Collaborative Modeling to Compare Different Breast Cancer Screening Strategies: A Decision Analysis for the US Preventive Services Task Force

Importance

The effects of breast cancer incidence changes and advances in screening and treatment on outcomes of different screening strategies are not well known.

Objective

To estimate outcomes of various mammography screening strategies.

Design, Setting, and Population

Comparison of outcomes using 6 Cancer Intervention and Surveillance Modeling Network (CISNET) models and national data on breast cancer incidence, mammography performance, treatment effects, and other-cause mortality in US women without previous cancer diagnoses.

Exposures

Thirty-six screening strategies with varying start ages (40, 45, 50 years) and stop ages (74, 79 years) with digital mammography or digital breast tomosynthesis (DBT) annually, biennially, or a combination of intervals. Strategies were evaluated for all women and for Black women, assuming 100% screening adherence and "real-world" treatment.

Main Outcomes and Measures

Estimated lifetime benefits (breast cancer deaths averted, percent reduction in breast cancer mortality, life-years gained), harms (false-positive recalls, benign biopsies, overdiagnosis), and number of mammograms per 1000 women.

Results

Biennial screening with DBT starting at age 40, 45, or 50 years until age 74 years averted a median of 8.2, 7.5, or 6.7 breast cancer deaths per 1000 women screened, respectively, vs no screening. Biennial DBT screening at age 40 to 74 years (vs no screening) was associated with a 30.0% breast cancer mortality reduction, 1376 false-positive recalls, and 14 overdiagnosed cases per 1000 women screened. Digital mammography screening benefits were similar to those for DBT but had more false-positive recalls. Annual screening increased benefits but resulted in more false-positive recalls and overdiagnosed cases. Benefit-to-harm ratios of continuing screening until age 79 years were similar or superior to stopping at age 74. In all strategies, women with higher-than-average breast cancer risk, higher breast density, and lower comorbidity level experienced greater screening benefits than other groups. Annual screening of Black women from age 40 to 49 years with biennial screening thereafter reduced breast cancer mortality disparities while maintaining similar benefit-to-harm trade-offs as for all women.

Conclusions

This modeling analysis suggests that biennial mammography screening starting at age 40 years reduces breast cancer mortality and increases life-years gained per mammogram. More intensive screening for women with greater risk of breast cancer diagnosis or death can maintain similar benefit-to-harm trade-offs and reduce mortality disparities.

Forgoing physician visits due to cost: regional clustering among cancer survivors by age, sex, and race/ethnicity

BACKGROUND

Innovative treatments have improved cancer survival but also increased financial hardship for patients. While demographic factors associated with financial hardship among cancer survivors are known in the USA, the role of geography is less clear.

METHODS

We evaluated prevalence of forgoing care due to cost within 12 months by US Census region (Northeast, North Central/Midwest [NCMW], South, West) by demographic factors (age, sex, race/ethnicity) among 217,981 cancer survivors aged 18 to 82 years from the 2015-2019 Behavioral Risk Factor Surveillance System survey. We summarized region- and group-specific prevalence of forgoing physician visits due to cost and used multilevel logistic regression models to compare regions.

RESULTS

The prevalence of forgoing physician visits due to cost was highest in the South (aged < 65 years: 19-38%; aged ≥ 65: 4-21%; adjusted odds ratios [OR], NCMW versus South, OR: 0.63 [0.56-0.71]; Northeast versus South, OR: 0.63 [0.55-0.73]; West versus South, OR: 0.73 [0.64-0.84]). Across the USA, including regions with broad Medicaid expansion, younger, female, and persons of color most often reported cost-related forgoing physician visits.

CONCLUSION

Forgoing physician visits due to cost among cancer survivors is regionally clustered, raising concerns for concentrated poor long-term cancer outcomes. Underlying factors likely include variation in regional population compositions and contextual factors, such as Medicaid expansion and social policies. Disproportionate cost burden among survivors of color in all regions highlight systemic barriers, underscoring the need to improve access to the entire spectrum of care for cancer survivors, and especially for those most vulnerable.

Change in alcohol consumption during the Covid-19 pandemic and associations with mental health and financial hardship: results from a survey of Wisconsin patients with cancer

PURPOSE

Alcohol consumption increases health risks for patients with cancer. The Covid-19 pandemic may have affected drinking habits for these individuals. We surveyed patients with cancer to examine whether changes in drinking habits were related to mental health or financial effects of the pandemic.

METHODS

From October 2020 to April 2021, adult patients (age 18-80 years at diagnosis) treated for cancer in southcentral Wisconsin were invited to complete a survey. Age-adjusted percentages for history of anxiety or depression, emotional distress, and financial impacts of Covid-19 overall and by change in alcohol consumption (non-drinker, stable, decreased, or increased) were obtained via logistic regression.

RESULTS

In total, 1,875 patients were included in the analysis (median age 64, range 19-87 years), including 9% who increased and 23% who decreased drinking. Compared to stable drinkers (32% of sample), a higher proportion of participants who increased drinking alcohol also reported anxiety or depression (45% vs. 26%), moderate to severe emotional distress (61% vs. 37%) and viewing Covid-19 as a threat to their community (67% vs. 55%). Decreased (vs. stable) drinking was associated with higher prevalence of depression or anxiety diagnosis, emotional distress, and negative financial impacts of the pandemic. Compared to non-drinkers (36% of sample), participants who increased drinking were more likely to report emotional distress (61% vs. 48%).

CONCLUSIONS

Patients with cancer from Wisconsin who changed their alcohol consumption during the Covid-19 pandemic were more likely to report poor mental health including anxiety, depression, and emotional distress than persons whose alcohol consumption was stable.

IMPLICATIONS FOR CANCER SURVIVORS

Clinicians working with cancer survivors should be aware of the link between poor mental health and increased alcohol consumption and be prepared to offer guidance or referrals to counseling, as needed.

Population simulation modeling of disparities in US breast cancer mortality

BACKGROUND

Populations of African American or Black women have persistently higher breast cancer mortality than the overall US population, despite having slightly lower age-adjusted incidence.

METHODS

Three Cancer Intervention and Surveillance Modeling Network simulation teams modeled cancer mortality disparities between Black female populations and the overall US population. Model inputs used racial group-specific data from clinical trials, national registries, nationally representative surveys, and observational studies. Analyses began with cancer mortality in the overall population and sequentially replaced parameters for Black populations to quantify the percentage of modeled breast cancer morality disparities attributable to differences in demographics, incidence, access to screening and treatment, and variation in tumor biology and response to therapy.

RESULTS

Results were similar across the 3 models. In 2019, racial differences in incidence and competing mortality accounted for a net ‒1% of mortality disparities, while tumor subtype and stage distributions accounted for a mean of 20% (range across models = 13%-24%), and screening accounted for a mean of 3% (range = 3%-4%) of the modeled mortality disparities. Treatment parameters accounted for the majority of modeled mortality disparities: mean = 17% (range = 16%-19%) for treatment initiation and mean = 61% (range = 57%-63%) for real-world effectiveness.

CONCLUSION

Our model results suggest that changes in policies that target improvements in treatment access could increase breast cancer equity. The findings also highlight that efforts must extend beyond policies targeting equity in treatment initiation to include high-quality treatment completion. This research will facilitate future modeling to test the effects of different specific policy changes on mortality disparities.

Racial Arrest Disparities in the USA by Rural-Urban Location and Region

BACKGROUND

Urban racial arrest disparities are well known. Emerging evidence suggests that rural policing shares similar patterns as urban policing in the USA, but without receiving the same public scrutiny, raising the risk of biased rural policing going unnoticed.

METHODS

We estimated adult and adolescent arrest rates and rate ratios (RR) by race, rural-urban status, and US region based on 2016 Uniform Crime Reporting Program arrest and US Census population counts using general estimating equation Poisson regression models with a 4-way interaction between race, region, age group, and urbanicity.

RESULTS

With few exceptions, arrest rates were highest in small towns and rural areas, especially among Black and American Indian populations. Arrest rates differed between US regions with highest rates and racial disparities in the Midwest. For example, arrest rates among Black adults in the rural Midwest were 148.6 arrests [per 1000 population], 95% CI 131.4-168.0, versus 94.4 arrests, 95% CI 77.2-115.4 in the urban Midwest; and versus corresponding rural Midwest arrests among white adults, 32.7 arrests, 95% CI 30.8-34.8, Black versus white rural RR 4.54, 95% CI 4.09-5.04. Racial arrest disparities in the South were lower but still high, e.g., rural South, Black versus White adults, RR 1.86, 95% CI 1.71-2.03.

