Since 2011, the process of incorporation of technologies into the Brazilian public health system (SUS) has been assisted by the National Commission for the Incorporation of Technologies in SUS (Conitec). The present work collected data of effectiveness, safety, cost-effectiveness, budget impact and other criteria from Conitec's reports to determine the influence of economic evaluations on issued recommendations.

Data was collected from drug recommendation reports published by Conitec between 2012 and 2016 and organized in a Microsoft Excel® spreadsheet. The association of the incremental cost-effectiveness ratio (ICER) and the chance of incorporation was assessed through a binary logistic regression in R®.

Two hundred and sixty-six reports were issued by Conitec between 2012 and 2016. Data were collected from 169 reports evaluating requisitions of incorporation of new medicines. Of these, there were ninety-nine which recommended the incorporation. The most common ATC classes analyzed were immunosuppressants (34 drugs), other antineoplastic agents (16 drugs) and direct-acting antivirals (15 drugs). Of the seventy negative recommendations, thirty-five were due to cost-effectiveness, thirty-one due to efficacy, twenty-nine due to safety, forty due to the budget impact, and thirty-two due to other reasons. In general, the reports were considered to be of poor quality. Only 21.9 percent of the reports had ICERs. The binary logistic regression analysis did not present a statistically significant difference for the influence of the ICER on the recommendation decision with outcomes reported in life years gained (OR = 0.9999732; 95% Confidence Interval [CI] = 0.9999304 to 1.000016) or quality-adjusted life years (OR = 0.9999789; 95% CI = 0.9999321 to 1.000026).

Economic evaluations appear to be a secondary criterion for Conitec's recommendations. Despite this, they are commonly used to justify non-incorporation of drugs into the public system.