African American (AA) persons are exposed to disproportionately high levels of social and environmental stressors, including low socioeconomic status (SES), social isolation, interpersonal and institutional discrimination, and residence in resource poor communities, which intersect at individual- and neighborhood- levels and are associated with poor lifestyle factors, healthcare utilization, and health outcomes. The glaring disparity in prostate cancer mortality between AA men and Non-Hispanic (NH) White men appears to be due to complex biological, socioeconomic, and socio-cultural determinants underlying disparities in presentation, diagnosis, treatment, and survival. Our ability to elucidate the impact of neighborhood contextual factors on health outcomes is complicated by differential research study participation rates by race/ethnicity, which persistently threatens generalizability of epidemiological findings. To examine the impact of neighborhood contextual factors on prostate cancer disparities, we conducted epidemiological studies, at statewide and regional levels. At a statewide level, we examined a population-based retrospective cohort of 17,787 AA and 112,591 NHW men diagnosed with prostate cancer men in California. We examined racial/ ethnic differences in occurrence of prostate cancer aggressiveness as defined by binary outcomes of high PSA, high Gleason score grade (GS), and high stage using multivariable logistic regression. We observed evidence that racial/ ethnic disparities in prostate cancer aggressiveness at diagnosis for AA men relative to NHW men in California were driven by high PSA, not high GS or stage. Specifically, AA men experienced an approximate 79% increase in odds of high PSA prostate cancer relative to NHW men, after full adjustment for year and age at diagnosis, marital status, and health insurance type, as well as stage and grade (odds ratio (OR)= 1.79; 95% confidence interval (CI)=1.69-1.90). On the other hand, when we added PSA as an independent variable to fully adjusted models, we observed that the OR for race (AA vs. NHW) was null for the high GS model (OR=0.97; 95% CI=0.92-1.03) and reduced for the high stage model (OR=0.86; 95% CI=0.80-0.91). At a regional level, we examined the population-based ChicagO Multiethnic Prevention and Surveillance Study (COMPASS). For our investigation of the impact of neighborhood contextual factors on bio-specimen research participation, we used the COMPASS household database (consolidated from postal service, commercial vendor and interviewer recruitment database information) enriched with a time-invariant census tract-level measure on neighborhood SES. In multivariable logistic regression models controlling for summarized data on households, interviewers, and design characteristics at addresses, we observed approximated three times the odds of research participation for predominantly AA households in the original target sample within low vs. average SES neighborhoods (OR=3.06; 95% CI=2.20-4.24) and no difference in odds of research participation for AA households in the original target sample within high vs. average SES neighborhoods (OR=0.94; 95% CI=0.71–1.25). These findings suggested that door-to-door recruitment and financial compensation ($50 in our study) were effective strategies to recruit traditionally under-represented racial/ ethnic minority participants in COMPASS. In our other COMPASS study, we examined the impact of self-reported lifestyle and healthcare factors on serum PSA levels based on clinical laboratory testing, among 928 AA men of predominantly low SES in COMPASS. Specifically, we examined the associations between self-reported cigarette smoking pack-years, other current regular tobacco use (including e-cigarettes and cigars), and current regular marijuana use on PSA in multivariable logistic regression models with outcome of elevated PSA 4.0+ ng/ mL and linear regression models with outcome of increasing PSA (continuous), after adjustment for age, marital status, individual and neighborhood SES, self-reported health, hypertension medication, body mass index (BMI), health insurance type, and quintiles of visits to doctor in last 12 months. Among fully adjusted stratified models of 430 AA men age 55+ years, we observed approximately 5 times the odds of elevated PSA among those with 1+ pack-years of cigarette smoking vs. never smokers (OR=5.03; 95% CI=1.56-16.2), a quarter the odds of elevated PSA among current marijuana users vs. non-users (OR=0.28; 95% CI-0.08-0.99), and a mean PSA increase of 1.25 ng/ mL among other current tobacco users vs. non-users. We interpreted these findings to suggest that cigarette smoking history and current tobacco use were adversely related to PSA risk profile among AA men in predominantly low SES neighborhoods, and that PSA testing may be an inappropriate biomarker of PSA risk profile among current marijuana users.