CONCLUSIONS

Rural areas and small towns are potential hotspots of racial arrest disparities across the USA, especially in the Midwest. Approaches to overcoming structural racism in policing must include strategies targeted at rural/small town communities. Our findings underscore the importance of dismantling racist policing in all US communities.

US Mass public shootings since Columbine: victims per incident by race and ethnicity of the perpetrator

White individuals in the United States (US) have historically had disproportionate access to firearms. The real-life availability of firearms, including those most lethal, may still be greater among White populations, manifesting in the number of victims in shootings. We compared the severity of US mass public shootings since Columbine by race and/or ethnicity of the perpetrator using The Violence Project Database of Mass Shooters, assessing fatalities (minimum four), total victims, type, and legal status of guns used. We used data visualization and Quasi-Poisson regression of victims minus four - accounting for truncation at 4 fatalities - to assess fatality and total victim rates comparing Non-Hispanic (NH) White with NH Black shooters, using winsorization to account for outlier bias from the 2017 Las Vegas shooting. In 104 total mass public shootings until summer 2021, NH White shooters had higher median fatalities (6 [IQR 5-9] versus 5 [IQR 4-6]) and total victims (9 [IQR 6-19] versus 7 [IQR 5-12]) per incident. Confidence intervals of NH Black versus NH White fatalities rate ratios (RR) ranged from 0.17-1.15, and of total victim RRs from 0.15-1.04. White shooters were overrepresented in mass public shootings with the most victims, typically involving legally owned assault rifles. To better understand the consequences when firearms are readily available, including assault rifles, we need a database of all US gun violence. Our assessment of total victims beyond fatalities emphasizes the large number of US gun violence survivors and the need to understand their experiences to capture the full impact of gun violence.

Mortality risk and physical activity across the lifespan in endometrial cancer survivors

PURPOSE

Physical activity (pre- and post-diagnosis) has been studied in prevention and survivorship contexts for endometrial cancer. However, the association of physical activity (PA) across the lifespan on mortality risk among endometrial cancer survivors is understudied. The study's objective was to identify the association of lifetime PA on mortality risk in endometrial cancer survivors.

METHODS

Seven hundred forty-five endometrial cancer survivors drawn from a population-based cancer registry (diagnosed between 1991 and 1994) reported the frequency (sessions/week) of moderate- and vigorous intensity physical activity (MVPA) at age 12, age 20, and 5 years pre-interview (post-diagnosis). Cox proportional hazards were used to estimate hazard ratios (HR) and 95% confidence intervals for the association between PA, all-cause, and cardiovascular disease mortality as assessed in 2016. MVPA was modeled using natural cubic splines.

RESULTS

Diagnosis age, body mass index, and smoking (pack-years) were each positively associated with increased all-cause mortality risk. Those who did one session of MVPA 5 years pre-interview had a lower mortality risk (HR 0.61; 95% CI 0.41-0.92) compared to those with no MVPA. Those reporting one session of MVPA was similarly observed at age 12 (HR 0.95; 95% CI 0.86-1.06) and at age 20 (HR 0.87; 95% CI 0.65-1.16).

CONCLUSION

Those who participated in PA, compared to those who did not, in the 5 years before diagnosis had a lower mortality risk. While PA was not independently protective against mortality risk at ages 12 or 20, PA is still important for endometrial cancer survivors for other non-mortality outcomes.

Identifying Equitable Screening Mammography Strategies for Black Women in the United States Using Simulation Modeling

BACKGROUND

Screening mammography guidelines do not explicitly consider racial differences in breast cancer epidemiology, treatment, and survival.

OBJECTIVE

To compare tradeoffs of screening strategies in Black women versus White women under current guidelines.

DESIGN

An established model from the Cancer Intervention and Surveillance Modeling Network simulated screening outcomes using race-specific inputs for subtype distribution; breast density; mammography performance; age-, stage-, and subtype-specific treatment effects; and non-breast cancer mortality.

SETTING

United States.

PARTICIPANTS

A 1980 U.S. birth cohort of Black and White women.

INTERVENTION

Screening strategies until age 74 years with varying initiation ages and intervals.

MEASUREMENTS

Outcomes included benefits (life-years gained [LYG], breast cancer deaths averted, and mortality reduction), harms (mammographies, false positives, and overdiagnoses), and benefit-harm ratios (tradeoffs) by race. Efficiency (benefits per unit resource), mortality disparity reduction, and equity in tradeoffs were evaluated. Equitable strategies for Black women were defined as those with tradeoffs closest to benchmark values for screening White women biennially from ages 50 to 74 years.

RESULTS

Biennial screening from ages 45 to 74 years was most efficient for Black women, whereas biennial screening from ages 40 to 74 years was most equitable. Initiating screening 10 years earlier in Black versus White women reduced Black-White mortality disparities by 57% with similar LYG per mammogram for both populations. Selection of the most equitable strategy was sensitive to assumptions about disparities in real-world treatment effectiveness: The less effective treatment was for Black women, the more intensively Black women could be screened before tradeoffs fell short of those experienced by White women.

LIMITATION

Single model.

CONCLUSION

Initiating biennial screening in Black women at age 40 years reduces breast cancer mortality disparities and yields benefit-harm ratios that are similar to tradeoffs of White women screened biennially from ages 50 to 74 years.

PRIMARY FUNDING SOURCE

National Cancer Institute at the National Institutes of Health.

Recent Changes in the Patterns of Breast Cancer as a Proportion of All Deaths According to Race and Ethnicity

BACKGROUND

Recent reports suggest that racial differences in breast cancer incidence rates have decreased. We examined whether these findings apply to breast cancer mortality while considering age, period, and cohort influences on both absolute and relative measures of breast cancer mortality.

METHODS

Using publicly available datasets (CDC WONDER, Human Mortality Database), we developed an age-period-cohort model of breast cancer mortality and breast cancer deaths as a proportion of all deaths during 1968-2019 among all women and by 5 race/ethnicity groups with sufficient numbers for estimation: Hispanic (all races), American Indian/Alaska Native and Asian/Pacific Islanders (regardless of ethnicity), non-Hispanic Black, and non-Hispanic White.

RESULTS

Initially increasing after 1968, age-adjusted breast cancer mortality rates have decreased among all racial/ethnic groups since 1988. The age-adjusted percent of all deaths due to breast cancer also has been declining for non-Hispanic White women since about 1990 while increasing or holding steady for other race/ethnic groups. In 2019, the age-adjusted percent of deaths due to breast cancer for women was highest for Asian/Pacific Islanders (5.6%) followed by non-Hispanic Black (4.5%), Hispanic (4.4%), non-Hispanic White (4.1%), and American Indian/Alaska Native women (2.6%).

CONCLUSIONS

Breast cancer mortality disparities are now greater on both relative and absolute scales for non-Hispanic Black women, and using the relative scale for Asian/Pacific Islander and Hispanic women, compared with non-Hispanic White women for the first time in 50 years.

Weapon Carrying Among Boys in US Schools by Race and/or Ethnicity: 1993-2019

BACKGROUND

High adolescent gun-related mortality, gun violence, pro-gun policies, white supremacy, and the long-term socioeconomic and other effects of racial oppression are intricately linked in the United States. Racist prejudice depicts male individuals of color as more prone to criminality than white male individuals. We described long-term patterns of weapon carrying in US schools among non-Hispanic (NH) white, NH Black/African American, and Hispanic boys, hypothesizing that in contrast to racist stereotypes, boys of color did not bring weapons into schools more often than NH white boys in recent years.

METHODS

We conducted a time series analysis using 1993-2019 Youth Risk Behavior Surveillance System data comparing boys' self-reported weapon carrying in a nationally representative sample of US high schools by race and/or ethnicity, age, and self-reported experience of safety and violence at school.

RESULTS

Weapon carrying in schools has declined among all boys. Comparing all schools, we found no significant differences in weapon carrying (4%-5%) by race and/or ethnicity in 2017 and 2019. Boys who reported experiencing violence or feeling unsafe at school were at least twice as likely to bring a weapon into school, and such negative experiences were more common among boys of color (8%-12%) than among NH white boys (4%-5%). In schools perceived as safer, NH white boys have been more likely to bring weapons into schools than NH Black/African American or Hispanic boys in the past 20 years.

