total (primary exposure) and individual alcoholic beverage (beer, wine, and liquor) intakes from validated dietary information every 2-4 years. Confirmed incident XFG/XFGS status using medical records. We used per-eye Cox proportional hazards models, accounting for intereye correlations, to estimate multivariate-adjusted relative risks (MVRRs) and 95% confidence intervals (CIs). During 6 877 823 eye-years of follow-up, 705 eyes with XFG/XFGS status were documented. Greater total alcohol consumption was associated significantly with higher XFG/XFGS status risk: the MVRR for XFG/XFGS status for cumulatively averaged alcohol consumption of ≧15 g/day or more versus nondrinking was 1.55 (95% CI, 1.17–2.07; P = 0.02 for trend). Long- and short-term alcohol intake was associated significantly with XFG/XFGS status risk, with the strongest associations with cumulatively averaged alcohol intake as of 4 years before diagnosis (MVRR ≥ 15 g/day vs. nondrinking, 1.65; 95% CI, 1.25–2.18; P = 0.002 for trend). Compared with nondrinkers, consuming ≧ 3.6 drinks of beer, wine, or liquor per week was associated with the following MVRRs for XFG/XFGS status: 1.26 (95% CI, 0.89–1.77; P = 0.40 for trend), 1.30 (95% CI, 1.00–1.68; P = 0.15 for trend), and 1.46 (95% CI, 1.15–1.85; P = 0.01 for trend), respectively. We did not observe interactions by age, latitude, residential tier, or intakes of folate or vitamin A (P > 0.40 for interaction); however, the association between alcohol and XFG/XFGS status was suggestively stronger for those without a family history of glaucoma (P = 0.10 for interaction). Long-term alcohol consumption was associated with a higher risk of XFG/XFGS status. Our findings provide further clues regarding the XFG/XFGS etiology. [ABSTRACT FROM AUTHOR]

Purpose: Metabolomic and preclinical studies suggest that branched‐chain amino acids (BCAA) may be inversely associated with neurodegenerative diseases including glaucoma. We therefore assessed the long‐term association between dietary intake of BCAA and incident primary open‐angle glaucoma (POAG) and POAG subtypes. Methods: We followed biennially participants of the Nurses' Health Study (NHS; 65 531 women: 1984–2016), Health Professionals Follow‐up Study (42 254 men: 1986–2016) and NHSII (66 904 women; 1991–2017). Eligible participants were 40+ years old and reported eye examinations. Repeated validated food frequency questionnaires were used to assess dietary intake of BCAA. Incident cases of POAG and POAG subtypes defined by visual field (VF) loss and untreated intraocular pressure (IOP) were confirmed by medical record review. Multivariable‐adjusted relative risks (MVRRs) and 95% confidence intervals (CIs) were estimated using Cox proportional hazards models. Results: We identified 1946 incident POAG cases. The pooled MVRRs of POAG for the highest quintile (Q5 = 17.1 g/day) versus lowest quintile (Q1 = 11.2 g/day) of total BCAA intake was 0.93 (95% CI, 0.73–1.19; ptrend = 0.45; pheterogeneity by sex = 0.24). For subtypes of POAG defined by IOP level or POAG with only peripheral VF loss, no associations were observed for men or women (ptrend ≥ 0.20); however, for the POAG subtype with early paracentral VF loss, there was a suggestion of an inverse association in women (MVRRQ5versusQ1 = 0.80 [95% CI, 0.57–1.12; ptrend = 0.12]) but not in men (MVRRQ5versusQ1 = 1.38 [95% CI, 0.81–2.34; ptrend = 0.28; pheterogeneity by sex = 0.06]). Conclusion: Higher dietary intake of BCAA was not associated with POAG risk. [ABSTRACT FROM AUTHOR]

Cardiovascular (CV) disease fatality rates are higher for women compared with men with diabetes despite lower rates of obstructive coronary artery disease (CAD). Impaired coronary flow reserve (CFR), the ratio of adenosine-stimulated to rest myocardial blood flow (MBF), is an indicator of coronary microvascular dysfunction and predicts major adverse CV events. We performed a post hoc analysis to determine whether there was a sex disparity in coronary microvascular dysfunction among 46 men and 27 women with well-controlled type 2 diabetes and without clinical evidence of obstructive CAD. We found that women had a higher rest MBF, lower CFR, and worse diastolic function compared with men. In addition, rest MBF was positively correlated with worse diastolic function in women. We previously showed that mineralocorticoid blockade improved CFR in men and women with type 2 diabetes, implicating aldosterone in the pathophysiology of coronary microvascular dysfunction. We therefore examined aldosterone levels and found that women had larger increases in aldosterone in response to an angiotensin-II infusion than did men. In conclusion, among individuals with type 2 diabetes and good cardiometabolic control, women had worse myocardial perfusion and diastolic function compared with men. The greater aldosterone responsivity in women may be a mechanism for this sex effect. [ABSTRACT FROM AUTHOR]

Background: The objective of this study was to evaluate the role of low‐carbohydrate diets after breast cancer diagnosis in relation to breast cancer–specific and all‐cause mortality. Methods: For 9621 women with stage I–III breast cancer from two ongoing cohort studies, the Nurses' Health Study and Nurses' Health Study II, overall low‐carbohydrate, animal‐rich low‐carbohydrate, and plant‐rich low‐carbohydrate diet scores were calculated by using food frequency questionnaires collected after breast cancer diagnosis. Results: Participants were followed up for a median 12.4 years after breast cancer diagnosis. We documented 1269 deaths due to breast cancer and 3850 all‐cause deaths. With the use of Cox proportional hazards regression and after controlling for potential confounding variables, we observed a significantly lower risk of overall mortality among women with breast cancer who had greater adherence to overall low‐carbohydrate diets (hazard ratio for quintile 5 vs. quintile 1 [HRQ5vsQ1], 0.82; 95% CI, 0.74–0.91; ptrend =.0001) and plant‐rich low‐carbohydrate diets (HRQ5vsQ1, 0.73; 95% CI, 0.66–0.82; ptrend <.0001) after breast cancer diagnosis but not animal‐rich low‐carbohydrate diets (HRQ5vsQ1, 0.93; 95% CI, 0.84–1.04; ptrend =.23). However, greater adherence to overall, animal‐rich, or plant‐rich low‐carbohydrate diets was not significantly associated with a lower risk of breast cancer–specific mortality. Conclusions: This study showed that greater adherence to low‐carbohydrate diets, especially plant‐rich low‐carbohydrate diets, was associated with better overall survival but not breast cancer–specific survival among women with stage I–III breast cancer. Greater adherence to low‐carbohydrate diets, especially plant‐rich low‐carbohydrate diets, was associated with better overall survival but not breast cancer–specific survival among women with stage I–III breast cancer. [ABSTRACT FROM AUTHOR]

