Specialist visits and initiation of cancer-directed treatment among a large cohort of men diagnosed with prostate cancer

Abstract

BackgroundThe urologist generally manages the treatment of men immediately following the diagnosis of prostate cancer (PCa). The role of other physician specialists in this setting is less clear. We investigated whether involvement of other physician specialty types immediately following diagnosis affects initiation of cancer-directed treatment. MethodsThis is a retrospective cohort study using linked cancer registry and claims data from 1999 to 2009, excluding stage I/II PCa. A physician visit index (PVI) served as the exposure variable and captured the “dispersion of care” across specialties, that is, the extent to which patient care involved different types of physician specialties such as the primary care physician, urologist, or oncologist. The PVI score was calculated using visits occurring within 30 days postdiagnosis. This score was dichotomized to measure “low PVI” (reflects seeing multiple specialist types). Competing risk Cox proportional hazard regression models provided adjusted hazard ratios (HR) for treatment receipt associated with a low PVI. ResultsThe sample included 33,380 patients: 4,910 metastatic and 28,470 nonmetastatic groups. The top 3 visit categories within 30 days postdiagnosis were “urologist only” (59%) and “urologist plus primary care physician” (21%) and no visit (6%). The median time to receipt of cancer-directed treatment was 51 days. Overall, 29% of individuals in the metastatic group and 38% in the nonmetastatic group were categorized as low PVI. A low PVI was associated with a shorter time to treatment receipt in the nonmetastatic (HR = 1.12 [95% CI: 1.09–1.15]) and metastatic (HR = 1.21 [95% CI: 1.14–1.29]) groups. ConclusionsMultispecialist involvement in the weeks following diagnosis is associated with a shorter time to treatment initiation, highlighting a role for exposure to different specialty types in the weeks following an initial diagnosis of PCa. This study provides important baseline data for future studies examining coordination of care across cancer and noncancer specialists.