CONCLUSIONS

Our findings contradict racist prejudice with regard to weapon carrying in schools, particularly in more favorable school environments. Making schools safer may reduce weapon carrying in schools where weapon carrying is most common.

Clustered spatio-temporal varying coefficient regression model

In regression analysis for spatio-temporal data, identifying clusters of spatial units over time in a regression coefficient could provide insight into the unique relationship between a response and covariates in certain subdomains of space and time windows relative to the background in other parts of the spatial domain and the time period of interest. In this article, we propose a varying coefficient regression method for spatial data repeatedly sampled over time, with heterogeneity in regression coefficients across both space and over time. In particular, we extend a varying coefficient regression model for spatial-only data to spatio-temporal data with flexible temporal patterns. We consider the detection of a potential cylindrical cluster of regression coefficients based on testing whether the regression coefficient is the same or not over the entire spatial domain for each time point. For multiple clusters, we develop a sequential identification approach. We assess the power and identification of known clusters via a simulation study. Our proposed methodology is illustrated by the analysis of a cancer mortality dataset in the Southeast of the U.S.

Patterns of farm exposure are associated with reduced incidence of atopic dermatitis in early life

BACKGROUND

Farm exposures may reduce the risk of atopic dermatitis (AD) in children, but this is controversial and US data are limited.

OBJECTIVE

This study was conducted to identify patterns of farm exposure in Wisconsin family farms that modify AD incidence and prevalence in early childhood.

METHODS

Environmental exposures, health history, and clinical outcomes were prospectively recorded for 111 farm families and 129 non-farm families enrolled in the Wisconsin Infant Study Cohort birth cohort study. Exposures from the prenatal and early postnatal (2-month) visits were evaluated together with parental report of AD diagnosis by a health care provider through age 24 months. Latent class analysis was performed with prenatal and early postnatal farm-exposure variables to assign farm children to 3 classes.

RESULTS

Overall, children of farm families had reduced AD incidence (P = .03). Within farm families, exposures including poultry (3% vs 28%; P = .003), pig (4% vs 25%; P = .04), feed grain (13% vs 34%; P = .02), and number of animal species were inversely associated with AD incidence. Among the latent class groups, children in families with diverse or more intense farm exposures (classes A and B) had reduced AD incidence, whereas low-exposure (class C) infants had AD incidence similar to that in nonfarm children.

CONCLUSIONS

Infants in Wisconsin farm families had reduced AD incidence, and patterns of farm exposures further defined AD risk. These findings suggest that exposure to diverse farm animals, feed, and bedding during the prenatal period and in early infancy reduce the risk of early-onset AD, a phenotype associated with multiple other atopic diseases.

Bicycling Rates and the Prevalence of Bicycle Helmet Usage in Wisconsin

INTRODUCTION

Bicycles are a source of transportation, recreation, and exercise throughout the world. Bicycling is associated with both health and environmental benefits but also poses a risk of injury. The use of bicycle helmets has been shown to reduce morbidity and mortality associated with cycling. It is unknown if helmet use differs across Wisconsin geographic areas and sociodemographic groups.

METHODS

Data were obtained from the Survey of the Health of Wisconsin (SHOW). Bicycle use and helmet use frequency were determined from a self-administered questionnaire that contained questions specific to preventative health behaviors. Descriptive statistics summarized overall bicycle ridership. Chi-square and Student t tests were performed to assess relationships between bicycle and helmet use across geographic categories and sociodemographic groups.

RESULTS

Differences between sex, race or ethnicity, and education level were found to be associated with bicycle ridership and the frequency of helmet use. Men were significantly more likely to report riding a bicycle and never wearing a helmet. Individuals from urban communities reported always wearing a helmet more often than rural communities. Higher education levels were associated with higher levels of bicycle and helmet use. Race or ethnicity was not associated with bicycle ridership but was associated with differences in helmet use frequency.

CONCLUSION

Nearly half of those who ride bicycles in Wisconsin report never wearing a helmet. Since bicycle ridership and helmet use were found to be associated with a number of sociodemographic characteristics, any solution should consider the role of equity when attempting to increase ridership or helmet use.

Abstract D002: Associations between multilevel health factors and cancer mortality according to rural residence

Surveillance reports consistently observe that cancer mortality rates are higher in rural than urban areas, yet data on the multi-level factors that impact rural disparities have not been fully leveraged to identify the areas of greatest need for research and policy changes. To address gaps in cancer data for rural communities, we adapted the County Health Rankings model of the multiple determinants of health to cancer. Using publicly available data, we compared health factors and cancer mortality for rural versus urban counties in Wisconsin. Counties were defined as rural (N=19) or non-rural (“urban”, N=53) based on Rural Urban Continuum Codes 7-9 and 1-6, respectively. Age-adjusted county-specific cancer mortality rates for all cancer sites combined were obtained from the state cancer registry. Health factor data were obtained from multiple sources in 4 categories: health behaviors (smoking, drinking alcohol, obesity, physical activity); clinical care (HPV vaccination; breast, cervical, and colorectal cancer screening; density of primary care physicians); socioeconomic factors (Area Deprivation Index based on 17 census items); and physical environment (access to grocery stores and alcohol outlets, air quality, pesticide use). Items were ranked for the 72 counties with lower-risk values having better ranks, e.g., higher values for screening and lower values for obesity ranked closer to 1. A composite health factor ranking was defined using County Health Rankings weights, equal to 0.3*(behavioral factors) + 0.2*(clinical factors) + 0.4*(socioeconomic factors) + 0.1*(physical environment). Cancer death rates were higher in rural than in urban counties (181 vs 164 per 100,000). The composite health ranking was positively associated with cancer mortality rates (Pearson correlation coefficient 0.38, 95% CI 0.17-0.57), with worse rankings for rural (average 44, interquartile range, IQR 39-51) than for urban counties (average 34, IQR 25-42). The difference in health factor category rankings between rural and urban counties was greatest for socioeconomic factors (rural average rank 50 vs urban average rank 32) followed by clinical care (rural average rank 43 vs urban average rank 34) and behavioral factors (rural average rank 40 vs urban average rank 35). Physical environment factor rankings were slightly better for rural (average 33) than urban (average 37) counties. In conclusion, we confirmed that cancer mortality in Wisconsin is higher in rural as compared with urban areas. Future analyses will (a) refine the set of health factors used to construct the composite health factor ranking (e.g., account more fully for distance to care) and (b) optimize the weights applied to the categories to calculate the composite ranking. These initial findings suggest that, to increase the impact of future research and policy efforts, clinical and behavioral interventions targeting cancer health disparities in rural counties should include strategies to address socioeconomic factors.

Citation Format: Amy Trentham-Dietz, Noelle K LoConte, Betsy Rolland, Lisa Cadmus-Bertram, Tracy M Downs, John M Eason, Cody M Fredrick, John M Hampton, Xiao Zhang, Ronald E Gangnon. Associations between multilevel health factors and cancer mortality according to rural residence [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 D002.

Effect of Patient Navigation on Transitions of HIV Care After Release from Prison: A Retrospective Cohort Study

Antiretroviral therapy is successfully administered to people living with HIV while they are incarcerated in most US prison systems, but interruptions in treatment are common after people are released. We undertook an observational cohort study designed to examine the clinical and psychosocial factors that influence linkage to HIV care and viral suppression after release from a single state prison system. In this report we describe baseline characteristics and 6-month post-incarceration HIV care outcomes for 170 individuals in Wisconsin. Overall, 114 (67%) individuals were linked to outpatient HIV care within 180 days of release from prison, and of these, 90 (79%) were observed to have HIV viral suppression when evaluated in the community. The strongest predictor of linkage to care in this study was participation in a patient navigation program: Those who received patient navigation were linked to care 84% of the time, compared to 60% of the individuals who received only standard release planning (adjusted OR 3.69, 95% CI 1.24, 10.96; P < 0.01). Findings from this study demonstrate that building and maintaining intensive patient navigation programs that support individuals releasing from prison is beneficial for improving transitions in HIV care.

Contribution of Breast Cancer to Overall Mortality for US Women

OBJECTIVE

Breast cancer simulation models must take changing mortality rates into account to evaluate the potential impact of cancer control interventions. We estimated mortality rates due to breast cancer and all other causes combined to determine their impact on overall mortality by year, age, and birth cohort.