Background: Epidemiologic and medical studies often rely on evaluators to obtain measurements of exposures or outcomes for study participants, and valid estimates of associations depends on the quality of data. Even though statistical methods have been proposed to adjust for measurement errors, they often rely on unverifiable assumptions and could lead to biased estimates if those assumptions are violated. Therefore, methods for detecting potential 'outlier' evaluators are needed to improve data quality during data collection stage. Methods: In this paper, we propose a two-stage algorithm to detect 'outlier' evaluators whose evaluation results tend to be higher or lower than their counterparts. In the first stage, evaluators' effects are obtained by fitting a regression model. In the second stage, hypothesis tests are performed to detect 'outlier' evaluators, where we consider both the power of each hypothesis test and the false discovery rate (FDR) among all tests. We conduct an extensive simulation study to evaluate the proposed method, and illustrate the method by detecting potential 'outlier' audiologists in the data collection stage for the Audiology Assessment Arm of the Conservation of Hearing Study, an epidemiologic study for examining risk factors of hearing loss in the Nurses' Health Study II. Results: Our simulation study shows that our method not only can detect true 'outlier' evaluators, but also is less likely to falsely reject true 'normal' evaluators. Conclusions: Our two-stage 'outlier' detection algorithm is a flexible approach that can effectively detect 'outlier' evaluators, and thus data quality can be improved during data collection stage. [ABSTRACT FROM AUTHOR]

It is well known that the sample correlation coefficient (Rxy ) is the maximum likelihood estimator of the Pearson correlation (ρxy ) for independent and identically distributed (i.i.d.) bivariate normal data. However, this is not true for ophthalmologic data where X (e.g., visual acuity) and Y (e.g., visual field) are available for each eye and there is positive intraclass correlation for both X and Y in fellow eyes. In this paper, we provide a regression-based approach for obtaining the maximum likelihood estimator of ρxy for clustered data, which can be implemented using standard mixed effects model software. This method is also extended to allow for estimation of partial correlation by controlling both X and Y for a vector U_ of other covariates. In addition, these methods can be extended to allow for estimation of rank correlation for clustered data by (i) converting ranks of both X and Y to the probit scale, (ii) estimating the Pearson correlation between probit scores for X and Y, and (iii) using the relationship between Pearson and rank correlation for bivariate normally distributed data. The validity of the methods in finite-sized samples is supported by simulation studies. Finally, two examples from ophthalmology and analgesic abuse are used to illustrate the methods. Copyright © 2017 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Background: The role of body fatness in the aetiology of breast cancer is complex. We evaluated the independent and synergistic effects of body fatness, at different stages throughout a woman's life course, on premenopausal breast cancer risk.Methods: Premenopausal participants of the Nurses' Health Study II (NHSII) were followed from 1991 up to 2009. Body fatness factors including birthweight, somatotype (a 9-level pictogram with level 1 being the leanest) at ages 5 and 10 years and body mass index (BMI) at age 18 were collected at baseline. Current BMI was updated biennially. Multivariate Cox regression models were used to evaluate the association between each body fatness factor as well as cross-classification of all factors and the incidence of breast cancer.Results: Based on 1574 incident premenopausal breast cancer cases and 1 133 893 person-years of follow-up, a lower incidence was associated with lower birthweight: hazard ratio (HR) [95% confidence interval (CI)] = 0.74 (0.58-0.95) for <2.5kg vs 3.9+kg, P for trend < 0.001; higher somatotype at age 5: HR=0.57 (95% CI 0.44-0.73) for 5-9 vs 1, P fortrend < 0.0001]; and at age 10: HR=0.61 (95% CI 0.49-0.75) for 5-9 vs 1, P for trend < 0.0001]; and BMI at age 18: HR=0.67 (95% 0.47-0.95) for ≥ 27.5 kg/m2 vs < 18.5 kg/m2, P for trend = 0.009], after adjusting for age and body fatness measures earlier in life and other risk factors, respectively. No significant interaction between body fatness measures was found. Women with the lowest birthweight, the highest somatotype at ages 5 and 10 and the highest BMI at age 18 and currently had a 72% (95% CI 54%-83%) lower incidence of invasive premenopausal breast cancer than women with the opposite extreme of each body fatness indicator.Conclusion: The lowest incidence of premenopausal breast cancer was associated with the lowest birthweight and the highest childhood, adolescent and early adult body fatness. [ABSTRACT FROM AUTHOR]

Background: The role of body fatness in the aetiology of breast cancer is complex. We evaluated the independent and synergistic effects of body fatness, at different stages throughout a woman's life course, on premenopausal breast cancer risk. Methods: Premenopausal participants of the Nurses' Health Study II (NHSII) were followed from 1991 up to 2009. Body fatness factors including birthweight, somatotype (a 9-level pictogram with level 1 being the leanest) at ages 5 and 10 years and body mass index (BMI) at age 18 were collected at baseline. Current BMI was updated biennially. Multivariate Cox regression models were used to evaluate the association between each body fatness factor as well as cross-classification of all factors and the incidence of breast cancer. Results: Based on 1574 incident premenopausal breast cancer cases and 1 133 893 personyears of follow-up, a lower incidence was associated with lower birthweight: hazard ratio (HR) [95% confidence interval (CI)]=0.74 (0.58-0.95) for <2.5kg vs 3.9þkg, P for trend<0.001; higher somatotype at age 5: HR=0.57 (95% CI 0.44-0.73) for 5-9 vs 1, P fortrend<0.0001]; and at age 10: HR=0.61 (95% CI 0.49-0.75) for 5-9 vs 1, P for trend<0.0001]; and BMI at age 18: HR=0.67 (95% 0.47-0.95) for ≥ 27.5 kg/m² vs < 18.5 kg/m², P for trend=0.009], after adjusting for age and body fatness measures earlier in life and other risk factors, respectively. No significant interaction between body fatness measures was found. Women with the lowest birthweight, the highest somatotype at ages 5 and 10 and the highest BMI at age 18 and currently had a 72% (95% CI 54%-83%) lower incidence of invasive premenopausal breast cancer than women with the opposite extreme of each body fatness indicator. Conclusion: The lowest incidence of premenopausal breast cancer was associated with the lowest birthweight and the highest childhood, adolescent and early adult body fatness. [ABSTRACT FROM AUTHOR]

Participant‐level meta‐analysis across multiple studies increases the sample size for pooled analyses, thereby improving precision in effect estimates and enabling subgroup analyses. For analyses involving biomarker measurements as an exposure of interest, investigators must first calibrate the data to address measurement variability arising from usage of different laboratories and/or assays. In practice, the calibration process involves reassaying a random subset of biospecimens from each study at a central laboratory and fitting models that relate the study‐specific "local" and central laboratory measurements. Previous work in this area treats the calibration process from the perspective of measurement error techniques and imputes the estimated central laboratory value among individuals with only a local laboratory measurement. In this work, we propose a repeated measures method to calibrate biomarker measurements pooled from multiple studies with study‐specific calibration subsets. We account for correlation between measurements made on the same person and between measurements made at the same laboratory. We demonstrate that the repeated measures approach provides valid inference, and compare it to existing calibration approaches grounded in measurement error techniques in an example describing the association between circulating vitamin D and stroke. [ABSTRACT FROM AUTHOR]

Screening mammography aims to identify breast cancer early and secondarily measures breast density to classify women at higher or lower than average risk for future breast cancer in the general population. Despite the strong association of individual mammography features to breast cancer risk, the statistical literature on mammogram imaging data is limited. While functional principal component analysis (FPCA) has been studied in the literature for extracting image‐based features, it is conducted independently of the time‐to‐event response variable. With the consideration of building a prognostic model for precision prevention, we present a set of flexible methods, supervised FPCA (sFPCA) and functional partial least squares (FPLS), to extract image‐based features associated with the failure time while accommodating the added complication from right censoring. Throughout the article, we hope to demonstrate that one method is favored over the other under different clinical setups. The proposed methods are applied to the motivating data set from the Joanne Knight Breast Health cohort at Siteman Cancer Center. Our approaches not only obtain the best prediction performance compared to the benchmark model, but also reveal different risk patterns within the mammograms. [ABSTRACT FROM AUTHOR]

