Dataset Information

Individual- and neighborhood-level predictors of mortality in Florida colorectal cancer patients.

ABSTRACT: We examined individual-level and neighborhood-level predictors of mortality in CRC patients diagnosed in Florida to identify high-risk groups for targeted interventions.Demographic and clinical data from the Florida Cancer Data System registry (2007-2011) were linked with Agency for Health Care Administration and US Census data (n = 47,872). Cox hazard regression models were fitted with candidate predictors of CRC survival and stratified by age group (18-49, 50-64, 65+).Stratified by age group, higher mortality risk per comorbidity was found among youngest (21%), followed by middle (19%), and then oldest (14%) age groups. The two younger age groups had higher mortality risk with proximal compared to those with distal cancer. Compared with private insurance, those in the middle age group were at higher death risk if not insured (HR = 1.35), or received healthcare through Medicare (HR = 1.44), Medicaid (HR = 1.53), or the Veteran's Administration (HR = 1.26). Only Medicaid in the youngest (52% higher risk) and those not insured in the oldest group (24% lower risk) were significantly different from their privately insured counterparts. Among 18-49 and 50-64 age groups there was a higher mortality risk among the lowest SES (1.17- and 1.23-fold higher in the middle age and 1.12- and 1.17-fold higher in the older age group, respectively) compared to highest SES. Married patients were significantly better off than divorced/separated (HR = 1.22), single (HR = 1.29), or widowed (HR = 1.19) patients.Factors associated with increased risk for mortality among individuals with CRC included being older, uninsured, unmarried, more comorbidities, living in lower SES neighborhoods, and diagnosed at later disease stage. Higher risk among younger patients was attributed to proximal cancer site, Medicaid, and distant disease; however, lower SES and being unmarried were not risk factors in this age group. Targeted interventions to improve survivorship and greater social support while considering age classification may assist these high-risk groups.

SUBMITTER: Tannenbaum SL

PROVIDER: S-EPMC4149543 | biostudies-other | 2014

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:ImportanceQuality improvement programs for colorectal cancer surgery have been introduced with benchmarking based on quality indicators, such as mortality. Detailed (pre)operative characteristics may offer relevant information for proper case-mix correction.ObjectiveTo investigate the added value of machine learning to predict quality indicators for colorectal cancer surgery and identify previously unrecognized predictors of 30-day mortality based on a large, nationwide colorectal cancer registry that collected extensive data on comorbidities.Design, setting, and participantsAll patients who underwent resection for primary colorectal cancer registered in the Dutch ColoRectal Audit between January 1, 2011, and December 31, 2016, were included. Multiple machine learning models (multivariable logistic regression, elastic net regression, support vector machine, random forest, and gradient boosting) were made to predict quality indicators. Model performance was compared with conventionally used scores. Risk factors were identified by logistic regression analyses and Shapley additive explanations (ie, SHAP values). Statistical analysis was performed between March 1 and September 30, 2020.Main outcomes and measuresThe primary outcome of this cohort study was 30-day mortality. Prediction models were trained on a training set by performing 5-fold cross-validation, and outcomes were measured by the area under the receiver operating characteristic curve on the test set. Machine learning was further used to identify risk factors, measured by odds ratios and SHAP values.ResultsThis cohort study included 62 501 records, most patients were male (35 116 [56.2%]), were aged 61 to 80 years (41 560 [66.5%]), and had an American Society of Anesthesiology score of II (35 679 [57.1%]). A 30-day mortality rate of 2.7% (n = 1693) was found. The area under the curve of the best machine learning model for 30-day mortality (0.82; 95% CI, 0.79-0.85) was significantly higher than the American Society of Anesthesiology score (0.74; 95% CI, 0.71-0.77; P < .001), Charlson Comorbidity Index (0.66; 95% CI, 0.63-0.70; P < .001), and preoperative score to predict postoperative mortality (0.73; 95% CI, 0.70-0.77; P < .001). Hypertension, myocardial infarction, chronic obstructive pulmonary disease, and asthma were comorbidities with a high risk for increased mortality. Machine learning identified specific risk factors for a complicated course, intensive care unit admission, prolonged hospital stay, and readmission. Laparoscopic surgery was associated with a decreased risk for all adverse outcomes.Conclusions and relevanceThis study found that machine learning methods outperformed conventional scores to predict 30-day mortality after colorectal cancer surgery, identified specific patient groups at risk for adverse outcomes, and provided directions to optimize benchmarking in clinical audits.