METHODS

Based on mortality rates from publicly available datasets, an age-period-cohort model was used to estimate the proportion of deaths due to breast cancer for US women aged 0 to 119 years, with birth years 1900 to 2000. Breast cancer mortality was calculated as all-cause mortality multiplied by the proportion of deaths due to breast cancer; other-cause mortality was the difference between all-cause and breast cancer mortality.

RESULTS

Breast cancer and other-cause mortality rates were higher for older ages and birth cohorts. The percent of deaths due to breast cancer increased across birth cohorts from 1900 to 1940 then decreased. Among 50-year-old women, in the 1920 birth cohort, 52 (9.9%) of 100,000 deaths (95% CI, 9.8% to 10.1%) were attributed to breast cancer whereas 476 of 100,000 were due to other causes; in the 1960 birth cohort, 22 (8.5%) of 100,000 deaths (95% CI, 8.3% to 8.7%) were attributed to breast cancer with 242 of 100,000 deaths due to other causes. The percentage of all deaths due to breast cancer was highest (4.1% to 12.9%) for women in their 40s and 50s for all birth cohorts.

CONCLUSIONS

This study offers evidence that advances in breast cancer screening and treatment have reduced breast cancer mortality for women across the age spectrum, and provides estimates of age-, year- and birth cohort-specific competing mortality rates for simulation models. Other-cause mortality estimates are important in these models because most women die from causes other than breast cancer.

Respiratory health, allergies, and the farm environment: design, methods and enrollment in the observational Wisconsin Infant Study Cohort (WISC): a research proposal

Epidemiologic and cross-sectional studies suggest that early life farming and animal exposures are associated with major health benefits, influencing immune development and modifying the subsequent risk of allergic diseases, including asthma. The Wisconsin Infant Study Cohort (WISC) study was established in central Wisconsin to test the hypothesis that early life animal farm exposures are associated with distinct innate immune cell maturation trajectories, decreased allergen sensitization and reduced respiratory viral illness burden during the first 2 years of life. Beginning in 2013, a total of 240 families have been enrolled, 16,522 biospecimens have been collected, and 4098 questionnaires have been administered and entered into a secure database. Study endpoints include nasal respiratory virus identification and respiratory illness burden score, allergic sensitization, expression of allergic disease, and anti-viral immune response maturation and profiles. The WISC study prospective design, broad biospecimen collections, and unique US rural community will provide insights into the role of environmental exposures on early life immune maturation profiles associated with protection from allergic sensitization and significant respiratory viral disease burden. The WISC study findings will ultimately inform development of new strategies to promote resistance to severe respiratory viral illnesses and design primary prevention approaches for allergic diseases for all infants.

The impact of chlorhexidine bathing on hospital-acquired bloodstream infections: a systematic review and meta-analysis

Background

Chlorhexidine gluconate (CHG) bathing of hospitalized patients may have benefit in reducing hospital-acquired bloodstream infections (HABSIs). However, the magnitude of effect, implementation fidelity, and patient-centered outcomes are unclear. In this meta-analysis, we examined the effect of CHG bathing on prevention of HABSIs and assessed fidelity to implementation of this behavioral intervention.

Methods

We undertook a meta-analysis by searching Medline, EMBASE, CINAHL, Scopus, and Cochrane’s CENTRAL registry from database inception through January 4, 2019 without language restrictions. We included randomized controlled trials, cluster randomized trials and quasi-experimental studies that evaluated the effect of CHG bathing versus a non-CHG comparator for prevention of HABSIs in any adult healthcare setting. Studies of pediatric patients, of pre-surgical CHG use, or without a non-CHG comparison arm were excluded. Outcomes of this study were HABSIs, patient-centered outcomes, such as patient comfort during the bath, and implementation fidelity assessed through five elements: adherence, exposure or dose, quality of the delivery, participant responsiveness, and program differentiation. Three authors independently extracted data and assessed study quality; a random-effects model was used.

Results

We included 26 studies with 861,546 patient-days and 5259 HABSIs. CHG bathing markedly reduced the risk of HABSIs (IRR = 0.59, 95% confidence interval [CI]: 0.52–0.68). The effect of CHG bathing was consistent within subgroups: randomized (0.67, 95% CI: 0.53–0.85) vs. non-randomized studies (0.54, 95% CI: 0.44–0.65), bundled (0.66, 95% CI: 0.62–0.70) vs. non-bundled interventions (0.51, 95% CI: 0.39–0.68), CHG impregnated wipes (0.63, 95% CI: 0.55–0.73) vs. CHG solution (0.41, 95% CI: 0.26–0.64), and intensive care unit (ICU) (0.58, 95% CI: 0.49–0.68) vs. non-ICU settings (0.56, 95% CI: 0.38–0.83). Only three studies reported all five measures of fidelity, and ten studies did not report any patient-centered outcomes.

Conclusions

Patient bathing with CHG significantly reduced the incidence of HABSIs in both ICU and non-ICU settings. Many studies did not report fidelity to the intervention or patient-centered outcomes. For sustainability and replicability essential for effective implementation, fidelity assessment that goes beyond whether a patient received an intervention or not should be standard practice particularly for complex behavioral interventions such as CHG bathing.

Trial registration

Study registration with PROSPERO CRD42015032523.

Electronic supplementary material

The online version of this article (10.1186/s12879-019-4002-7) contains supplementary material, which is available to authorized users.

Oxidized Low-density Lipoprotein and the Incidence of Age-related Macular Degeneration

PURPOSE

To examine the relationship between serum oxidized low-density lipoprotein (ox-LDL) cholesterol and the incidence of age-related macular degeneration (AMD) over a 25-year period in a sample of persons from the population-based Beaver Dam Eye Study (BDES).

DESIGN

Observational prospective cohort study.

PARTICIPANTS

A total of 4972 people from the BDES (aged 43-84 years and living in Beaver Dam, Wisconsin in 1988) seen during at least 1 of 6 examination phases at approximately 5-year intervals between 1988 and 2016.

METHODS

A 50% random sample of participants (N = 2468) was selected for ox-LDL measurements. Stored frozen specimens from every examination phase were processed using an enzyme-linked immunosorbent assay from a single batch. All available intervals were included for a person, resulting in 6586 person-visits.

MAIN OUTCOME MEASURES

Age-related macular degeneration was assessed using the Wisconsin Age-related Maculopathy Grading System, and severity was defined using a 5-step severity scale. The severity of the worse eye at each examination was used for analyses. A multi-state Markov (MSM) model was fit to simultaneously assess the ox-LDL relationship to all AMD transitions, including incidence of any AMD, incidence of late AMD, and worsening and improvement of AMD over the 25 years of the study.

RESULTS

The mean (standard deviation) level of ox-LDL was 75.3 (23.1) U/L at the baseline examination. When adjusting for age, sex, ARMS2 and CFH risk alleles, and examination phase, the ox-LDL at the beginning of a period was not statistically significantly associated with the incidence of any AMD (hazard ratio per 10 U/L ox-LDL was 1.03, 95% confidence interval 0.98,1.09). Furthermore, ox-LDL was not associated with worsening anywhere along the AMD severity scale, nor with incidence of late AMD. The lack of relationships of ox-LDL to the incidence of any AMD or worsening of AMD remained after adjustment for history of statin use, smoking status, body mass index, and history of cardiovascular disease (data not shown).

CONCLUSIONS

Our findings do not provide evidence for statistically significant relationships between ox-LDL and AMD disease development or worsening of AMD.

Obesity and mortality after locoregional breast cancer diagnosis

PURPOSE

Higher mortality after a breast cancer diagnosis has been observed among women who are obese. We investigated the relationships between body mass index (BMI) and all-cause or breast cancer-specific mortality after a diagnosis of locoregional breast cancer.

METHODS

Women diagnosed in 2004 with AJCC Stage I, II, or III breast cancer (n = 5394) were identified from a population-based National Program of Cancer Registries (NPCR) patterns of care study (POC-BP) drawing from registries in seven U.S. states. Differences in overall and breast cancer-specific mortality were investigated using Cox proportional hazards regression models adjusting for demographic and clinical covariates, including age- and stage-based subgroup analyses.

