Lifetime alcohol consumption patterns and young-onset breast cancer by subtype among Non-Hispanic Black and White women in the Young Women's Health History Study

PURPOSE

The role of alcohol in young-onset breast cancer (YOBC) is unclear. We examined associations between lifetime alcohol consumption and YOBC in the Young Women's Health History Study, a population-based case-control study of breast cancer among Non-Hispanic Black and White women < 50 years of age.

METHODS

Breast cancer cases (n = 1,812) were diagnosed in the Metropolitan Detroit and Los Angeles County SEER registry areas, 2010-2015. Controls (n = 1,381) were identified through area-based sampling and were frequency-matched to cases by age, site, and race. Alcohol consumption and covariates were collected from in-person interviews. Weighted multivariable logistic regression was conducted to calculate adjusted odds ratios (aOR) and 95% confidence intervals (CI) for associations between alcohol consumption and YOBC overall and by subtype (Luminal A, Luminal B, HER2, or triple negative).

RESULTS

Lifetime alcohol consumption was not associated with YOBC overall or with subtypes (all ptrend ≥ 0.13). Similarly, alcohol consumption in adolescence, young and middle adulthood was not associated with YOBC (all ptrend ≥ 0.09). An inverse association with triple-negative YOBC, however, was observed for younger age at alcohol use initiation (< 18 years vs. no consumption), aOR (95% CI) = 0.62 (0.42, 0.93). No evidence of statistical interaction by race or household poverty was observed.

CONCLUSIONS

Our findings suggest alcohol consumption has a different association with YOBC than postmenopausal breast cancer-lifetime consumption was not linked to increased risk and younger age at alcohol use initiation was associated with a decreased risk of triple-negative YOBC. Future studies on alcohol consumption in YOBC subtypes are warranted.

Calibration of an Adaptive Genetic Algorithm for Modeling Opinion Diffusion

Genetic algorithms mimic the process of natural selection in order to solve optimization problems with minimal assumptions and perform well when the objective function has local optima on the search space. These algorithms treat potential solutions to the optimization problem as chromosomes, consisting of genes which undergo biologically-inspired operators to identify a better solution. Hyperparameters or control parameters determine the way these operators are implemented. We created a genetic algorithm in order to fit a DeGroot opinion diffusion model using limited data, making use of selection, blending, crossover, mutation, and survival operators. We adapted the algorithm from a genetic algorithm for design of mixture experiments, but the new algorithm required substantial changes due to model assumptions and the large parameter space relative to the design space. In addition to introducing new hyperparameters, these changes mean the hyperparameter values suggested for the original algorithm cannot be expected to result in optimal performance. To make the algorithm for modeling opinion diffusion more accessible to researchers, we conduct a simulation study investigating hyperparameter values. We find the algorithm is robust to the values selected for most hyperparameters and provide suggestions for initial, if not default, values and recommendations for adjustments based on algorithm output.

Using social contact data to improve the overall effect estimate of a cluster-randomized influenza vaccination program in Senegal

This study estimates the overall effect of two influenza vaccination programs consecutively administered in a cluster-randomized trial in western Senegal over the course of two influenza seasons from 2009-2011. We apply cutting-edge methodology combining social contact data with infection data to reduce bias in estimation arising from contamination between clusters. Our time-varying estimates reveal a reduction in seasonal influenza from the intervention and a nonsignificant increase in H1N1 pandemic influenza. We estimate an additive change in overall cumulative incidence (which was 6.13% in the control arm) of -0.68 percentage points during Year 1 of the study (95% CI: -2.53, 1.18). When H1N1 pandemic infections were excluded from analysis, the estimated change was -1.45 percentage points and was significant (95% CI, -2.81, -0.08). Because cross-cluster contamination was low (0-3% of contacts for most villages), an estimator assuming no contamination was only slightly attenuated (-0.65 percentage points). These findings are encouraging for studies carefully designed to minimize spillover. Further work is needed to estimate contamination - and its effect on estimation - in a variety of settings.