Background Endometrial cancer risk stratification may help target interventions, screening, or prophylactic hysterectomy to mitigate the rising burden of this cancer. However, existing prediction models have been developed in select cohorts and have not considered genetic factors. Methods We developed endometrial cancer risk prediction models using data on postmenopausal White women aged 45-85 years from 19 case-control studies in the Epidemiology of Endometrial Cancer Consortium (E2C2). Relative risk estimates for predictors were combined with age-specific endometrial cancer incidence rates and estimates for the underlying risk factor distribution. We externally validated the models in 3 cohorts: Nurses' Health Study (NHS), NHS II, and the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Results Area under the receiver operating characteristic curves for the epidemiologic model ranged from 0.64 (95% confidence interval [CI] = 0.62 to 0.67) to 0.69 (95% CI = 0.66 to 0.72). Improvements in discrimination from the addition of genetic factors were modest (no change in area under the receiver operating characteristic curves in NHS; PLCO = 0.64 to 0.66). The epidemiologic model was well calibrated in NHS II (overall expected-to-observed ratio [E/O]   =   1.09, 95% CI = 0.98 to 1.22) and PLCO (overall E/O   =   1.04, 95% CI = 0.95 to 1.13) but poorly calibrated in NHS (overall E/O   =   0.55, 95% CI = 0.51 to 0.59). Conclusions Using data from the largest, most heterogeneous study population to date (to our knowledge), prediction models based on epidemiologic factors alone successfully identified women at high risk of endometrial cancer. Genetic factors offered limited improvements in discrimination. Further work is needed to refine this tool for clinical or public health practice and expand these models to multiethnic populations. [ABSTRACT FROM AUTHOR]

Matrix metalloproteinase 2 (MMP2) is an enzyme with important functions in breast cancer invasion and metastasis. However, it is unclear whether circulating MMP2 levels may predict breast cancer risk. We conducted a prospective nested case-control analysis in the Nurses' Health Study among 1136 cases who were diagnosed with invasive breast cancer between 1992 and 2004 and 1136 matched controls. All participants provided blood samples in 1989-1990, and a subset (170 cases, 170 controls) contributed an additional sample in 2000-2002. Pre-diagnostic plasma MMP2 levels were measured via immunoassay, and conditional logistic regression was performed to calculate odds ratios (ORs) and 95% confidence intervals (95% CIs), adjusted for breast cancer risk factors. No association was observed between plasma MMP2 levels and risk of total invasive breast cancer (top vs. bottom quartile, OR = 1.0; 95% CI: 0.7, 1.2; ptrend = 0.89). Findings did not vary significantly by time since blood draw, body mass index, postmenopausal hormone use, or menopausal status at either blood draw or breast cancer diagnosis. MMP2 was associated with a greater risk of nodal metastases at diagnosis (top vs. bottom quartile, OR = 1.5; 95% CI: 1.0, 2.2; p-heterogeneity, any vs. no lymph nodes = 0.002), but no significant associations were observed with other tumor characteristics or with recurrent or fatal cancers. Plasma MMP2 levels do not appear to be predictive of total invasive breast cancer risk, although associations with aggressive disease warrant further study. [ABSTRACT FROM AUTHOR]

Nutrient intake is often measured with substantial error both in commonly used surrogate instruments such as a food frequency questionnaire (FFQ) and in gold standard-type instruments such as a diet record (DR). If there is a correlated error between the FFQ and DR, then standard measurement error correction methods based on regression calibration can produce biased estimates of the regression coefficient ( λ) of true intake on surrogate intake. However, if a biomarker exists and the error in the biomarker is independent of the error in the FFQ and DR, then the method of triads can be used to obtain unbiased estimates of λ, provided that there are replicate biomarker data on at least a subsample of validation study subjects. Because biomarker measurements are expensive, for a fixed budget, one can use a either design where a large number of subjects have one biomarker measure and only a small subsample is replicated or a design that has a smaller number of subjects and has most or all subjects validated. The purpose of this paper is to optimize the proportion of subjects with replicated biomarker measures, where optimization is with respect to minimizing the variance of [ABSTRACT FROM AUTHOR]

Relationships between some risk factors and breast cancer incidence are known to vary by tumor subtype. However, breast tumors can be classified according to a number of markers, which may be correlated, making it difficult to identify heterogeneity of risk factors with specific tumor markers when using standard competing-risk survival analysis. In this paper, we propose a constrained competing-risk survival model that allows for assessment of heterogeneity of risk factor associations according to specific tumor markers while controlling for other markers. These methods are applied to Nurses’ Health Study data from 1980–2006, during which 3,398 incident invasive breast cancers occurred over 1.4 million person-years of follow-up. Results suggested that when estrogen receptor (ER) and progesterone receptor (PR) status are mutually considered, some risk factors thought to be characteristic of “estrogen-positive tumors,” such as high body mass index during postmenopause and increased height, are actually significantly associated with PR-positive tumors but not ER-positive tumors, while other risk factors thought to be characteristic of “estrogen-negative tumors,” such as late age at first birth, are actually significantly associated with PR-negative rather than ER-negative breast cancer. This approach provides a strategy for evaluating heterogeneity of risk factor associations by tumor marker levels while controlling for additional tumor markers. [ABSTRACT FROM PUBLISHER]

Objective: To investigate associations between dietary omega-3 fatty acids and other fat intake, genes related to age-related macular degeneration (AMD), and progression to geographic atrophy (GA). Design: Observational analysis of a prospective cohort. Participants: A total of 2531 individuals from the Age-Related Eye Disease Study, among which 525 eyes progressed to GA and 4165 eyes did not. Methods: Eyes without advanced AMD at baseline were evaluated for progression to GA. Behavioral data, including smoking and body mass index measurements, were collected at baseline using questionnaires. Dietary data were collected from food frequency questionnaires (FFQs) at baseline. Omega-3 fatty acids (docosahexaenoic acid [DHA] and eicosapentaenoic acid [EPA]), omega-6 fatty acids, monounsaturated, saturated, polyunsaturated, and total fat were adjusted for sex and calories and divided into quintiles (Q). Eight single nucleotide polymorphisms in 7 genes (CFH, ARMS2/HTRA1, CFB, C2, C3, CFI, and LIPC) were genotyped. Cox proportional hazards models were used to test for associations between incident GA and intake of dietary lipids and interaction effects between dietary fat intake and genetic variation on risk of GA. Main Outcome Measures: Associations between dietary fat intake reported from FFQs, genetic variants, and incident GA. Results: Increased intake of DHA was significantly associated with reduced risk of progression to GA in models with behavioral factors (model A) plus genetic variants (model B) (P trend = 0.01 and 0.03, respectively). Total omega-3 long chain polyunsaturated (DHA + EPA) fatty acid intake was significantly associated with reduced risk of progression in model B (P trend = 0.02). Monounsaturated fat was associated with increased risk in model A (P trend = 0.05). DHA intake was significantly associated with reduced risk of incident GA among those with the ARMS2/HTRA1 homozygous risk genotype (hazard ratio [HR] Q5 vs Q1, 0.4; P = 0.002; P for interaction between gene and fat intake = 0.05). DHA was not associated with reduced risk of GA among those with the homozygous ARMS2/HTRA1 nonrisk genotype (HR, 1.0; P = 0.90). Conclusions: Increased self-reported dietary intake of omega-3 fatty acids is associated with reduced risk of GA and may modify genetic susceptibility for progression to GA. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. [Copyright &y& Elsevier]

