Development of a Multilevel Model to Identify Patients at Risk for Delay in Starting Cancer Treatment.

Importance: Delays in starting cancer treatment disproportionately affect vulnerable populations and can influence patients' experience and outcomes. Machine learning algorithms incorporating electronic health record (EHR) data and neighborhood-level social determinants of health (SDOH) measures may identify at-risk patients. Objective: To develop and validate a machine learning model for estimating the probability of a treatment delay using multilevel data sources. Design, Setting, and Participants: This cohort study evaluated 4 different machine learning approaches for estimating the likelihood of a treatment delay greater than 60 days (group least absolute shrinkage and selection operator [LASSO], bayesian additive regression tree, gradient boosting, and random forest). Criteria for selecting between approaches were discrimination, calibration, and interpretability/simplicity. The multilevel data set included clinical, demographic, and neighborhood-level census data derived from the EHR, cancer registry, and American Community Survey. Patients with invasive breast, lung, colorectal, bladder, or kidney cancer diagnosed from 2013 to 2019 and treated at a comprehensive cancer center were included. Data analysis was performed from January 2022 to June 2023. Exposures: Variables included demographics, cancer characteristics, comorbidities, laboratory values, imaging orders, and neighborhood variables. Main Outcomes and Measures: The outcome estimated by machine learning models was likelihood of a delay greater than 60 days between cancer diagnosis and treatment initiation. The primary metric used to evaluate model performance was area under the receiver operating characteristic curve (AUC-ROC). Results: A total of 6409 patients were included (mean [SD] age, 62.8 [12.5] years; 4321 [67.4%] female; 2576 [40.2%] with breast cancer, 1738 [27.1%] with lung cancer, and 1059 [16.5%] with kidney cancer). A total of 1621 (25.3%) experienced a delay greater than 60 days. The selected group LASSO model had an AUC-ROC of 0.713 (95% CI, 0.679-0.745). Lower likelihood of delay was seen with diagnosis at the treating institution; first malignant neoplasm; Asian or Pacific Islander or White race; private insurance; and lacking comorbidities. Greater likelihood of delay was seen at the extremes of neighborhood deprivation. Model performance (AUC-ROC) was lower in Black patients, patients with race and ethnicity other than non-Hispanic White, and those living in the most disadvantaged neighborhoods. Though the model selected neighborhood SDOH variables as contributing variables, performance was similar when fit with and without these variables. Conclusions and Relevance: In this cohort study, a machine learning model incorporating EHR and SDOH data was able to estimate the likelihood of delays in starting cancer therapy. Future work should focus on additional ways to incorporate SDOH data to improve model performance, particularly in vulnerable populations.

Defining aggressive prostate cancer: a geospatial perspective.

BACKGROUND: Spatial analysis can identify communities where men are at risk for aggressive prostate cancer (PCan) and need intervention. However, there are several definitions for aggressive PCan. In this study, we evaluate geospatial patterns of 3 different aggressive PCan definitions in relation to PCan-specific mortality and provide methodologic and practical insights into how each definition may affect intervention targets. METHODS: Using the Pennsylvania State Cancer Registry data (2005-2015), we used 3 definitions to assign "aggressive" status to patients diagnosed with PCan. Definition one (D1, recently recommended as the primary definition, given high correlation with PCan death) was based on staging criteria T4/N1/M1 or Gleason score ≥ 8. Definition two (D2, most frequently-used definition in geospatial studies) included distant SEER summary stage. Definition three (D3) included Gleason score ≥ 7 only. Using Bayesian spatial models, we identified geographic clusters of elevated odds ratios for aggressive PCan (binomial model) for each definition and compared overlap between those clusters to clusters of elevated hazard ratios for PCan-specific mortality (Cox regression). RESULTS: The number of "aggressive" PCan cases varied by definition, and influenced quantity, location, and extent/size of geographic clusters in binomial models. While spatial patterns overlapped across all three definitions, using D2 in binomial models provided results most akin to PCan-specific mortality clusters as identified through Cox regression. This approach resulted in fewer clusters for targeted intervention and less sensitive to missing data compared to definitions that rely on clinical TNM staging. CONCLUSIONS: Using D2, based on distant SEER summary stage, in future research may facilitate consistency and allow for standardized comparison across geospatial studies.

The effect of social determinants of health on utilization of surgical treatment for hepatocellular carcinoma patients.

BACKGROUND: A paucity of data exists on how social determinants of health (SDOH) influence treatment for Hepatocellular carcinoma (HCC). We investigated associations between SDOH (healthcare access, education, social/community context, economic stability, and built/neighborhood environment) and receipt of surgery. METHODS: The Pennsylvania Liver Cancer Registry was linked with neighborhood SDOH from the American Community Survey. Multilevel logistic regression models with patient and neighborhood SDOH variables were developed. RESULTS: Of 9423 HCC patients, 2393 were stage I. Only 36.3% of stage I patients received surgery. Black patients had significantly lower odds of surgery vs Whites (OR = 0.73; p < 0.01), but not after adjustments for SDOH. All 5 SDOH domains were associated with odds of surgery overall; 2 domains were associated in Stage I patients, social context (e.g., racial concentration, p = 0.03) and insurance access (p < 0.01). CONCLUSIONS: SDOH impact utilization of surgery for HCC. Findings can guide healthcare professionals to create programs for populations at risk for poor liver cancer outcomes.