Performance of a Genetic Algorithm for Estimating DeGroot Opinion Diffusion Model Parameters for Health Behavior Interventions

Leveraging social influence is an increasingly common strategy to change population behavior or acceptance of public health policies and interventions; however, assessing the effectiveness of these social network interventions and projecting their performance at scale requires modeling of the opinion diffusion process. We previously developed a genetic algorithm to fit the DeGroot opinion diffusion model in settings with small social networks and limited follow-up of opinion change. Here, we present an assessment of the algorithm performance under the less-than-ideal conditions likely to arise in practical applications. We perform a simulation study to assess the performance of the algorithm in the presence of ordinal (rather than continuous) opinion measurements, network sampling, and model misspecification. We found that the method handles alternate models well, performance depends on the precision of the ordinal scale, and sampling the full network is not necessary to use this method. We also apply insights from the simulation study to investigate notable features of opinion diffusion models for a social network intervention to increase uptake of pre-exposure prophylaxis (PrEP) among Black men who have sex with men (BMSM).

Using a novel genetic algorithm to assess peer influence on willingness to use pre-exposure prophylaxis in networks of Black men who have sex with men

The DeGroot model for opinion diffusion over social networks dates back to the 1970s and models the mechanism by which information or disinformation spreads through a network, changing the opinions of the agents. Extensive research exists about the behavior of the DeGroot model and its variations over theoretical social networks; however, research on how to estimate parameters of this model using data collected from an observed network diffusion process is much more limited. Existing algorithms require large data sets that are often infeasible to obtain in public health or social science applications. In order to expand the use of opinion diffusion models to these and other applications, we developed a novel genetic algorithm capable of recovering the parameters of a DeGroot opinion diffusion process using small data sets, including those with missing data and more model parameters than observed time steps. We demonstrate the efficacy of the algorithm on simulated data and data from a social network intervention leveraging peer influence to increase willingness to take pre-exposure prophylaxis in an effort to decrease transmission of human immunodeficiency virus among Black men who have sex with men.

Abstract C030: Alcohol consumption and risk of young-onset breast cancer among non-Hispanic Black and White women in the Young Women’s Health History Study

The incidence of young-onset breast cancer (YOBC) is increasing, and is higher among Non-Hispanic Black (NHB) than Non-Hispanic White (NHW) women. While alcohol consumption is consistently linked to breast cancer risk, its role in YOBC specifically, is poorly understood. Moreover, it is unclear whether associations of alcohol and YOBC vary by tumor subtype or whether it contributes to disparities by race and/or socioeconomic position (SEP). We examined associations of lifetime alcohol use and YOBC overall and by tumor subtype, and assessed whether associations varied by race, and SEP. We hypothesized that alcohol consumption would be positively associated with YOBC risk, with stronger associations for estrogen receptor (ER)-positive subtypes and similar associations across race and SEP categories. Data are from the Young Women’s Health History study, a large population-based case-control study of breast cancer among NHB and NHW women &lt; 50 years of age, from the Los Angeles and Detroit SEER registries. Controls (n=1,381) were identified through area-based sampling and frequency matched to invasive breast cancer cases (n=1,812) by age-group, study site and race. Lifetime average alcohol consumption was collected from in-person interviews, using life history calendars to facilitate accurate recall. Molecular subtypes were defined as: Luminal A (ER-positive and/or progesterone receptor (PR)-positive and human epidermal growth factor 2 (HER2)-negative with grades 1 or 2); luminal B (ER-positive and/or PR-positive and HER2-positive or ER-positive and/or PR-positive and HER2-negative with grade 3); HER2-type (ER-negative and/or PR-negative and HER2-positive); and triple negative (TN; negative for ER, PR, and HER2). Weighted multivariable logistic regression models, adjusted for established risk factors, were used to estimate odds ratios (OR) and 95% confidence intervals (CI) for associations of alcohol consumption and YOBC by subtype. Statistical interaction of alcohol by race and SEP was evaluated using the Wald’s test for the cross-product term. Women who consumed &gt;20 g/day of alcohol were slightly leaner, less likely to have a college education and have a family history of breast cancer and more likely to have smoked than nondrinkers. Alcohol consumption was not significantly associated with YOBC risk overall, or within molecular subtype [LumA (&gt;20 g/day vs. nondrinkers OR (95%CI) = 0.66 (0.43, 1.00), LumB=0.83 (0.53, 1.31), HER2=1.43 (0.63, 3.23), and TN=0.82 (0.52, 1.29)]. We did not observe any interaction by race (pint=0.62) or SEP (pint=0.94) overall or in models of tumor subtypes (all pint&gt;0.28). Alcohol consumption of &gt;20 g/day was inversely associated with LumA among NHW [OR (95%CI) =0.50 (0.28, 0.89)], however, no significant association was observed for LumA in NHB [OR (95%CI) =1.14 (0.61, 2.12)]. Our findings suggest that alcohol consumption is not strongly linked to YOBC risk, and that associations do not appear to vary by molecular subtype, race, or SEP.