Objective: A study was undertaken to relate dietary fat types to cognitive change in healthy community-based elders. Methods: Among 6,183 older participants in the Women's Health Study, we related intake of major fatty acids (saturated [SFA], monounsaturated [MUFA], total polyunsaturated [PUFA], trans-unsaturated) to late-life cognitive trajectory. Serial cognitive testing, conducted over 4 years, began 5 years after dietary assessment. Primary outcomes were global cognition (averaging tests of general cognition, verbal memory, and semantic fluency) and verbal memory (averaging tests of recall). We used analyses of response profiles and logistic regression to estimate multivariate-adjusted differences in cognitive trajectory and risk of worst cognitive change (worst 10%) by fat intake. Results: Higher SFA intake was associated with worse global cognitive ( p for linear trend = 0.008) and verbal memory ( p for linear trend = 0.01) trajectories. There was a higher risk of worst cognitive change, comparing highest versus lowest SFA quintiles; the multivariate-adjusted odds ratio (OR) with 95% confidence interval (CI) was 1.64 (1.04-2.58) for global cognition and 1.65 (1.04-2.61) for verbal memory. By contrast, higher MUFA intake was related to better global cognitive ( p for linear trend < 0.001) and verbal memory ( p for linear trend = 0.009) trajectories, and lower OR (95% CI) of worst cognitive change in global cognition (0.52 [0.31-0.88]) and verbal memory (0.56 [0.34-0.94]). Total fat, PUFA, and trans-fat intakes were not associated with cognitive trajectory. Interpretation: Higher SFA intake was associated with worse global cognitive and verbal memory trajectories, whereas higher MUFA intake was related to better trajectories. Thus, different consumption levels of the major specific fat types, rather than total fat intake itself, appeared to influence cognitive aging. ANN NEUROL 2012; [ABSTRACT FROM AUTHOR]

Purpose: The eye is often the unit of measurement for outcomes, and frequently also for covariates, in vision research, and measurements in the two eyes of the same person are often strongly but far from perfectly correlated. Advances have occurred in the development and accessibility of analytic approaches to evaluate determinants of eye-specific outcomes including information from both eyes of some subjects. Methods: We illustrate available regression approaches to analyze correlated outcomes from both eyes in datasets with both eye- and subject-specific exposures and potential confounding variables. We consider cross-sectional and longitudinal study designs, and discrete, continuous, and time-to-event outcomes. Results: Across a range of study designs and measurement scales for the outcome variable, we show the under-estimation of P-values and widths of confidence intervals that occurs when the correlation between paired eyes in a person is ignored, and the reduced precision that occurs in separate analyses of right or left eyes, or in analyses of persons rather than eyes. By comparison, regression models with the eye as the unit of analysis and appropriate consideration of the correlation between paired outcomes generally offer maximal use of available data, enhanced interpretability of covariate-outcome associations, and efficient use of information from subjects who contribute only one eye to analyses. Conclusions: For many studies in vision research, the now widely available regression models that appropriately treat the eye as the unit of analysis offer the best analytic approach. [ABSTRACT FROM AUTHOR]

The article discusses a prospective cohort study of the relationship between breast cancer risk (BCR) and moderate alcohol consumption during adult life (AL). The authors analyzed BCR data for 105,986 women enrolled in the Nurses' Health Study from 1980 to 2008. They observed that there was a ten percent increase in BCR with every ten grams daily intake of alcohol. They found that binge drinking was linked with BCR after controlling for cumulative alcohol intake (AI). They concluded that AI, both earlier and later in AL, was separately associated with risk.

The Wilcoxon rank sum test is widely used for two-group comparisons of nonnormal data. An assumption of this test is independence of sampling units both within and between groups, which will be violated in the clustered data setting such as in ophthalmological clinical trials, where the unit of randomization is the subject, but the unit of analysis is the individual eye. For this purpose, we have proposed the clustered Wilcoxon test to account for clustering among multiple subunits within the same cluster (, Biometrics , 1089-1098; 2006, Biometrics , 1251-1259). However, power estimation is needed to plan studies that use this analytic approach. We have recently published methods for estimating power and sample size for the ordinary Wilcoxon rank sum test (, Biometrics , 188-197). In this article we present extensions of this approach to estimate power for the clustered Wilcoxon test. Simulation studies show a good agreement between estimated and empirical power. These methods are illustrated with examples from randomized trials in ophthalmology. Enhanced power is achieved with use of the subunit as the unit of analysis instead of the cluster using the ordinary Wilcoxon rank sum test. [ABSTRACT FROM AUTHOR]

Purpose: Age at menopause, a major marker in the reproductive life, may bias results for evaluation of breast cancer risk after menopause. Methods: We followed 38,948 premenopausal women in 1980 and identified 2,586 who reported hysterectomy without bilateral oophorectomy and 31,626 who reported natural menopause during 22 years of follow-up. We evaluated risk factors for natural menopause, imputed age at natural menopause for women reporting a hysterectomy without bilateral oophorectomy, and estimated the hazard of reaching natural menopause in the next 2 years. We applied this imputed age at menopause to both increase sample size and to evaluate the relation between postmenopausal exposures and risk of breast cancer. Results: Age, cigarette smoking, age at menarche, pregnancy history, body mass index, history of benign breast disease, and history of breast cancer were each significantly related to age at natural menopause; duration of oral contraceptive use and family history of breast cancer were not. The imputation increased sample size substantially, and although some risk factors after menopause were weaker in the expanded model (height, and alcohol use), use of hormone therapy is less biased. Conclusions: Imputing age at menopause increases sample size, broadens generalizability making it applicable to women with hysterectomy, and reduces bias. [ABSTRACT FROM AUTHOR]

The articles presents information on a study which evaluated whether incorporating benign breast diseases (BBD) histologic lesions would improve the discriminatory power of risk classification models in breast cancer. A review of the related literature on BBD is offered. Data analyzed were from the Nurses' Health Study. The findings of the study are presented. It concludes that the clinical utility of BBD model will depend on the consistency of histologic classification of BBD lesions.

Objective: A genetic variant in the high-density lipoprotein (HDL) cholesterol pathway, hepatic lipase (LIPC), was discovered to be associated with advanced age-related macular degeneration (AMD) in a genome-wide association study. In this study, we evaluated whether LIPC is associated with serum lipids, and whether this gene and serum lipids are independently associated with AMD. Design: Case-control study. Participants: A total of 458 participants from the Progression Study of Macular Degeneration and the Age-Related Eye Disease Ancillary Biomarker Study, including 318 advanced AMD cases with either geographic atrophy (n = 123) or neovascular disease (n = 195) and 140 controls. Methods: Participants were genotyped for 8 variants associated with AMD: 2 CFH variants, C2, CFB, C3, CFI, the ARMS2/HTRA1 gene region, and LIPC. Fasting blood specimens were obtained at study onset, and serum levels of total cholesterol, low-density lipoprotein (LDL), HDL, and triglycerides were determined. Logistic and linear regression were used to evaluate associations between serum lipids, LIPC genotype, and AMD. Main Outcome Measures: LIPC and serum lipid associations with AMD. Results: The minor T allele of the LIPC gene was associated with a reduced risk of AMD (odds ratio, 0.4; 95% confidence interval, 0.2–0.9; P = 0.01, trend for number of T alleles, controlling for age and gender). Mean level of HDL was lower (P = 0.05) and mean level of LDL (P = 0.03) was higher in cases of advanced AMD compared with controls. Higher total cholesterol and LDL levels were associated with increased risk of AMD, with P for trend = 0.01 for both, in models controlling for environmental and genetic covariates. The T allele of LIPC was associated with higher levels of HDL, although LIPC was associated with advanced AMD independent of HDL level. Conclusions: The HDL-raising allele of the LIPC gene (T) was associated with a reduced risk of AMD. Higher total cholesterol and LDL levels were associated with increased risk, whereas higher HDL levels tended to reduce the risk of AMD. The specific mechanisms underlying the association between AMD and LIPC require further investigation. Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references. [Copyright &y& Elsevier]

