Analysis of Delays in Breast Cancer Treatment and Late-Stage Diagnosis in Kazakhstan By Using Electronic Cancer Registry Data

Abstract

Abstract 27 Background: Although Kazakhstan (KZ) has made significant investments to improve population health, high cancer rates persist, with breast cancer as the most prevalent. We assessed factors that contribute to delays in treatment and late staging for patients with breast cancer. Methods: A retrospective follow-up study design was used. By using 2014 registry data, we identified 4,248 patients with breast cancer who were treated at cancer centers in the 16 KZ regions. Patients with delays in treatment as a result of medical errors and other reasons were identified. We used logistic regression to estimate associations of delays with patient demographics, occupation, and cancer center region; and associations of late-stage (III and IV) cancer diagnosis with delays in treatment while controlling for patient demographics, occupation, and cancer center region. Results: Approximately 9% (n = 378) of patients experienced delayed treatment. Older and Russian patients as well as those treated in regions further away from Almaty City had significantly higher adjusted risk of delayed treatment. However, risk of late-stage diagnosis was greater for patients who were treated in Almaty City and for those who were Russian, unemployed, or who had delayed treatment. Conclusion: The main driver of delayed treatment was cancer center region. Patients who were treated in Almaty City, where the national cancer research and treatment center is located, had fewer delays in treatment but a higher likelihood of late-stage diagnosis, likely as a result of referrals of sicker patients from neighboring regions. These findings suggest that referrals to new tertiary care centers being developed in KZ may reduce treatment delays but affect facility case mix. Future research will examine the role played by distance to cancer centers in access to specialty care and whether tertiary care is associated with improved outcomes conditional on case mix. AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST No COIs from the authors.