Citation Format: Kelly A Hirko, Darek R Lucas, Dorothy R Pathak, Nicole Bohme Carnegie, Richard T Houang, Kendra Schwartz, Ellen M Velie. Alcohol consumption and risk of young-onset breast cancer among non-Hispanic Black and White women in the Young Women’s Health History Study [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr C030.

Measuring the predictability of life outcomes with a scientific mass collaboration

How predictable are life trajectories? We investigated this question with a scientific mass collaboration using the common task method; 160 teams built predictive models for six life outcomes using data from the Fragile Families and Child Wellbeing Study, a high-quality birth cohort study. Despite using a rich dataset and applying machine-learning methods optimized for prediction, the best predictions were not very accurate and were only slightly better than those from a simple benchmark model. Within each outcome, prediction error was strongly associated with the family being predicted and weakly associated with the technique used to generate the prediction. Overall, these results suggest practical limits to the predictability of life outcomes in some settings and illustrate the value of mass collaborations in the social sciences.

Effects of contact network structure on epidemic transmission trees: implications for data required to estimate network structure

Understanding the dynamics of disease spread is key to developing effective interventions to control or prevent an epidemic. The structure of the network of contacts over which the disease spreads has been shown to have a strong influence on the outcome of the epidemic, but an open question remains as to whether it is possible to estimate contact network features from data collected in an epidemic. The approach taken in this paper is to examine the distributions of epidemic outcomes arising from epidemics on networks with particular structural features to assess whether that structure could be measured from epidemic data and what other constraints might be needed to make the problem identifiable. To this end, we vary the network size, mean degree, and transmissibility of the pathogen, as well as the network feature of interest: clustering, degree assortativity, or attribute-based preferential mixing. We record several standard measures of the size and spread of the epidemic, as well as measures that describe the shape of the transmission tree in order to ascertain whether there are detectable signals in the final data from the outbreak. The results suggest that there is potential to estimate contact network features from transmission trees or pure epidemic data, particularly for diseases with high transmissibility or for which the relevant contact network is of low mean degree. Copyright © 2017 John Wiley & Sons, Ltd.

A flexible, interpretable framework for assessing sensitivity to unmeasured confounding

When estimating causal effects, unmeasured confounding and model misspecification are both potential sources of bias. We propose a method to simultaneously address both issues in the form of a semi-parametric sensitivity analysis. In particular, our approach incorporates Bayesian Additive Regression Trees into a two-parameter sensitivity analysis strategy that assesses sensitivity of posterior distributions of treatment effects to choices of sensitivity parameters. This results in an easily interpretable framework for testing for the impact of an unmeasured confounder that also limits the number of modeling assumptions. We evaluate our approach in a large-scale simulation setting and with high blood pressure data taken from the Third National Health and Nutrition Examination Survey. The model is implemented as open-source software, integrated into the treatSens package for the R statistical programming language. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Estimation of the Overall Treatment Effect in the Presence of Interference in Cluster-Randomized Trials of Infectious Disease Prevention