Previous epidemiologic studies suggest that the major histologic subtypes of epithelial ovarian cancer may have different risk factor profiles; however, no known prospective study has systematically examined differences in risk by subtype. The authors used Cox proportional hazards regression, stratified by histologic subtype and time period, to examine the association between ovarian cancer risk factors and incidence of serous invasive, endometrioid, and mucinous ovarian cancers in the US Nurses’ Health Study (1976–2006) and Nurses’ Health Study II (1989–2005). For each exposure, they calculated P-heterogeneity using a likelihood ratio test comparing models with separate estimates for the 3 subtypes versus a single estimate across subtypes. Analysis included 221,866 women and 721 cases with the histologies of interest (496 serous invasive, 139 endometrioid, 86 mucinous). In analyses of reproductive/hormonal exposures, the associations with age, duration of breastfeeding, age at natural menopause, and duration of estrogen use differed significantly by subtype (all P-heterogeneity ≤0.05). The associations with several nonreproductive exposures also appeared to vary by subtype, but only the association with smoking differed significantly (P-heterogeneity = 0.03). Results suggest that associations with several ovarian cancer risk factors vary by subtype, and these differences are consistent with known similarities between each major histologic subtype and its normal tissue counterpart. [ABSTRACT FROM PUBLISHER]

Few studies have evaluated the validity of adolescent diet recall after many decades. Between 1943 and 1970, yearly diet records were completed by parents of adolescents participating in an ongoing US study. In 2005–2006, study participants who had been 13–18 years of age when the diet records were collected were asked to complete a food frequency questionnaire regarding their adolescent diet. Food frequency questionnaires and diet records were available for 72 participants. The authors calculated Spearman correlation coefficients between food, food group, and nutrient intakes from the diet records and food frequency questionnaire and deattenuated them to account for the effects of within-person variation measured in the diet records on the association. The median deattenuated correlation for foods was 0.30, ranging from −0.53 for a beef, pork, or lamb sandwich to 0.99 for diet soda. The median deattenuated correlation for food groups was 0.31 (range: −0.48 for breads to 0.70 for hot beverages); for nutrient intakes, it was 0.25 (range: −0.08 for iron to 0.82 for vitamin B12). Some dietary factors were reasonably recalled 3–6 decades later. However, this food frequency questionnaire did not validly measure overall adolescent diet when completed by middle-aged and older adults on average 48 years after adolescence. [ABSTRACT FROM PUBLISHER]

Screening mammography can distort estimated effects in breast cancer risk models due to associations with other risk factors. Mammography information was available in the Nurses’ Health Study from 1988, and 1,815 incident breast cancers were accrued through 2000 among 55,625 women with risk factor data. Logistic models were fit for screening mammography, and inverse probability weighting was used to adjust parameters in an established breast cancer risk model. Approximately 80% of women in each 2-year follow-up period had screening mammograms, which were positively associated with history of benign breast disease, family history of breast cancer, hormone therapy, alcohol use, physical activity, multivitamins, and calcium supplements, and negatively associated with postmenopause, current smoking, and body mass index. Markers of medical attention, including hypertension, high cholesterol, and osteoarthritis, were positively associated, while cardiovascular disease was negative. Inverse probability weighting led to small changes in effects of benign breast disease, family history, and hormone therapy. An apparent reduced risk associated with current smoking in unadjusted models was eliminated after weighting. Thus, several risk factors for breast cancer and cancer diagnosis are associated with mammographic screening. Adjustment for screening had some impact on breast cancer prediction in this cohort, especially for hormone therapy and smoking. [ABSTRACT FROM PUBLISHER]

Purpose: In prospective studies of incident cataract, a person may have developed two or all three of cortical, nuclear sclerotic and posterior subscapsular cataract at the time when clinically significant cataract is apparent. One can compare the impact of risk factors on different types of cataract with methods of competing risk survival analysis that account for tied events. Methods: We studied risk factors for incident cataract among 20,599 participants in the Physicians' Health Study; 1,923 of whom developed cataract that reduced visual acuity over a median follow-up of 13.0 years. Among incident cases, 45% had two types of cataract and an additional 12% had all three. Proportional hazards models, adjusted for tied failures through use of generalized estimating equations, provided comparisons across different cataract types. Stratified analyses by type with age as the time scale were used to accommodate the strong but different relationships of age with occurrence of different cataract types. Results: Analyses found different relationships of diabetes and body mass index, but similar relationships of smoking, family history of coronary disease and multivitamin use, across different cataract types. A simplified model provided common estimates for relative risks associated with those characteristics with similar effects across cataract types. Conclusions: Whereas polytomous logistic regression has proven useful for comparisons of risks across cataract types in retrospective and cross-sectional studies, this approach based on methods of survival analysis has greater flexibility in prospective studies with variable follow-up and frequently tied event times for different cataract types. [ABSTRACT FROM AUTHOR]

The authors compared, in the context of diabetic retinopathy, alternative methods of quantifying the extent to which disease progression in one eye increases the risk of subsequent progression in the other eye. Data were gathered on 478 US patients with insulin-dependent diabetes mellitus who participated in the 1983–1988 Sorbinil Retinopathy Trial and were followed up for a median of 41 months. During that time, diabetic retinopathy progressed in 93 right eyes and 77 left eyes. Crude incidence rates of progression for right eyes were 7.7 times higher after the left eye had progressed and, for left eyes, were 4.4 times higher after the right eye had progressed. In eye-specific proportional hazards models that adjusted for increasing rates of progression over time and for baseline risk factors, the comparable relative risks associated with progression in the other eye were 2.6 (95% confidence interval (Cl): 1.5, 4.7) for right eyes and 1.4 (95% Cl: 0.72, 2.9) for left eyes. Two alternative proportional hazards models that included data on both eyes and accounted for their correlation produced estimated relative risks of 1.9 (95% Cl: 1.2, 2.9) and 2.7 (95% Cl: 1.8, 3.5), respectively. The more complex models for joint survival integrate information on both eyes and provide more stable estimates than do separate analyses of right or left eyes. Am J Epidemiol 2000; 151: 965-74. [ABSTRACT FROM PUBLISHER]

