Pertussis, a severe respiratory infection caused by Bordetella pertussis, is distributed globally. Vaccination has been crucial to annual reductions in the number of cases. However, disease reemergence has occurred over the last decade in several countries, including Brazil. Here we describe the clinical and epidemiological aspects of suspected pertussis cases in Salvador, Brazil, and evaluate factors associated with case confirmation. This descriptive and retrospective study was conducted in the five hospitals in Salvador that reported the highest number of pertussis cases between 2011-2016. Demographic and clinical data were recorded for each patient. Bivariate analysis was performed to evaluate differences between groups (confirmed vs. unconfirmed cases) using Pearson's Chi-square test or Fisher's exact test. Results: Of 529 suspected pertussis cases, 29.7% (157/529) were confirmed by clinical, clinical-epidemiological or laboratory criteria, with clinical criteria most frequently applied (63.7%; 100/157). Unvaccinated individuals (43.3%; 68/157) were the most affected, followed by age groups 2-3 months (37.6%; 59/157) and <2 months (31.2%; 49/157). Overall, ≤50% of the confirmed cases presented a complete vaccination schedule. All investigated cases presented cough in association with one or more symptoms, especially paroxysmal cough (66.9%; 105/529) (p = 0.001) or cyanosis (66.2%; 104/529) (p<0.001). Our results indicate that pertussis occurred mainly in infants and unvaccinated individuals in Salvador, Brazil. The predominance of clinical criteria used to confirm suspected cases highlights the need for improvement in the laboratory tools used to perform rapid diagnosis. Fluctuations in infection prevalence demonstrate the importance of vaccination strategies in improving the control and prevention of pertussis.