Individual-level home values and cancer mortality in a statewide registry.

BACKGROUND: Prior work assessing disparities in cancer outcomes has relied on regional socioeconomic metrics. These metrics average data across many individuals, resulting in a loss of granularity and confounding with other regional factors. METHODS: Using patients' addresses at the time of diagnosis from the Ohio Cancer Incidence Surveillance System, we retrieved individual home price estimates from an online real estate marketplace. This individual-level estimate was compared with the Area Deprivation Index (ADI) at the census block group level. Multivariable Cox proportional hazards models were used to determine the relationship between home price estimates and all-cause and cancer-specific mortality. RESULTS: A total of 667 277 patients in Ohio Cancer Incidence Surveillance System were linked to individual home prices across 16 cancers. Increasing home prices, adjusted for age, stage at diagnosis, and ADI, were associated with a decrease in the hazard of all-cause and cancer-specific mortality (hazard ratio [HR] = 0.92, 95% confidence interval [CI] = 0.92 to 0.93, and HR = 0.95, 95% CI = 0.94 to 0.95, respectively). Following a cancer diagnosis, individuals with home prices 2 standard deviations above the mean had an estimated 10-year survival probability (7.8%, 95% CI = 7.2% to 8.3%) higher than those with home prices 2 standard deviations below the mean. The association between home price and mortality was substantially more prominent for patients living in less deprived census block groups (Pinteraction < .001) than for those living in more deprived census block groups. CONCLUSION: Higher individual home prices were associated with improved all-cause and cancer-specific mortality, even after accounting for regional measures of deprivation.

Impact of Left-Digit Age Bias in the Treatment of Localized Prostate Cancer.

PURPOSE: Left-digit bias is a phenomenon in which the leftmost digit of a number disproportionately influences decision making. We measured the effect of left-digit age bias on treatment recommendations for localized prostate cancer. MATERIALS AND METHODS: We included men with clinically localized prostate adenocarcinoma in Surveillance, Epidemiology, and End Results from 2004 to 2018 and the National Cancer Database from 2004 to 2016. Primary outcomes were recommendations for radiation therapy and radical prostatectomy. Regression discontinuity was used to assess whether age increase from 69 to 70 years was associated with disproportionate changes in treatment recommendations. RESULTS: In Surveillance, Epidemiology, and End Results, discontinuities were found in the proportion of patients recommended for radiation among the entire cohort (effect size 2.2%, P < .01) and among patients with Gleason 6 (1.6%, P < .01), Gleason 7 (2.5%, P < .01), and Gleason ≥8 (2.1%, P < .01) cancer, while the proportion recommended for prostatectomy decreased in the entire cohort (-1.4%, P < .01) and in patients with Gleason 7 cancer (-2.4%, P < .01). In the National Cancer Database, discontinuity from age 69 to 70 was found in recommendations for radiation in the entire cohort (effect size: 3.1%, P < .01) and in patients with Gleason 6 (2.2%, P < .01), Gleason 7 (4.0%, P < .01), and Gleason ≥8 (2.3%, P < .02) cancer, while the proportion recommended for prostatectomy decreased at this cutoff in the entire cohort (effect size: -2.7%, P < .01) and patients with Gleason 6 (-2.2%, P < .01) and Gleason 7 (-3.7%, P < .01) cancer. CONCLUSIONS: In patients with localized prostate cancer, left-digit age change from 69 to 70 was associated with disproportionately increased recommendations for radiation and decreased recommendations for prostatectomy.

Risk Factors for 30-day Hospital Readmission for Diverticular Hemorrhage.

INTRODUCTION: The 2010 Affordable Care Act introduced the Hospital Readmissions Reduction Program to reduce health care utilization. Diverticular disease and its complications remain a leading cause of hospitalization among gastrointestinal disease. We sought to determine risk factors for 30-day hospital readmissions after hospitalization for diverticular bleeding. MATERIALS AND METHODS: We utilized the 2013 National Readmission Database sponsored by the Agency for Healthcare Research and Quality focusing on hospitalizations with the primary or secondary discharge diagnosis of diverticular hemorrhage or diverticulitis with hemorrhage. We excluded repeat readmissions, index hospitalizations during December and those resulting in death. Our primary outcome was readmission within 30 days of index hospital discharge. Secondary outcomes of interest included medical and procedural comorbid risk factors. The data were analyzed using logistic regression analysis. RESULTS: In total, 29,090 index hospitalizations for diverticular hemorrhage were included. There were 3484 (12%) 30-day readmissions with recurrent diverticular hemorrhage diagnosed in 896 (3%).Index admissions with renal failure [odds ratio (OR), 1.31; 95% confidence interval (CI), 1.19-1.43], congestive heart failure (OR, 1.30; 95% CI, 1.17-1.43), chronic pulmonary disease (OR, 1.19; 95% CI, 1.09-1.29), coronary artery disease (OR, 1.12; 95% CI, 1.03-1.21), atrial fibrillation (OR, 1.12; 95% CI, 1.02-1.22) cirrhosis (OR, 1.95; 95% CI, 1.29-2.93, performance of blood transfusion (OR, 1.23; 95% CI, 1.15-1.33), and abdominal surgery (OR, 1.24; 95% CI, 1.03-1.49) had increased risk of 30-day readmission. CONCLUSIONS: The 30-day readmission rate for diverticular hemorrhage was 12% with multiple identified comorbidities increasing readmission risk.

Vasectomy and Risk of Prostate Cancer in a Screening Trial.

Background: Vasectomy has been implicated as a risk factor for prostate cancer in multiple epidemiologic studies over the past 25 years. Whether this relationship is causal remains unclear. This study examines the association between vasectomy and prostate cancer in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, which randomized men to usual care or annual prostate cancer screening.Methods: We performed a retrospective analysis of 13-year screening and outcomes data from the PLCO trial. Multivariable Cox proportional hazards regression stratified by study arm and age at vasectomy was performed.Results: There was an increased risk of prostate cancer in men who had undergone a vasectomy and were randomized to the usual care arm of the study (adjusted HR, 1.11; 95% confidence interval, 1.03-1.20; P = 0.008). There was no association between vasectomy and diagnosis of prostate cancer in men randomized to the prostate cancer screening arm. Only men undergoing vasectomy at an older age in the usual care arm of the study, but not the prostate cancer screening arm, were at increased risk of being diagnosed with prostate cancer.Conclusions: Vasectomy was not associated with prostate cancer risk among men who were screened for prostate cancer as part of a clinical trial, but was associated with prostate cancer detection in men receiving usual care.Impact: The positive association between vasectomy and prostate cancer is likely related to increased detection of prostate cancer based on patterns of care rather than a biological effect of vasectomy on prostate cancer development. Cancer Epidemiol Biomarkers Prev; 26(11); 1653-9. ©2017 AACR.

Tracking employment shocks using mobile phone data.

Can data from mobile phones be used to observe economic shocks and their consequences at multiple scales? Here we present novel methods to detect mass layoffs, identify individuals affected by them and predict changes in aggregate unemployment rates using call detail records (CDRs) from mobile phones. Using the closure of a large manufacturing plant as a case study, we first describe a structural break model to correctly detect the date of a mass layoff and estimate its size. We then use a Bayesian classification model to identify affected individuals by observing changes in calling behaviour following the plant's closure. For these affected individuals, we observe significant declines in social behaviour and mobility following job loss. Using the features identified at the micro level, we show that the same changes in these calling behaviours, aggregated at the regional level, can improve forecasts of macro unemployment rates. These methods and results highlight promise of new data resources to measure microeconomic behaviour and improve estimates of critical economic indicators.