Authors: Debarchana (Debs) Ghosh*, University of Connecticut, Shamayeta Bhattacharya*, University of Connecticut
Topics: Spatial Analysis & Modeling, Social Theory, Religion
Keywords: Cancer screenings, adaptive KDE, religious support, African Americans
Session Type: Paper
Start / End Time: 9:55 AM / 11:35 AM
Room: Marshall West, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Health geographers are increasingly incorporating point-pattern analysis into research as geocoded cases of screening, disease, or treatment become widely available. In this paper we enhance point-pattern analysis using a two-step process. First, we identify areas with high (and low) spatial relative probabilities of breast, cervical, prostate, and colorectal cancer screening practices amongst African-Americans. Second, we evaluate the associations between the identified spatial pattern of screening with participants’ religious support. We chose African-Americans because this group, on average, tends to express greater religious involvement than others and is also disproportionately impacted by many chronic conditions.
The study used yes/no encoded data of cancer-screenings geocoded at the point-level from a national survey (n=3,011). The cancer-screening tests were mammography, pap-smear, prostate-specific antigen (PSA), digital rectum exam (DRE), fecal occult blood test (FOBT), and colonoscopy. Support was measured by social, religious support, and religious involvement scales. Adaptive kernel density functions were used to measure variation in spatial relative probabilities for each of the cancer-screenings. We then used multivariate regression models to examine the association between spatial relative probabilities of screening and social and religious support after controlling for SES factors.
Higher spatial relative probabilities of screening were identified in the West South Central, East South Central, and South Atlantic divisions of the US. Positive associations between support and cancer screening were consistent but statistically significant associations were found only for some screens such as mammography, PSA, and DRE. Our unique approach highlights spatial heterogeneity in cancer screening behaviors indicating the need for targeted interventions.