An issue that remains challenging in the field of causal inference is how to relax the assumption of no interference between units. Interference occurs when the treatment of one unit can affect the outcome of another, a situation which is likely to arise with outcomes that may depend on social interactions, such as occurrence of infectious disease. Existing methods to accommodate interference largely depend upon an assumption of “partial interference” – interference only within identifiable groups but not among them. There remains a considerable need for development of methods that allow further relaxation of the no-interference assumption. This paper focuses on an estimand that is the difference in the outcome that one would observe if the treatment were provided to all clusters compared to that outcome if treatment were provided to none – referred as the overall treatment effect. In trials of infectious disease prevention, the randomized treatment effect estimate will be attenuated relative to this overall treatment effect if a fraction of the exposures in the treatment clusters come from individuals who are outside these clusters. This source of interference – contacts sufficient for transmission that are with treated clusters – is potentially measurable. In this manuscript, we leverage epidemic models to infer the way in which a given level of interference affects the incidence of infection in clusters. This leads naturally to an estimator of the overall treatment effect that is easily implemented using existing software.

An approximation method for improving dynamic network model fitting

There has been a great deal of interest recently in the modeling and simulation of dynamic networks, i.e., networks that change over time. One promising model is the separable temporal exponential-family random graph model (ERGM) of Krivitsky and Handcock, which treats the formation and dissolution of ties in parallel at each time step as independent ERGMs. However, the computational cost of fitting these models can be substantial, particularly for large, sparse networks. Fitting cross-sectional models for observations of a network at a single point in time, while still a non-negligible computational burden, is much easier. This paper examines model fitting when the available data consist of independent measures of cross-sectional network structure and the duration of relationships under the assumption of stationarity. We introduce a simple approximation to the dynamic parameters for sparse networks with relationships of moderate or long duration and show that the approximation method works best in precisely those cases where parameter estimation is most likely to fail-networks with very little change at each time step. We consider a variety of cases: Bernoulli formation and dissolution of ties, independent-tie formation and Bernoulli dissolution, independent-tie formation and dissolution, and dependent-tie formation models.

Targeting pre-exposure prophylaxis among men who have sex with men in the United States and Peru: partnership types, contact rates, and sexual role

BACKGROUND

We aim to identify optimal strategies for deploying pre-exposure prophylaxis among men who have sex with men (MSM) in the United States and Peru to maximize population-level effectiveness in an efficient manner. We use epidemic models to simulate the impact of targeting strategies. Most studies have focused on targeting either the general population or high-risk MSM. Alternative strategies, including serodiscordant couples, may better balance effectiveness and efficiency.

METHODS

We use dynamic stochastic sexual network models based on exponential-family random graph modeling, parameterized from behavioral surveys of MSM in the United States and Peru. These models represent main partnerships and casual contacts separately, permitting modeling of interventions targeting men whose risk derives from combinations of relational types. We also model varying rates of uptake and adherence to pre-exposure prophylaxis (PrEP). We assess sensitivity of results to risk compensation through increases in condomless casual contacts and condomless sex in main partnerships.

RESULTS

Targeting all men who are not exclusively insertive has the largest impact on HIV incidence, but targeting only those with high levels of casual activity yields comparable results using fewer person-years on PrEP. The effect is robust to risk compensation in the United States, but less so in Peru. Targeting serodiscordant main partnerships does not significantly impact incidence, but requires fewer person-years on PrEP per infection averted than other strategies.

CONCLUSIONS

PrEP could be effective in reducing new infections at the population level in both settings. Serodiscordant partnerships are an attractive component of a targeting program, but targeting should include other high-risk men.

Linkage of viral sequences among HIV-infected village residents in Botswana: estimation of linkage rates in the presence of missing data