No large national studies of ethnic differences in blood pressure among children accounting for body size differences have been published, to the author's knowledge. This report details the similarities and differences in systolic and diastolic blood pressures betweenn Black children and White children in the United States and examines the effects of age, sex, and body size on ethnic differences in blood pressure levels. Standardized measurements of seated systolic and diastolic pressures from eight large epidemiologic studies published between 1978 and 1991 that included measurements of 47, 196 children on 68, 556 occasions for systolic pressure and for 38, 184 children on 52, 053 occasions for diastolic pressure were used; 51 percent (24, 048 children) were boys and 37 percent (17, 466 children) were Black. Overall, there appear to be few substantive ethnic differences in either systolic or diastolic pressure during childhood and adolescence. The differences that were observed were small, inconsistent, and often expalained by differences in body size. There was an ethnic group-body mass index (BMI) interaction that resulted in these findings that at lower levels of BMI Blacks have higher blood pressure and more hypertension than do Whites, but that at the highest levels of BMI, Whites have more hypertension (systolic or diastolic pressure) than do Blacks. Am J Epidemiol 2000: 151: 1007-19. [ABSTRACT FROM PUBLISHER]

Nutritional exposures are often measured with considerable error in commonly used surrogate instruments such as the food frequency questionnaire (FFQ) (denoted by Q i for the ith subject). The error can be both systematic and random. The diet record (DR) denoted by R i for the ith subject is considered an alloyed gold standard. However, some authors have reported both systematic and random errors with this instrument as well. One goal in measurement error research is to estimate the regression coefficient of T i (true intake for the ith subject) on Q i denoted by λ TQ. If the systematic errors in Q i and R i (denoted by q i and r i) are uncorrelated, then one can obtain an unbiased estimate of λ TQ by λ RQ obtained by regressing R i on Q i. However, if Corr( q i, r i)>0, then λ RQ>λ TQ. In this paper, we propose a method for indirectly estimating λ TQ even in the presence of correlated systematic error based on a longitudinal design where Q i (surrogate measure of dietary intake), R i (a reference measure of dietary intake), and M i (a biomarker) are available on the same subjects at 2 time points. In addition, between-person variation in mean levels of M i among people with the same dietary intake is also accounted for. The methodology is illustrated for dietary vitamin C intake based on longitudinal data from 323 subjects in the European Prospective Investigation of Cancer (EPIC)-Norfolk study who provided two measures of dietary vitamin C intake from the FFQ ( Q i) and a 7-day DR ( R i) and plasma vitamin C ( M i) 4 years apart. Copyright © 2008 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Purpose: To extend the Wilcoxon rank sum test to the analysis of clustered observational data with a binary exposure variable where not all cluster members (or subunits) have the same exposure status, as in ophthalmologic data where fellow eyes may differ in exposure status. Methods: The Wilcoxon rank sum test statistic (Wc) is defined as the sum of ranks of all exposed subunits over all clusters. The large sample distribution of Wc is derived based on a two-stage permutation distribution approach which maintains the clustering structure of the exposure distribution in the original data. The procedure is applied to data from the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR) for assessing differences in retinopathy grade between eyes with high vs. low intraocular pressure (IOP). Results: Data analyses based on the WESDR data indicate significant positive associations between IOP and retinopathy grade as assessed by the clustered Wilcoxon test (p = 0.019), but non-significant effects with mixed model procedures based on raw scores (p = 0.30). Conclusions: The clustered Wilcoxon procedure is an effective approach for analyzing clustered non-normal outcome data with a binary eye-specific exposure which may vary in 2 eyes of an individual. [ABSTRACT FROM AUTHOR]

The article discusses following the cardiovascular disease paradigm in which interventions are targeted according to the estimated risk of disease development as opposed to applying a one size fits all approach to breast cancer chemoprevention. Tamoxifen and raloxifene have been shown to reduce the incidence of invasive breast cancer by approximately 50%, but their use has been limited due to concerns over adverse effects. Using epidemiologic risk models is suggested to estimate breast cancer risk.

The Spearman (ρ s) and Kendall (τ) rank correlation coefficient are routinely used as measures of association between non-normally distributed random variables. However, confidence limits for ρ s are only available under the assumption of bivariate normality and for τ under the assumption of asymptotic normality of . In this paper, we introduce another approach for obtaining confidence limits for ρ s or τ based on the arcsin transformation of sample probit score correlations. This approach is shown to be applicable for an arbitrary bivariate distribution. The arcsin-based estimators for ρ s and τ (denoted by s, a, a) are shown to have asymptotic relative efficiency (ARE) of 9/π2 compared with the usual estimators s and when ρ s and τ are, respectively, 0. In some nutritional applications, the Spearman rank correlation between nutrient intake as assessed by a reference instrument versus nutrient intake as assessed by a surrogate instrument is used as a measure of validity of the surrogate instrument. However, if only a single replicate (or a few replicates) are available for the reference instrument, then the estimated Spearman rank correlation will be downwardly biased due to measurement error. In this paper, we use the probit transformation as a tool for specifying an ANOVA-type model for replicate ranked data resulting in a point and interval estimate of a measurement error corrected rank correlation. This extends previous work by Rosner and Willett for obtaining point and interval estimates of measurement error corrected Pearson correlations. Copyright © 2006 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

The Wilcoxon rank sum test is widely used for two-group comparisons for nonnormal data. An assumption of this test is independence of sampling units both between and within groups. In ophthalmology, data are often collected on two eyes of an individual, which are highly correlated. In ophthalmological clinical trials, randomization is usually performed at the subject level, but the unit of analysis is the eye. If the eye is used as the unit of analysis, then a modification to the usual Wilcoxon rank sum variance formula must be made to account for the within-cluster dependence. For some clustered data designs, where the unit of analysis is the subunit, group membership may be defined at the subunit level. For example, in some randomized ophthalmologic clinical trials, different treatments may be applied to fellow eyes of some patients, while the same treatment may be applied to fellow eyes of other patients. In general, binary eye-specific covariates may be present (scored as exposed or unexposed) and one wishes to compare nonnormally distributed outcomes between exposed and unexposed eyes using the Wilcoxon rank sum test while accounting for the clustering. In this article, we present a corrected variance formula for the Wilcoxon rank sum statistic in the setting of eye (subunit)-specific covariates. We apply it to compare ocular itching scores in ocular allergy patients between eyes treated with active versus placebo eye drops, where some patients receive the same eye drop in both eyes, while other patients receive different eye drops in fellow eyes. We also present comparisons between the clustered Wilcoxon test and each of the signed rank tests and mixed model approaches and show dramatic differences in power in favor of the clustered Wilcoxon test for some designs. [ABSTRACT FROM AUTHOR]

The Wilcoxon signed rank test is a frequently used nonparametric test for paired data (e.g., consisting of pre- and posttreatment measurements) based on independent units of analysis. This test cannot be used for paired comparisons arising from clustered data (e.g., if paired comparisons are available for each of two eyes of an individual). To incorporate clustering, a generalization of the randomization test formulation for the signed rank test is proposed, where the unit of randomization is at the cluster level (e.g., person), while the individual paired units of analysis are at the subunit within cluster level (e.g., eye within person). An adjusted variance estimate of the signed rank test statistic is then derived, which can be used for either balanced (same number of subunits per cluster) or unbalanced (different number of subunits per cluster) data, with an exchangeable correlation structure, with or without tied values. The resulting test statistic is shown to be asymptotically normal as the number of clusters becomes large, if the cluster size is bounded. Simulation studies are performed based on simulating correlated ranked data from a signed log-normal distribution. These studies indicate appropriate type I error for data sets with ≥20 clusters and a superior power profile compared with either the ordinary signed rank test based on the average cluster difference score or the multivariate signed rank test of Puri and Sen (1971 , Nonparametric Methods in Multivariate Analysis, New York: John Wiley). Finally, the methods are illustrated with two data sets, (i) an ophthalmologic data set involving a comparison of electroretinogram (ERG) data in retinitis pigmentosa (RP) patients before and after undergoing an experimental surgical procedure, and (ii) a nutritional data set based on a randomized prospective study of nutritional supplements in RP patients where vitamin E intake outside of study capsules is compared before and after randomization to monitor compliance with nutritional protocols. [ABSTRACT FROM AUTHOR]