RESULTS

In women 70 or older, higher BMI was associated with lower overall mortality (HR for a 5 kg/m² difference in BMI = 0.85, 95% CI 0.75-0.95). There was no significant association between BMI and overall mortality for women under 70. BMI was not associated with breast cancer death in the full sample, but among women with Stage I disease; those in the highest BMI category had significantly higher breast cancer mortality (HR for BMI ≥ 35 kg/m² vs. 18.5-24.9 kg/m² = 4.74, 95% CI 1.78-12.59).

CONCLUSIONS

Contrary to our hypothesis, greater BMI was not associated with higher overall mortality. Among older women, BMI was inversely related to overall mortality, with a null association among younger women. Higher BMI was associated with breast cancer mortality among women with Stage I disease, but not among women with more advanced disease.

Optimal Bayesian point estimates and credible intervals for ranking with application to county health indices

It is fairly common to rank different geographic units, e.g. counties in the USA, based on health indices. In a typical application, point estimates of the health indices are obtained for each county, and the indices are then simply ranked as if they were known constants. Several authors have considered optimal rank estimators under squared error loss on the rank scale as a default method for general purpose ranking, e.g. situations where ranking units across the full spectrum of performance (low, medium, high) is important. While computationally convenient, squared error loss on the rank scale may not represent the true inferential goals of rank consumers. We construct alternative loss functions based on three components: (1) the inferential goal (rank position or pairwise comparisons), (2) the scale (original, log-transformed or rank) and (3) the (positional or pairwise) loss function (0/1, squared error or absolute error). We can obtain optimal ranks for loss functions based on rank positions and nearly optimal ranks for loss functions based on pairwise comparisons paired with highest posterior density (HPD) credible intervals. We compare inferences produced by the various ranking methods, both optimal and heuristic, using low birth weight data for counties in the Midwestern United States, from 2006 to 2012.

The Relationship of Retinal Vessel Geometric Characteristics to the Incidence and Progression of Diabetic Retinopathy

PURPOSE

To examine the relationships of retinal vessel geometric characteristics (RVGCs) to the incidence and progression of diabetic retinopathy (DR).

DESIGN

Observational, prospective cohort study.

PARTICIPANTS

Nine hundred ninety-six persons with type 1 diabetes mellitus (T1DM) and 1370 persons with type 2 diabetes mellitus (T2DM) seen at a baseline examination who were eligible for follow-up examinations at subsequent 5-year intervals. A total of 3846 person-interval data from these follow-up examinations are the basis for the analyses.

METHODS

Diabetic retinopathy and macular edema were assessed by grading of 30° stereoscopic color fundus photographs. Retinal vessel geometric characteristics were assessed using the Singapore I Vessel Assessment program from a digitized copy of 1 of the field 1 fundus photographs obtained at baseline and follow-up.

MAIN OUTCOME MEASURES

The 5-year incidence of any DR, progression of DR, and incidence of proliferative diabetic retinopathy (PDR) and clinically significant macular edema (CSME) in right eyes.

RESULTS

Incident DR occurred in 45%, progression in 32%, PDR in 10%, and CSME in 5%. While adjusting for glycated hemoglobin, duration of diabetes, and other factors, retinal arteriolar simple tortuosity was associated significantly with the incidence of any DR (odds ratio [OR], 1.17; 95% confidence interval [CI], 1.01-1.35). Retinal venular branching angle was associated significantly with progression of DR (OR, 1.18; 95% CI, 1.03-1.36), retinal venular curvature tortuosity was associated significantly with the incidence of PDR (OR, 1.15; 95% CI, 1.01-1.30), and retinal venular branching angle (OR, 1.41; 95% CI, 1.10-1.82) was associated significantly with the incidence of CSME. There were no significant associations of other RVGCs with any of the DR outcomes in the full multivariate model. Inclusion of all possible RVGCs did not improve the predictive value of the models that already included retinal vessel diameter and baseline DR severity level.

CONCLUSIONS

Retinal vessel geometric characteristics of the retinal venules were associated with progression of DR; however, most of the RVGCs measured from digitized fundus photographs added little to the assessment of risk of incidence and progression of DR when other risk factors were considered in T1DM and T2DM.

The University of Wisconsin Breast Cancer Epidemiology Simulation Model: An Update

The University of Wisconsin Breast Cancer Epidemiology Simulation Model (UWBCS), also referred to as Model W, is a discrete-event microsimulation model that uses a systems engineering approach to replicate breast cancer epidemiology in the US over time. This population-based model simulates the lifetimes of individual women through 4 main model components: breast cancer natural history, detection, treatment, and mortality. A key feature of the UWBCS is that, in addition to specifying a population distribution in tumor growth rates, the model allows for heterogeneity in tumor behavior, with some tumors having limited malignant potential (i.e., would never become fatal in a woman's lifetime if left untreated) and some tumors being very aggressive based on metastatic spread early in their onset. The model is calibrated to Surveillance, Epidemiology, and End Results (SEER) breast cancer incidence and mortality data from 1975 to 2010, and cross-validated against data from the Wisconsin cancer reporting system. The UWBCS model generates detailed outputs including underlying disease states and observed clinical outcomes by age and calendar year, as well as costs, resource usage, and quality of life associated with screening and treatment. The UWBCS has been recently updated to account for differences in breast cancer detection, treatment, and survival by molecular subtypes (defined by ER/HER2 status), to reflect the recent advances in screening and treatment, and to consider a range of breast cancer risk factors, including breast density, race, body-mass-index, and the use of postmenopausal hormone therapy. Therefore, the model can evaluate novel screening strategies, such as risk-based screening, and can assess breast cancer outcomes by breast cancer molecular subtype. In this article, we describe the most up-to-date version of the UWBCS.

Urinary Magnesium and Other Elements in Relation to Mammographic Breast Density, a Measure of Breast Cancer Risk

PURPOSE

Heavy metals and other elements may act as breast carcinogens due to estrogenic activity. We investigated associations between urine concentrations of a panel of elements and breast density.

METHODS

Mammographic density categories were abstracted from radiology reports of 725 women aged 40-65 yr in the Avon Army of Women. A panel of 27 elements was quantified in urine using high resolution magnetic sector inductively coupled plasma mass spectrometry. We applied LASSO (least absolute shrinkage and selection operator) logistic regression to the 27 elements and calculated odds ratios (OR) and 95% confidence intervals (CI) for dense vs. nondense breasts, adjusting for potential confounders.

RESULTS

Of the 27 elements, only magnesium (Mg) was selected into the optimal regression model. The odds ratio for dense breasts associated with doubling the Mg concentration was 1.24 (95% CI 1.03-1.49). Doubling the calcium-to-magnesium ratio was inversely associated with dense breasts (OR 0.83, 95% CI 0.70-0.98).

CONCLUSIONS

Our cross-sectional study found that higher levels of urinary magnesium were associated with greater breast density. Prospective studies are needed to confirm whether magnesium as evaluated in urine is prospectively associated with breast density and, more importantly, breast cancer.

The association between post-diagnosis health behaviors and long-term quality of life in survivors of ductal carcinoma in situ: a population-based longitudinal cohort study

PURPOSE

Women diagnosed with ductal carcinoma in situ (DCIS) often experience adverse changes in health-related behaviors following diagnosis. The impact of health behaviors on long-term quality of life (QoL) in DCIS survivors has not been investigated.

METHODS

We examined the association of post-diagnosis body mass index (BMI), physical activity, alcohol, and smoking with QoL among 1448 DCIS survivors aged 20-74 enrolled in the population-based Wisconsin in situ Cohort from 1997 to 2006. Health behaviors and QoL were self-reported during biennial post-diagnosis interviews. Physical and mental QoL were measured using the validated SF-36 questionnaire. Generalized linear regression was used to determine the association between behaviors and QoL with adjustment for confounders. Lagged behavior variables were used to predict QoL during follow-up and avoid reverse causation.

RESULTS

Women reported 3,536 QoL observations over an average 7.9 years of follow-up. Women maintaining a healthy BMI had on average a significantly higher summary measure score of physical QoL than obese women (normal versus obese: β = 3.02; 2.18, 3.85). Physical QoL scores were also elevated among those who were physically active (5 + h/week vs. none: β = 1.96; 0.72, 3.20), those consuming at least seven drinks/week of alcohol (vs. none; β = 1.40; 0.39, 2.41), and nonsmokers (vs. current smokers: β = 1.80; 0.89, 2.71). Summary measures of mental QoL were significantly higher among women who were moderately physically active (up to 2 h/week vs. none: β = 1.11; 0.30, 1.92) and nonsmokers (vs. current smokers: β = 1.49;0.45, 2.53).

