Proximity to clinical care and time to resolution following an abnormal cancer screening in an urban setting
Barriers to care have been identified as a major factor in cancer health disparities. Previous research at Boston Medical Center (BMC) found that women referred from community health centers (CHCs) following abnormal breast cancer screening took longer to achieve diagnostic resolution than women referred from a BMC-based practice, consistent with research showing longer delays and worse outcomes for disadvantaged urban populations. It is not known whether this difference relates to the additional distance to BMC. To evaluate the effect of proximity from subjects' residence to the site of clinical care on time to diagnostic resolution in this urban setting we conducted a secondary analysis using data collected as part of the Boston Patient Navigation Research Program (PNRP). The database included all women who had a breast or cervical cancer screening abnormality at six Federally-qualified CHCs from January 2007 to June 2009. Using geocoded home address data captured at the time of registration, we calculated straight-line distances to the location of the diagnostic evaluation, which was the CHC for subjects with a cervical abnormality or BMC for subjects with a breast abnormality, and plotted the time to diagnostic resolution versus distance to site of care. We used proportional hazards regression models to examine the effect of distance to site of care on time to resolution, adjusting for CHC, subject age, race/ethnicity, language, and insurance. Results. We geocoded addresses for 1512 of 1544 subjects (98%). Among the diverse group of subjects with a breast screening abnormality (36% Black, 33% Hispanic; 44% non-English speaking), there was no significant difference in adjusted hazard ratios based on distance to care in 1,000 meter units (adjusted Hazard Ratio 1.00, 95% CI 0.99 -1.01). Similarly, among those with a cervical screening abnormality (22% Black, 21% Hispanic; 15% non-English), there was no significant difference in adjusted hazard ratios based on distance to care in 1,000 meter units (adjusted Hazard Ratio 1.01, 95% CI 1.00- 1.02). Conclusions. Increased distance between residence and clinic alone is not a barrier to diagnostic resolution for this vulnerable urban population receiving care at a CHC who had an abnormal cancer screening exam.
Thesis (M.S.)--Boston University