Association of pre-diagnosis primary care utilization with metastatic disease, treatment adherence, and mortality in colorectal cancer.

Abstract

3664 Background: In addition to timely evaluation and screening in patients with colorectal cancer (CRC), routine primary care provider (PCP) engagement may improve CRC survival outcomes through earlier diagnosis and increased treatment adherence. We evaluated the association between pre-diagnostic PCP utilization and metastatic disease at diagnosis, cancer-specific mortality (CSM), and cancer treatment adherence. Methods: Using the Surveillance, Epidemiology, and End Results (SEER)-Medicare database, we conducted a retrospective cohort study of adults aged 68-75 diagnosed with CRC between 2010-2017. PCP visits in the three years preceding diagnosis were categorized as none (0 visits), occasional (1–2), or annual (3). Multivariable logistic regression models evaluated associations between PCP use and metastatic CRC at diagnosis, as well as treatment adherence. Locally advanced (Stage III) treatment adherence was defined as curative therapy ≤1 year from diagnosis, and metastatic treatment adherence was defined as systemic therapy ≤6 months from diagnosis. A competing risk Cox proportional hazards model evaluated cancer-specific mortality (CSM), with non-cancer death as a competing event. Results: Of 18,604 patients with CRC, 12.9% had no prior PCP visits, 48.6% had occasional visits, and 38.5% had annual visits. Compared to patients with no pre-diagnostic PCP use, occasional and annual PCP visits were respectively associated with 43% and 57% lower odds (p < 0.001) of metastatic disease at diagnosis and 48% and 57% lower odds (p < 0.001) of CSM (Table 1). Among patients with locally advanced CRC, occasional and annual PCP use were associated with 64% and 150% higher odds of receiving curative treatment, respectively (Table 1). Similarly, occasional and annual PCP use in patients with metastatic CRC were associated with 2-fold and 3-fold higher odds, respectively, of receiving systemic treatment (Table 1). Conclusions: Primary care utilization prior to CRC diagnosis is significantly associated with lower odds of metastatic disease, lower risk of CSM, and higher odds of treatment adherence, with the highest effects observed with annual PCP utilization. Increasing PCP engagement is an essential strategy to improving outcomes in patients with CRC. Multivariate estimates for the effect of pre-diagnostic primary care provider (PCP) utilization prior to cancer diagnosis. Metastatic Disease at DiagnosisOR, [95% CI], p-value Cancer-Specific MortalitysHR, [95% CI], p-value Received Treatment (Locally advanced CRC)OR, [95% CI], p-value Received Treatment (Metastatic CRC)OR, [95% CI], p-value 1-2 PCP visits (Occasional) 0.57 [0.51-0.64], p < 0.001 0.52 [0.48-0.56], p < 0.001 2.64 [2.10-3.30], p < 0.001 2.08 [1.70-2.56], p < 0.001 3+ PCP visits(Annual) 0.43 [0.40-0.47], p < 0.001 0.43 [0.40-0.47], p < 0.001 3.50 [2.76-4.24], p < 0.001 2.90 [2.31-3.64], p < 0.001