CONCLUSIONS

Our results demonstrate that maintaining healthy behaviors following DCIS treatment is associated with modest improvements in long-term QoL. These results inform interventions aimed at promoting healthy behaviors and optimizing QoL in DCIS survivors.

Collagen Alignment as a Predictor of Recurrence after Ductal Carcinoma In Situ

Background: Collagen fibers surrounding breast ducts may influence breast cancer progression. Syndecan-1 interacts with constituents in the extracellular matrix, including collagen fibers, and may contribute to cancer cell migration. Thus, the orientation of collagen fibers surrounding ductal carcinoma in situ (DCIS) lesions and stromal syndecan-1 expression may predict recurrence.Methods: We evaluated collagen fiber alignment and syndecan-1 expression in 227 women diagnosed with DCIS in 1995 to 2006 followed through 2014 (median, 14.5 years; range, 0.7-17.6). Stromal collagen alignment was evaluated from diagnostic tissue slides using second harmonic generation microscopy and fiber analysis software. Univariate analysis was conducted using χ2 tests and ANOVA. The association between collagen alignment z-scores, syndecan-1 staining intensity, and time to recurrence was evaluated using HRs and 95% confidence intervals (CIs).Results: Greater fiber angles surrounding DCIS lesions, but not syndecan-1 staining intensity, were related to positive HER2 (P = 0.002) status, comedo necrosis (P = 0.03), and negative estrogen receptor (P = 0.002) and progesterone receptor (P = 0.02) status. Fiber angle distributions surrounding lesions included more angles closer to 90 degrees than normal ducts (P = 0.06). Collagen alignment z-scores for DCIS lesions were positively related to recurrence (HR = 1.25; 95% CI, 0.84-1.87 for an interquartile range increase in average fiber angles).Conclusions: Although collagen alignment and stromal syndecan-1 expression did not predict recurrence, collagen fibers perpendicular to the duct perimeter were more frequent in DCIS lesions with features typical of poor prognosis.Impact: Follow-up studies are warranted to examine whether additional features of the collagen matrix may more strongly predict patient outcomes. Cancer Epidemiol Biomarkers Prev; 27(2); 138-45. ©2017 AACR.

Geographic access to mammography facilities and frequency of mammography screening

PURPOSE

To assess the association between geographic access to mammography facilities and women's mammography utilization frequency.

METHODS

Using data from the population-based 1995-2007 Wisconsin Women's Health study, we used proportional odds and logistic regression to test whether driving times to mammography facilities and the number of mammography facilities within 10 km of women's homes were associated with mammography frequency among women aged 50-74 years and whether associations differed between Rural-Urban Commuting Areas and income and education groups.

RESULTS

We found evidence for nonlinear relationships between geographic access and mammography utilization (nonlinear effects of driving times and facility density, P-values .01 and .005, respectively). Having at least one nearby mammography facility was associated with greater mammography frequency among urban women (1 vs. 0 facilities, odds ratio 1.26, 95% confidence interval, 1.09-1.47), with similar effects among rural women. Adding more facilities had decreasing marginal effects. Long driving times tended to be associated with lower mammography frequency. We found no effect modification by income, education, or urbanicity. In rural settings, mammography nonuse was higher, facility density smaller, and driving times to facilities were longer.

CONCLUSIONS

Having at least one mammography facility near one's home may increase mammography utilization, with decreasing effects per each additional facility.

Skin Intrinsic Fluorescence and Age-Related Macular Degeneration: The Beaver Dam Eye Study

Purpose

To determine if skin intrinsic fluorescence (SIF), a noninvasive measure of advanced glycation endproducts and oxidative stress in skin is associated with AMD.

Methods

SIF was measured with the SCOUT DS skin fluorescence spectrometer in a cross-sectional cohort study of 969 persons aged 68 to 102 years from the 1181 who participated in the 25-year follow-up examination in the Beaver Dam Eye Study (BDES) in 2014 to 2016. The SCOUT DS skin fluorescence spectrometer uses five light-emitting diodes, centered at 375 nm to 456 nm. AMD was assessed by grading of digital color 45° stereoscopic fundus photographs of the macula using the Wisconsin Age-Related Maculopathy grading scheme. Analyses included logistic regression with generalized estimating equations to account for correlation between the eyes of a person.

Results

There were data for 1827 eyes for analyses. Early AMD was present in 22% and late AMD in 4% of the eyes. While adjusting for age, sex, smoking status, and history of cardiovascular disease, there were no significant associations of any SIF measure with any AMD or exudative AMD. SIF01 (odds ratio per 1 SD difference on the log scale, 95% confidence interval) (1.66, 1.00-2.74, P = 0.05) and SIF03 (1.81, 1.16-2.81, P = 0.008) were associated with geographic atrophy.

Conclusions

There was a suggestive relationship of two SIF measures, SIF01 and SIF03, using different correction factors from the excitation centered at 375 nm, with the prevalence of geographic atrophy in the BDES. Longitudinal follow-up is indicated to assess a temporal relationship.

Insurance Coverage for CT Colonography Screening: Impact on Overall Colorectal Cancer Screening Rates

RSNA, 2017.

Health-related behaviors and mortality outcomes in women diagnosed with ductal carcinoma in situ

PURPOSE

Women diagnosed with ductal carcinoma in situ (DCIS) of the breast are at greater risk of dying from cardiovascular disease and other causes than from breast cancer, yet associations between health-related behaviors and mortality outcomes after DCIS have not been well studied.

METHODS

We examined the association of body mass index, physical activity, alcohol consumption, and smoking with mortality among 1925 women with DCIS in the Wisconsin In Situ Cohort study. Behaviors were self-reported through baseline interviews and up to three follow-up questionnaires. Cox proportional hazards regression was used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for mortality after DCIS, with adjustment for patient sociodemographic, comorbidity, and treatment factors.

RESULTS

Over a mean of 6.7 years of follow-up, 196 deaths occurred. All-cause mortality was elevated among women who were current smokers 1 year prior to diagnosis (HR = 2.17 [95% CI 1.48, 3.18] vs. never smokers) and reduced among women with greater physical activity levels prior to diagnosis (HR = 0.55 [95% CI: 0.35, 0.87] for ≥5 h per week vs. no activity). Moderate levels of post-diagnosis physical activity were associated with reduced all-cause mortality (HR = 0.31 [95% CI 0.14, 0.68] for 2-5 h per week vs. no activity). Cancer-specific mortality was elevated among smokers and cardiovascular disease mortality decreased with increasing physical activity levels.

CONCLUSIONS

There are numerous associations between health-related behaviors and mortality outcomes after a DCIS diagnosis.

IMPLICATIONS FOR CANCER SURVIVORS

Women diagnosed with DCIS should be aware that their health-related behaviors are associated with mortality outcomes.

Predictors of primary care provider adoption of CT colonography for colorectal cancer screening

PURPOSE

To examine factors influencing primary care provider (PCP) adoption of CT colonography (CTC) for colorectal cancer (CRC) screening.

MATERIALS AND METHODS

We performed a retrospective cohort study linking electronic health record (EHR) data with PCP survey data. Patients were eligible for inclusion if they were not up-to-date with CRC screening and if they had CTC insurance coverage in the year prior to survey administration. PCPs were included if they had at least one eligible patient in their panel and completed the survey (final sample N = 95 PCPs; N = 6245 patients). Survey data included perceptions of CRC screening by any method, as well as CTC specifically. Multivariate logistic regression estimated odds ratios and 95% confidence intervals for PCP and clinic predictors of CRC screening by any method and screening with CTC.

RESULTS

Substantial variation in CTC use was seen among PCPs and clinics (range 0-16% of CRC screening). Predictors of higher CTC use were PCP perceptions that CTC is effective in reducing CRC mortality, higher number of perceived advantages to screening with CTC, and Internal Medicine specialty. Factors not associated with CTC use were PCP perceptions of less organizational capacity to meet demand for colonoscopy, number of perceived disadvantages to screening with CTC, PCP age and gender, and clinic factors.

CONCLUSION

Significant variation in PCP adoption of CTC exists. PCP perceptions of CTC and specialty practice were related to CTC adoption. Strategies to increase PCP adoption of CTC for CRC screening should include emphasis on the effectiveness and advantages of CTC.