Estimates of diet-disease relative risks (RRs) suffer from inaccuracies introduced by dietary measurement errors. Using the "method of triads," by which the validity coefficient (VC) of the dietary assessment method and "true" long-term intake could be estimated from 3 pairwise correlations between the FFQ, the reference method, and the biomarker, the authors evaluated the performance of a newly developed FFQ. Over a period of 13 mo (September 2000 to September 2001), 161 participants completed 3 FFQs and six 24-h recalls (24HRs), and supplied 2 blood samples and three 24-h urine collections. For protein, beta-carotene, and folic acid, the VCs of the FFQ with the "true intake" (0.77, 0.65, and 0.72, respectively) were relatively higher than the VCs of 24HRs (0.68, 0.60 and 0.39, respectively). Among the biomarkers, the VCs of serum beta-carotene and folic acid with the "true intake" (0.65 and 0.65) were higher than the VCs of urinary nitrogen and alpha-tocopherol (0.44 and 0.34, respectively). The DEARR study showed that the newly developed FFQ is a valid and reproducible instrument for assessing dietary intake. The VCs obtained can be used for future adjustment of diet-disease RR estimates in this population. [ABSTRACT FROM AUTHOR]

Tracking correlations of blood pressure, particularly childhood measures, may be attenuated by within-person variability. Combining multiple measurements can reduce this error substantially. The area under the curve (AUC) computed from longitudinal growth curve models can be used to improve the prediction of young adult blood pressure from childhood measures. Quadratic random-effects models over unequally spaced repeated measures were used to compute the area under the curve separately within the age periods 5-14 and 20-34 years in the Bogalusa Heart Study. This method adjusts for the uneven age distribution and captures the underlying or average blood pressure, leading to improved estimates of correlation and risk prediction. Tracking correlations were computed by race and gender, and were approximately 0.6 for systolic, 0.5-0.6 for K4 diastolic, and 0.4-0.6 for K5 diastolic blood pressure. The AUC can also be used to regress young adult blood pressure on childhood blood pressure and childhood and young adult body mass index (BMI). In these data, while childhood blood pressure and young adult BMI were generally directly predictive of young adult blood pressure, childhood BMI was negatively correlated with young adult blood pressure when childhood blood pressure was in the model. In addition, racial differences in young adult blood pressure were reduced, but not eliminated, after controlling for childhood blood pressure, childhood BMI, and young adult BMI, suggesting that other genetic or lifestyle factors contribute to this difference. Copyright © 2004 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

In many ophthalmologic studies, progression of diseases such as diabetic retinopathy, age-related maculopathy, cataract, and glaucoma is only noted when each eye is examined at intervals that commonly vary between subjects. Such data are often analysed using continuous time survival methods with observed progression assumed to occur at the end of the interval. Tied times of progression can lead to substantial bias in estimation of the association between progression in right and left eyes. We describe a multiple imputation strategy to create multiple data sets without ties, based on drawing interval-censored progression times from a parametric gamma frailty model that accounts for continuous and discrete covariates. We illustrate the method with data from 478 patients with insulin-dependent diabetes mellitus who were followed for progression of diabetic retinopathy in the Sorbinil Retinopathy Trial. Resolution of tied failure times allows for valid estimation of the hazard of progression in one eye given the progression status of the other eye. A simulation study suggests that the method performs well. Results highlight the advantage of multiple imputation that data imputed under one model can be analysed under several alternative models. Copyright © 2004 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Background: Evaluations of epidemiologic risk factors in relation to breast cancer classified jointly by estrogen receptor (ER) and progesterone receptor (PR) status have been inconsistent. To address this issue, we conducted a prospective evaluation of risk factors for breast cancer classified according to receptor status. Methods: During 1 029 414 person-years of follow-up of 66 145 women participating in the Nurses' Health Study from 1980 through 2000, we identified 2096 incident cases of breast cancer for which information on ER/PR status was available: 1281 were ER+/ PR+, 318 were ER+/PR-, 80 were ER-/PR+, and 417 were ER-/PR-. We fit a log-incidence model of breast cancer and used polychotomous logistic regression to compare coefficients for breast cancer risk factors in patients with different ER/PR status. To test for differences in risk factor odds ratios based on marginal ER/PR categories, we evaluated ER status controlling for PR status and vice versa. The predictive ability of our log-incidence model to discriminate between women who would develop ER+/PR+ breast cancer and those who would not (and similarly for ER-/ PR-breast cancer) was evaluated by using receiver operator characteristic curve analysis. All statistical tests were two-sided. Results: We observed statistically significant heterogeneity among the four ER/PR categories for some risk factors (age, menopausal status, body mass index [BMI] after menopause, the one-time adverse effect of first pregnancy, and past use of postmenopausal hormones) but not for others (benign breast disease, family history of breast cancer, alcohol use, and height). The one-time adverse association of first pregnancy with incidence was present for PR- but not for PR+ tumors after controlling for ER status (P = .007). However, the association of BMI after menopause with incidence was present for PR+ but not PR-tumors (P = .005). Statistically significant differences in the incidence of ER+ and ER-tumors were seen with age, both before and after menopause (P = .003), and with past use of postmenopausal hormones (P = .01). Area under the receiver operator characteristic curve, adjusted for age, was 0.64 (95% confidence interval [CI] = 0.63 to 0.66) for ER+/PR+ tumors and 0.61 (95% CI = 0.58 to 0.64) for ER-/PR-tumors. Conclusions: Incidence rates and risk factors for breast cancer differ according to ER and PR status. Thus, to accurately estimate breast cancer risk, breast cancer cases should be divided according to the ER and PR status of the tumor. [ABSTRACT FROM AUTHOR]

Objective: Both randomized and observational studies commonly examine composite end points, but the literature on model development and criticism in this setting is limited.Study Design and Setting: We examined approaches for evaluating heterogeneity in the effects of risk factors for different components of the end point, and determining the impact of heterogeneity on the ability to predict the composite end point. A specific example considered the composite cardiovascular disease end point in the Physicians'' Health Study that occurred in 1,542 (myocardial infarction, n = 716; stroke, n = 557; cardiovascular death, n = 269) of 16,688 participants with complete information on baseline covariates. The strategy compared alternative polytomous logistic regression models assuming different effects of risk factors on components of the end point and a comparable logistic model assuming common effects.Results: Likelihood ratio tests identified heterogeneity in the effects of age, alcohol consumption, and diabetes across components of the outcome, but comparability in the effects of other risk factors. However, a model assuming uniform effects explained over 90% of the log-likelihood change in the best polytomous model, and the two models also performed similarly based on a comparison of ROC curves.Conclusion: The overall strategy may be helpful for evaluating the validity of a composite end point analysis and identifying heterogeneity in risk factors. [Copyright &y& Elsevier]