Linkage analysis is useful in investigating disease transmission dynamics and the effect of interventions on them, but estimates of probabilities of linkage between infected people from observed data can be biased downward when missingness is informative. We investigate variation in the rates at which subjects' viral genotypes link across groups defined by viral load (low/high) and antiretroviral treatment (ART) status using blood samples from household surveys in the Northeast sector of Mochudi, Botswana. The probability of obtaining a sequence from a sample varies with viral load; samples with low viral load are harder to amplify. Pairwise genetic distances were estimated from aligned nucleotide sequences of HIV-1C env gp120. It is first shown that the probability that randomly selected sequences are linked can be estimated consistently from observed data. This is then used to develop estimates of the probability that a sequence from one group links to at least one sequence from another group under the assumption of independence across pairs. Furthermore, a resampling approach is developed that accounts for the presence of correlation across pairs, with diagnostics for assessing the reliability of the method. Sequences were obtained for 65% of subjects with high viral load (HVL, n = 117), 54% of subjects with low viral load but not on ART (LVL, n = 180), and 45% of subjects on ART (ART, n = 126). The probability of linkage between two individuals is highest if both have HVL, and lowest if one has LVL and the other has LVL or is on ART. Linkage across groups is high for HVL and lower for LVL and ART. Adjustment for missing data increases the group-wise linkage rates by 40–100%, and changes the relative rates between groups. Bias in inferences regarding HIV viral linkage that arise from differential ability to genotype samples can be reduced by appropriate methods for accommodating missing data.

The analysis of viral genomes has great potential for investigating transmission of disease, including the identification of risk factors and transmission clusters, and can thereby aid in targeting interventions. To make use of genetic data in this way, it is necessary to make inferences about population-level patterns of viral linkage. As with any rigorous statistical inference from sampled data to a population, it is important to consider the effect of the sampling strategy and the occurrence of missing data on the final inferences made. In this paper we highlight the effects of missing data on the resulting estimates of population level linkage rates and develop methods for adjusting for the presence of missing data. As an example, we consider comparing the rates of linkage of HIV sequences from subjects with high viral load, low viral load, or on antiretroviral treatment, and show that comparative inferences are compromised when adjustment is not made for missing sequences and bias in inferences can be reduced with proper adjustment.

Size matters: concurrency and the epidemic potential of HIV in small networks

Background

Generalized heterosexual epidemics are responsible for the largest share of the global burden of HIV. These occur in populations that do not have high rates of partner acquisition, and research suggests that a pattern of fewer, but concurrent, partnerships may be the mechanism that provides the connectivity necessary for sustained transmission. We examine how network size affects the impact of concurrency on network connectivity.

Methodology/Principal Findings

We use a stochastic network model to generate a sample of networks, varying the size of the network and the level of concurrency, and compare the largest components for each scenario to the asymptotic expected values. While the threshold for the growth of a giant component does not change, the transition is more gradual in the smaller networks. As a result, low levels of concurrency generate more connectivity in small networks.

Conclusions/Significance

Generalized HIV epidemics are by definition those that spread to a larger fraction of the population, but the mechanism may rely in part on the dynamics of transmission in a set of linked small networks. Examples include rural populations in sub-Saharan Africa and segregated minority populations in the US, where the effective size of the sexual network may well be in the hundreds, rather than thousands. Connectivity emerges at lower levels of concurrency in smaller networks, but these networks can still be disconnected with small changes in behavior. Concurrency remains a strategic target for HIV combination prevention programs in this context.

Bootstrap confidence intervals and bias correction in the estimation of HIV incidence from surveillance data with testing for recent infection

The incidence of new infections is a key measure of the status of the HIV epidemic, but accurate measurement of incidence is often constrained by limited data. Karon et al. (Statist. Med. 2008; 27:4617–4633) developed a model to estimate the incidence of HIV infection from surveillance data with biologic testing for recent infection for newly diagnosed cases. This method has been implemented by public health departments across the United States and is behind the new national incidence estimates, which are about 40 per cent higher than previous estimates. We show that the delta method approximation given for the variance of the estimator is incomplete, leading to an inflated variance estimate. This contributes to the generation of overly conservative confidence intervals, potentially obscuring important differences between populations. We demonstrate via simulation that an innovative model-based bootstrap method using the specified model for the infection and surveillance process improves confidence interval coverage and adjusts for the bias in the point estimate. Confidence interval coverage is about 94–97 per cent after correction, compared with 96–99 per cent before. The simulated bias in the estimate of incidence ranges from −6.3 to +14.6 per cent under the original model but is consistently under 1 per cent after correction by the model-based bootstrap. In an application to data from King County, Washington in 2007 we observe correction of 7.2 per cent relative bias in the incidence estimate and a 66 per cent reduction in the width of the 95 per cent confidence interval using this method. We provide open-source software to implement the method that can also be extended for alternate models.