Cluster detection of spatial regression coefficients

Popular approaches to spatial cluster detection, such as the spatial scan statistic, are defined in terms of the responses. Here, we consider a varying-coefficient regression and spatial clusters in the regression coefficients. For varying-coefficient regression, such as the geographically weighted regression, different regression coefficients are obtained for different spatial units. It is often of interest to the practitioners to identify clusters of spatial units with distinct patterns in a regression coefficient, but there is no formal statistical methodology for that. Rather, cluster identification is often ad-hoc such as by eyeballing the map of fitted regression coefficients and discerning patterns. In this paper, we develop new methodology for spatial cluster detection in the regression setting based on hypotheses testing. We evaluate our methods in terms of power and coverages for true clusters via simulation studies. For illustration, our methodology is applied to a cancer mortality dataset. Copyright © 2016 John Wiley & Sons, Ltd.

Tailoring Breast Cancer Screening Intervals by Breast Density and Risk for Women Aged 50 Years or Older: Collaborative Modeling of Screening Outcomes

BACKGROUND

Biennial screening is generally recommended for average-risk women aged 50 to 74 years, but tailored screening may provide greater benefits.

OBJECTIVE

To estimate outcomes for various screening intervals after age 50 years based on breast density and risk for breast cancer.

DESIGN

Collaborative simulation modeling using national incidence, breast density, and screening performance data.

SETTING

United States.

PATIENTS

Women aged 50 years or older with various combinations of breast density and relative risk (RR) of 1.0, 1.3, 2.0, or 4.0.

INTERVENTION

Annual, biennial, or triennial digital mammography screening from ages 50 to 74 years (vs. no screening) and ages 65 to 74 years (vs. biennial digital mammography from ages 50 to 64 years).

MEASUREMENTS

Lifetime breast cancer deaths, life expectancy and quality-adjusted life-years (QALYs), false-positive mammograms, benign biopsy results, overdiagnosis, cost-effectiveness, and ratio of false-positive results to breast cancer deaths averted.

RESULTS

Screening benefits and overdiagnosis increase with breast density and RR. False-positive mammograms and benign results on biopsy decrease with increasing risk. Among women with fatty breasts or scattered fibroglandular density and an RR of 1.0 or 1.3, breast cancer deaths averted were similar for triennial versus biennial screening for both age groups (50 to 74 years, median of 3.4 to 5.1 vs. 4.1 to 6.5 deaths averted; 65 to 74 years, median of 1.5 to 2.1 vs. 1.8 to 2.6 deaths averted). Breast cancer deaths averted increased with annual versus biennial screening for women aged 50 to 74 years at all levels of breast density and an RR of 4.0, and those aged 65 to 74 years with heterogeneously or extremely dense breasts and an RR of 4.0. However, harms were almost 2-fold higher. Triennial screening for the average-risk subgroup and annual screening for the highest-risk subgroup cost less than $100 000 per QALY gained.

LIMITATION

Models did not consider women younger than 50 years, those with an RR less than 1, or other imaging methods.

CONCLUSION

Average-risk women with low breast density undergoing triennial screening and higher-risk women with high breast density receiving annual screening will maintain a similar or better balance of benefits and harms than average-risk women receiving biennial screening.

PRIMARY FUNDING SOURCE

National Cancer Institute.

Genetic associations with viral respiratory illnesses and asthma control in children

BACKGROUND

Viral respiratory infections can cause acute wheezing illnesses in children and exacerbations of asthma.

OBJECTIVE

We sought to identify variation in genes with known antiviral and pro-inflammatory functions to identify specific associations with more severe viral respiratory illnesses and the risk of virus-induced exacerbations during the peak fall season.

METHODS

The associations between genetic variation at 326 SNPs in 63 candidate genes and 10 phenotypes related to viral respiratory infection and asthma control were examined in 226 children enrolled in the RhinoGen study. Replication of asthma control phenotypes was performed in 2128 children in the Copenhagen Prospective Study on Asthma in Childhood (COPSAC). Significant associations in RhinoGen were further validated using virus-induced wheezing illness and asthma phenotypes in an independent sample of 122 children enrolled in the Childhood Origins of Asthma (COAST) birth cohort study.

RESULTS

A significant excess of P values smaller than 0.05 was observed in the analysis of the 10 RhinoGen phenotypes. Polymorphisms in 12 genes were significantly associated with variation in the four phenotypes showing a significant enrichment of small P values. Six of those genes (STAT4, JAK2, MX1, VDR, DDX58, and EIF2AK2) also showed significant associations with asthma exacerbations in the COPSAC study or with asthma or virus-induced wheezing phenotypes in the COAST study.

CONCLUSIONS

We identified genetic factors contributing to individual differences in childhood viral respiratory illnesses and virus-induced exacerbations of asthma. Defining mechanisms of these associations may provide insight into the pathogenesis of viral respiratory infections and virus-induced exacerbations of asthma.

Patterns of sedentary behavior and physical function in older adults

BACKGROUND/AIMS

The purposes of this study were to examine the relationship between various objectively measured sedentary behavior (SB) variables and physical function in older adults, examine the measurement properties of an SB questionnaire, and describe the domains of SB in our sample.

METHODS

Forty-four older adults (70 ± 8 years, 64 % female) had their SB measured via activPAL activity monitor and SB questionnaire for 1 week followed by performance-based tests of physical function.

RESULTS

The pattern of SB was more important than total SB time. Where a gender by SB interaction was found, increasing time in SB and fewer breaks were associated with worse function in the males only. The SB questionnaire had acceptable test-retest reliability but poor validity compared to activPAL-measured SB. The majority of SB time was spent watching television, using the computer and reading.

DISCUSSION/CONCLUSIONS

This study provides further evidence for the association between SB and physical function and describes where older adults are spending their sedentary time. This information can be used in the design of future intervention to reduce sedentary time and improve function in older adults.

Relationships among aeroallergen sensitization, peripheral blood eosinophils, and periostin in pediatric asthma development

BACKGROUND

Biomarkers, preferably noninvasive, that predict asthma inception in children are lacking.

OBJECTIVE

Little is known about biomarkers of type 2 inflammation in early life in relation to asthma inception. We evaluated aeroallergen sensitization, peripheral blood eosinophils, and serum periostin as potential biomarkers of asthma in children.

METHODS

Children enrolled in the Childhood Origins of ASThma study were followed prospectively from birth. Blood samples were collected at ages 2, 4, 6, and 11 years, and serum-specific IgE levels, blood eosionophil counts, and periostin levels were measured in 244 children. Relationships among these biomarkers, age, and asthma were assessed.

RESULTS

Serum periostin levels were approximately 2- to 3-fold higher in children than previously observed adult levels. Levels were highest at 2 years (145 ng/mL), and did not change significantly between 4 and 11 years (128 and 130 ng/mL). Age 2 year periostin level of 150 ng/mL or more predicted asthma at age 6 years (odds ratio [OR], 2.3; 95% CI, 1.3-4.4). Eosinophil count of 300 cells/μL or more and aeroallergen sensitization at age 2 years were each associated with increased risk of asthma at age 6 years (OR, 3.1; 95% CI, 1.7-6.0 and OR, 3.3; 95% CI, 1.7-6.3). Children with any 2 of the biomarkers had a significantly increased risk of developing asthma by school age (≥2 biomarkers vs none: OR, 6.6; 95% CI, 2.7-16.0).

CONCLUSIONS

Serum periostin levels are significantly higher in children than in adults, likely due to bone turnover, which impairs clinical utility in children. Early life aeroallergen sensitization and elevated blood eosinophils are robust predictors of asthma development. Children with evidence of activation of multiple pathways of type 2 inflammation in early life are at greatest risk for asthma development.

Emerging trends in surgical and adjuvant radiation therapies among women diagnosed with ductal carcinoma in situ

BACKGROUND

The use of surgery and radiation therapy in treating ductal carcinoma in situ (DCIS) is directed by treatment guidelines and evidence from research. This study investigated recent patterns in DCIS treatment by demographic factors.

METHODS

Data for women diagnosed with DCIS between 1998 and 2011 (n = 416,232) in the National Cancer Data Base were assessed for trends in treatment patterns by age group, calendar year, ancestral/ethnic group, and geographic region. The likelihood of receiving specific treatment modalities was analyzed with multivariable logistic regression.