The Wilcoxon rank sum test is frequently used in statistical practice for the comparison of measures of location when the underlying distributions are far from normal or not known in advance. An assumption of the ordinary rank sum test is that individual sampling units are independent. In many ophthalmologic clinical trials, the Early Treatment for Diabetic Retinopathy Scale (ETDRS) is a principal endpoint used for measuring the level of diabetic retinopathy. This is an ordinal scale, and it is natural to consider the Wilcoxon rank sum test for the comparison of the level of diabetic retinopathy between treatment groups. However, under this design, unlike the usual Wilcoxon rank sum test, the subject is the unit of randomization, but the eye is the unit of analysis. Furthermore, a person will tend to have different, but correlated, ETDRS scores for fellow eyes. Thus, we propose a correction to the variance of the Wilcoxon rank sum statistic that accounts for clustering effects and that can be used for both balanced (same number of subunits per cluster) or unbalanced (different number of subunits per cluster) data, both in the presence or absence of ties, with p-value adjusted accordingly. In this article, we present large-sample theory and simulation results for this test procedure and apply it to diabetic retinopathy data from type I diabetics in the Sorbinil Retinopathy Trial. [ABSTRACT FROM AUTHOR]

In epidemiological studies, continuous covariates often are measured with error and categorical covariates often are misclassified. Using the logistic regression model to represent the relationship between the binary outcome and the perfectly measured and classified covariates, the model for the observed main study data is derived. This derivation relies on the assumption that the error in the continuous covariates is multivariate normally distributed and uses a chain of logistic regression models to describe the misclassification processes. These model assumptions are empirically verified in the validation study, where the misclassified and mismeasured covariates are validated using perfectly measured and classified data. The full data likelihood, including contributions from both the main study and the validation study, is maximized to obtain the maximum likelihood estimates for the parameters of the underlying logistic regression model and of the measurement error model and reclassification models simultaneously. Standard asymptotic theory is applied. An example of this methodology is presented from the Nurses' Health Study investigating the relationship between cumulative incidence of breast cancer and saturated fat, total energy, and alcohol intake. A detailed simulation study was conducted to investigate the small-sample properties of these likelihood-based estimates and inferential quantities. No single estimation/inference option performed satisfactorily when the main study/validation study size was representative of that typically encountered in practice. When the validation size was twice or larger than from the usual one, features of asymptotic optimality were more apparent. By example and through simulation, the procedures appeared to be robust to misspecification of the order of the chain of conditional measurement error/reclassification models. [ABSTRACT FROM AUTHOR]

Develops a model for the distribution of number of batters faced and number of runs scored against starting pitchers in Major League Baseball games based on data from the year 1990. Details of the model; Estimation.

Examines the impact of risk factors for breast cancer among women with and without a family history of the disease. Data on family history; Methods; Parameters from the multiple births model fitted to women; Cumulative risk of breast cancer; Age, susceptibility to DNA damage and breast cancer risk.

Studies the development of a mathematical model to quantify the effects of reproductive risk factors on breast cancer incidence and validation of the model. Methods; Breast cancer log-incidence model based on follow-up from 1976 to 1990; Derivation of the likelihood; Observed and expected relative risk.

Abstract. Previous studies of childhood blood pressure have shown tracking correlations, which estimate the magnitude of association between initial and subsequent measurements, to be lower than corresponding adult values. Inasmuch as this disparity could arise from failing to account for a larger week-to-week variability in children, blood pressure was measured for 4 successive years, on four weekly visits in each year, and with three measurements at each visit, using a random-zero sphygmomanometer, in a cohort of 333 schoolchildren aged 8 through 15 at entry. Ninety percent of subjects had measurements in 1 or more years of follow-up. For all follow-up periods (1, 2, and 3 years from baseline), the Pearson correlation coefficient (r) for both systolic and diastolic blood pressure rose substantially with the number of weekly visits used to calculate each subject's yearly blood pressure (P < .0001). For systolic pressure, the 3-year r values for 1, 2, 3, and 4 visits were .45, .55, .64, and .69, respectively. For diastolic pressure (Korotkoff phase 4), the corresponding values were .28, .41, .47, and .54. These higher multiple-visit estimates of tracking approximate published adult values and raise the possibility that prediction of adult blood pressure from childhood measurements may be improved by averaging readings from multiple weekly visits. [ABSTRACT FROM AUTHOR]

Aircraft accidents with survivors were examined to determine the relative risk at mortality for unrestrained infant passengers w, seat-belted adult passengers. The crude relative risk was estimated to he 7.1, based on US data. and 7.4. based on worldwide data. More refined estimates allow rot possible effects of seat location and for differences in lethality between crashes. Using such estimates, unbelted infant passengers have relative risk of 5.9 based on US data and 9.6 based on worldwide data. The injury experience of restrained vs unrestrained automobile passengers suggests that the observed excess risk to infant air passengers may be related to the absence of a mechanical restraint system. Recommendations for a mechanical restraint system are made and the cost and benefits of implementing these recommendations in the US are discussed. (Am J Public Health 1981:71: 1242-1246.) [ABSTRACT FROM AUTHOR]

Abstract. The blood pressures and pulse rates of 257 normal full-term infants and their mothers were measured two to four days after birth. Birthweight was correlated with systolic (P = .038), but not with diastolic blood pressure. Infants who were asleep had significantly lower systolic and diastolic blood pressure than infants who were awake (P < .001). Sex, body length, and feedings did not appear to influence infant's blood pressure nor did the anesthesia given to the mothers. Maternal diastolic pressure correlated with infant's diastolic pressure (regression coefficient, .128) (P < .01), whereas for systolic pressure the regression coefficient between maternal and infant pressure was .085 (P = NS). The aggregation between the diastolic blood pressures of infants and mothers was not influenced by birthweight, age of the mother, or medication administered to the mother. The pulse rates of black infants were significantly higher than those of white infants (P < .002). There was no correlation between the pulse rates and blood pressures in infants. [ABSTRACT FROM AUTHOR]

Assesses the value of chest X-ray in the diagnosis of bronchial asthma. Evaluation of the medical records of 58 patients with mild to moderate bronchial asthma; Performance of simple spirometry and chest X-ray; Chest X-ray more sensitive than spirometry for bronchial asthma; Importance of the inflammatory response in asthma; Value of chest X-ray in diagnosing mild bronchial asthma.

Clustered data occur frequently in statistical practice. In some areas of application, such as ophthalmology, clustered data are the rule rather than the exception. In this setting, standard multivariate methods such as logistic regression are invalid, because of the lack of independence among outcomes for individual sample points within a cluster. In Rosner (1984), a polychotomous logistic regression model was presented to control for the effect of both cluster and individual-specific covariates while accounting for the correlation among units within a cluster. This model reduces to a beta-binomial model in the absence of covariates, and to an ordinary logistic model for clusters of size 1 and for larger clusters when no correlation is present. A 1 level of nesting correlation structure is assumed, as is typically found in ophthalmology, where the person is the cluster and the individual eyes are the units within a cluster. In this article, this model is generalized to a 2 level of nesting structure. This model is motivated by an ophthalmologic application, where data are collected in family units and one wishes to model simultaneously the correlation at the first level of nesting among persons within a family, and at the second level of nesting between eyes within a person. In the absence of covariates, the model is characterized by a compound beta-binomial distribution that generalizes the betabinomial distribution to more than one level of nesting. The compound beta-binomial distribution is then augmented in a regression setting, to allow for the presence of family-, person-, and eye-specific covariates while controlling for correlation at each level of nesting. Extensions to more than two levels of nesting are also considered. [ABSTRACT FROM AUTHOR]