E133: Using Quantitative and Geospatial Analysis to Identify Populations at Risk for Late Diagnosis of Cancer

Over 50% of certain cancers are diagnosed at an advanced stage with metastasis. These patients have a very poor prognosis with no curative options. This may be due in part to certain demographic and social factors. This study aimed to use quantitative and geospatial analysis to identify neighborhoods and populations at risk of presenting with metastasis.

Methods:

We used Texas and California cancer registries, merged with CDC’s Social Vulnerability Index (SVI) database. We identified patients diagnosed with the top 5 leading causes of cancer deaths from 2004 to 2019 for adult males and females. This included 7 cancers namely breast, pancreatic, liver, colorectal, lungs, prostate, and ovarian cancer. We then used the patient’s geographical information to ascertain the SVI of their census tract. To determine the association of demographic and social vulnerability factors with metastatic cancer at presentation (at or within first 4 months of diagnosis), univariate and multivariate logistic regression analysis (MVA) was performed, stratified by cancer type. We then geospatially mapped the census tracts with proportions of metastatic cancer and SVI, both categorized as tertiles.

Results:

We included 2,196,828 patients and of these, 473,506 (21.5%) patients presented with metastasis. This included more than 55% of all pancreatic, lung and ovarian cancers, and less than 7% of all breast and prostate cancers. On MVA, age was significantly associated with odds of metastatic cancer at presentation however the trends across age groups varied for each cancer. Across all cancer types, Black patients, uninsured and patients on Medicaid had increased odds of metastatic cancer at presentation. Compared with the most affluent neighborhoods (SVI Tertile 1) odds of metastatic cancer at presentation increased with increasing social vulnerability with the highest for tertile 3 for prostate (OR 1.41, 95%CI 1.36 – 1.45), breast (OR 1.33, 95%CI 1.29 – 1.37), liver (OR 1.22 95%CI 1.16-1.29), lungs (OR 1.12 95%CI 1.10 – 1.13), pancreas (OR 1.09 95%CI 1.06 – 1.13) and colorectal (OR 1.04, 95%CI 1.02 – 1.06) cancers.

Conclusions:

Increasing social vulnerability and multiple demographic factors are independently associated with odds of having metastatic cancer at presentation. In addition, we geospatially mapped census tracts with high social vulnerability and proportions of metastatic cancer. Screening and early detection interventions aimed towards these at-risk neighborhoods offer an opportunity for timely diagnosis and potentially improved survival.