RESULTS

DCIS cases were most frequently treated with breast-conserving surgery (BCS) and adjuvant radiation (45.6%). After an initial rise, the use of adjuvant radiation after BCS plateaued at approximately 70% after 2007, with increasing utilization of mastectomy beyond 2005. In addition, there was an increasing trend in postmastectomy reconstruction over time, and women of African ancestry (odds ratio [OR], 0.69; 95% confidence interval [CI], 0.66-0.72) and Hispanic women (OR, 0.83; 95% CI, 0.78-0.89) were less likely to undergo reconstruction in comparison with women of European ancestry. A similar trend was observed in contralateral risk-reducing mastectomy utilization, with women of European ancestry having a more rapid rise in the utilization of contralateral risk-reducing mastectomy in comparison with all other ancestral/ethnic groups.

CONCLUSIONS

Recent trends demonstrate a plateau in radiation therapy administration after BCS along with increasing utilization of mastectomy, reconstruction, and contralateral risk-reducing mastectomy. There are substantial differences in treatment utilization according to ancestry/ethnicity and geographical region. Further studies examining patient-physician decision making surrounding DCIS treatment are warranted. Cancer 2016. © 2016 American Cancer Society. Cancer 2016;122:2810-2818. © 2016 American Cancer Society.

Early life rhinovirus wheezing, allergic sensitization, and asthma risk at adolescence

Background

Early life rhinovirus (RV) wheezing illnesses and aeroallergen sensitization increase the risk of asthma at school age. Whether these remain risk factors for the persistence of asthma out to adolescence is not established.

Objective

We sought to define the relationships among specific viral illnesses and the type and timing of aeroallergen sensitization with the persistence of asthma into adolescence.

Methods

A total of 217 children were followed prospectively from birth to age 13 years. The etiology and timing of viral wheezing illnesses during the first 3 years of life were assessed along with patterns of allergen sensitization. The associations between viral wheezing illnesses, presence and pattern of aeroallergen sensitization, and asthma diagnosis at age 13 years were evaluated.

Results

When adjusted for all viral etiologies, wheezing with RV (odds ratio = 3.3; 95% CI, 1.5-7.1), but not respiratory syncytial virus (odds ratio = 1.0; 95% CI, 0.4-2.3), was associated with asthma at age 13 years. Age of aeroallergen sensitization also influenced asthma risk; 65% of children sensitized by age 1 year had asthma at age 13 years, compared with 40% of children not sensitized at age 1 year but sensitized by age 5 years, and 17% of children not sensitized at age 5 years. Early life aeroallergen sensitization and RV wheezing had additive effects on asthma risk at adolescence.

Conclusions

In a high-risk birth cohort, the persistence of asthma at age 13 years was most strongly associated with outpatient wheezing illnesses with RV and aeroallergen sensitization in early life.

Stepwise and stagewise approaches for spatial cluster detection

Spatial cluster detection is an important tool in many areas such as sociology, botany and public health. Previous work has mostly taken either a hypothesis testing framework or a Bayesian framework. In this paper, we propose a few approaches under a frequentist variable selection framework for spatial cluster detection. The forward stepwise methods search for multiple clusters by iteratively adding currently most likely cluster while adjusting for the effects of previously identified clusters. The stagewise methods also consist of a series of steps, but with a tiny step size in each iteration. We study the features and performances of our proposed methods using simulations on idealized grids or real geographic areas. From the simulations, we compare the performance of the proposed methods in terms of estimation accuracy and power. These methods are applied to the the well-known New York leukemia data as well as Indiana poverty data.

Variability in Spectral-Domain Optical Coherence Tomography over 4 Weeks by Age

PURPOSE

To quantify variation in spectral-domain optical coherence tomography (SD-OCT) measures of total retinal thickness (top of inner limiting membrane to top of retinal pigment epithelium, RPE) and RPE thickness measures over a 4-week period and by age.

METHODS

A total of 76 volunteers aged 40-85 years were seen at three visits over 4 weeks. Two Topcon SD-OCT scans were taken at each visit. Following grid re-centration, total retinal and RPE thickness were determined in nine subfields. Multilevel modeling was used to quantify variance between scans and by age.

RESULTS

In the central circle, mean total retinal thickness was 237.9 µm (standard deviation, SD, 23.5 µm) and RPE thickness was 46.0 µm (SD 5.3 µm). Intraclass correlation coefficient in the central circle was 0.988 for total retinal thickness and 0.714 for RPE thickness. Pairwise measures taken within 4 weeks were strongly correlated (p > 0.95). Within-subject variation of total retinal thickness increased significantly with age. Subjects in the oldest age group had significantly increased among- and within-subject variability in measures of RPE thickness.

CONCLUSIONS

Correlation between retinal thickness measures was very high (>0.95) over a period of 4 weeks with small changes likely due to variation in measurement. Increasing variability in total retinal and RPE thickness measures with age suggest that the use of more and/or higher quality images to calculate mean thickness to reduce variability may benefit the study of these measures in older persons. This may also impact sample size calculations for future studies of SD-OCT measures in older adults.

Trends in Health-Related Quality of Life After a Diagnosis of Ductal Carcinoma In Situ

PURPOSE

Studies of quality of life (QoL) are scarce among survivors of ductal carcinoma in situ (DCIS). The objective of this study was to assess long-term QoL in DCIS survivors in relation to age at diagnosis, time since diagnosis, and treatments received.

METHODS

We assessed physical and mental measures of health-related QoL in 1,604 patients with DCIS diagnosed in 1997 to 2006 with up to four follow-up interviews. We further compared baseline QoL to 1,055 control patients without DCIS. QoL was measured using the validated Medical Outcomes Study Short Form 36 Health Status Survey questionnaire. Among patients with DCIS, we examined trends in QoL over time since diagnosis using generalized linear regression models, adjusting for confounders. We tested for effect modification by surgical treatment choice, post-treatment endocrine therapy use, and age at diagnosis.

RESULTS

Both physical and mental measures of QoL among DCIS survivors at fewer than 2 years after diagnosis were comparable to controls. Mental measures of QoL among patients with DCIS declined at ≥ 10 years after diagnosis and were significantly lower than at less than 2 years after diagnosis (47.4 v 52.0; P < .01). In the first 5 years after a DCIS diagnosis, mental QoL was significantly higher among women diagnosed at ages 50 to 74 years compared with those diagnosed at ages 28 to 49 years, although this difference was not sustained in later time periods.

CONCLUSION

QoL after a DCIS diagnosis was generally comparable to that of women of similar age without a personal history of DCIS. Our findings suggest that DCIS survivors, and particularly those diagnosed at a younger age, may benefit from support for mental QoL.

The Association Between Post-Diagnosis Health Behaviors and Quality of Life in Survivors of Ductal Carcinoma In Situ

Background: Survivors of ductal carcinoma in situ (DCIS), an early stage breast cancer, tend to decrease physical activity, gain weight, and maintain alcohol use following treatment. However, the impact of these health behaviors on long-term quality of life (QoL) in DCIS survivors has not been investigated. Methods: We examined the association of post-diagnosis body mass index (BMI), physical activity and smoking with QoL among 1,448 DCIS survivors aged 20–74, who were diagnosed during 1995–2006 and enrolled in the population-based Wisconsin In Situ Cohort. Health behaviors and QoL were self-reported during biennial post- diagnosis interviews. Physical and mental QoL were measured using the validated SF-36 questionnaire (higher scores reflect more positive QoL). Generalized linear regression was used to establish QoL mean scores in cross-sectional analyses, with multivariable adjustment for age, comorbidity status, education, and income. Results: Women reported 3,444 QoL observations over an average 7.9 years of follow-up. Physical health summary scale measures of QoL were significantly higher among women with healthy BMI (46.5 for healthy weight versus 40.5 for obese, P = 0.02) and those who were physically active (45.9 for active women versus 42.6 for inactive, P = 0.03). Mental health summary scale scores were significantly higher among non-smokers (51.2 for non-smokers versus 47.1 for current smokers, P &lt; 0.01). These associations were consistent over increasing time since treatment up to 15 years. Conclusion: Our preliminary analysis suggests that maintaining healthy behaviors following DCIS treatment is associated with improved long-term QoL. Longitudinal analysis using cross-lagged regression is underway to evaluate the temporal association between health behavior and QoL. Understanding factors that impact QoL in DCIS survivors may inform interventions aimed at preventing negative health behaviors and optimizing long term quality of life following a DCIS diagnosis.