Prediction of attendance to the "60 days law" within the patient navigation program to breast cancer in Rio de Janeiro, Brazil.

Abstract

e13100 Background: An applied study was conducted on how the use of machine learning techniques can help in the process of identifying compliance with the "60 Day Law", which states that all patients with cancer within the public system must initiate the treatment within 60 days after the diagnosis of cancer. Within the Patient Navigation Program (PNP) for breast cancer in Rio de Janeiro, the study aims to: 1) identify barriers to compliance with the Law; 2) ensure that at least 70% of patients recruited with breast cancer initiate treatment within the mandatory 60-day period; and 3) to construct a model that accurately predicts whether or not a patient meets the period established in the Law. Methods: From August 2017 to May 2018, 105 patients aged 33-80 years (mean 59 years) were recruited for navigation. For the development of the statistical analysis, three learning models were used AdaBoost, Decision Tree and GaussianNB. Results: Patients presented 0-I (17%), II-III (78%) and IV (5%) staging. The main barriers to compliance with the Law were fear and fatalistic thoughts (99%), financial problems (79%), and uncoordinated health care (76%).The PNP had 100% patient satisfaction and in 52% of the cases it helped at the beginning of the treatment within the period established by law. The AdaBoost learning model had superior results in relation to accuracy and f-score (0.8889 and 0.8333, respectively). Conclusions: The PNP generated a positive experience in the public health system, because it is an intentional and proactive process of individual assistance through the health system, accessing services, and actively overcoming the barriers to quality care. The study did not reach the success rate of 70% compliance with the Law as intended (having reached 52%). However, the barriers that NP cannot overcome, such as the lack of human resources and medical supplies, have been reported to health authorities and hospital administrators. We identified 38 important attributes for compliance with the Law, which simplifies the information required for model learning. In the Brazilian context, the PNP may represent an opportunity to adequately implement existing legislation and, as such, would have great potential for integration into federal, state, and local health systems. Clinical trial information: 62728616.5.0000.5274.