Project description:BackgroundThe influence of community context and individual socioeconomic status on health is widely recognized. However, the dynamics of how the relationship of neighborhood context on health varies by individual socioeconomic status is less well understood.ObjectiveTo examine the relationship between neighborhood context and mortality among older adults and examine how the influence of neighborhood context on mortality differs by individual socioeconomic status, using two measures of income-level and homeownership.Research design and subjectsA retrospective study of 362,609 Medicare Advantage respondents to the 2014-2015 Medicare Health Outcomes Survey aged 65 and older.MeasuresNeighborhood context was defined using the deciles of the Area Deprivation Index. Logistic regression was used to analyze mortality with interaction terms between income/homeownership and neighborhood deciles to examine cross-level relationships, controlling for age, gender, race/ethnicity, number of chronic conditions, obese/underweight, difficulties in activities of daily living, smoking status, and survey year. Predicted mortality rates by group were calculated from the logistic model results.ResultsLow-income individuals (8.9%) and nonhomeowners (9.1%) had higher mortality rates compared to higher-income individuals (5.3%) and homeowners (5.3%), respectively, and the differences were significant across all neighborhoods even after adjustment. With regression adjustment, older adults residing in less disadvantaged neighborhoods showed lower predicted 2-year mortality among high-income (4.86% in the least disadvantaged neighborhood; 6.06% in the most disadvantaged neighborhood; difference p-value<0.001) or homeowning individuals (4.73% in the least disadvantaged neighborhood; 6.25% in the most disadvantaged neighborhood; difference p-value<0.001). However, this study did not observe a significant difference in predicted mortality rates among low-income individuals by neighborhood (8.7% in the least disadvantaged neighborhood; 8.61% in the most disadvantaged neighborhood; difference p-value = 0.825).ConclusionsLow-income or non-homeowning older adults had a higher risk of mortality regardless of neighborhood socioeconomic status. While living in a less disadvantaged neighborhood provided a protective association for higher-income or homeowning older adults, low-income older adults did not experience an observable benefit.

Project description:BackgroundResearch on psychosocial stress and risk of breast cancer has produced conflicting results. Few studies have assessed this relation by breast cancer subtype or specifically among Black women, who experience unique chronic stressors.MethodsWe used prospective data from the Black Women's Health Study, an ongoing cohort study of 59,000 US Black women, to assess neighborhood- and individual-level psychosocial factors in relation to risk of breast cancer. We used factor analysis to derive two neighborhood score variables after linking participant addresses to US Census data (2000 and 2010) on education, employment, income and poverty, female-headed households, and Black race for all households in each residential block group. We used Cox proportional hazards regression to estimate hazard ratios (HR) and 95% confidence intervals (CI) adjusted for established breast cancer risk factors.ResultsDuring follow-up from 1995 to 2017, there were 2167 incident invasive breast cancer cases (1259 estrogen receptor positive (ER +); 687 ER negative (ER-)). For ER- breast cancer, HRs were 1.26 (95% CI 1.00-1.58) for women living in the highest quartile of neighborhood disadvantage relative to women in the lowest quartile, and 1.24 (95% CI 0.98-1.57) for lowest versus highest quartile of neighborhood socioeconomic status (SES). For ER+ breast cancer, living in the lowest quartile of neighborhood SES was associated with a reduced risk of ER+ breast cancer (HR = 0.83, 95% CI 0.70-0.98). With respect to individual-level factors, childhood sexual abuse (sexual assault ≥ 4 times vs. no abuse: HR = 1.35, 95% CI 1.01-1.79) and marital status (married/living together vs. single: HR = 1.29, 95% CI 1.08-1.53) were associated with higher risk of ER+, but not ER- breast cancer.ConclusionNeighborhood disadvantage and lower neighborhood SES were associated with an approximately 25% increased risk of ER- breast cancer in this large cohort of Black women, even after control for multiple behaviors and lifestyle factors. Further research is need to understand the underlying reasons for these associations. Possible contributing factors are biologic responses to the chronic stress/distress experienced by individuals who reside in neighborhoods characterized by high levels of noise, crime and unemployment or the direct effects of environmental toxins.

Dataset Information

Individual- and neighborhood-level predictors of mortality in Florida colorectal cancer patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets