Purpose: To assess delayed-type cutaneous reactions (DTCRs) related to drugs, using a case-control approach to qualify drug risks.

Methods: The study used the Tunisian pharmacovigilance database of Monastir. The association between drugs and DTCRs was assessed using a case/non-case method. Drugs were grouped according to the ATC Classification System. Patients were defined as "cases" if they have developed DTCRs regardless of the causality assessment. All other reports were "non-cases". Association between reactions and drugs was calculated using the reporting odds ratio (ROR) with 95% confidence intervals (CIs). A p value < 0.05 was considered significant.

Results: The analysis was carried out on 1798 reports, of which 867 concerned DTCRs (cases) and 931 concerned non-cases. The calculated risk estimates were significant for cefotaxime (ROR 2.1; 95% CI 1.5 to 3), pristinamycin (ROR 4; 95% CI 2 to 7.9), sulfamethoxazole (ROR 4.4; 95% CI 1.6 to 11.7), oxacillin (ROR 2.2; 95% CI 1.2 to 3.8), doxycycline (ROR 10.8; 95% CI 1.4 to 84.9), carbamazepine (ROR 3.3; 95% CI 1.7 to 6.2), phenobarbital (ROR 2.3; 95% CI 1.03 to 5.1), allopurinol (ROR 3.6; 95% CI 1.8 to 7.2), furosemide (ROR 2.4; 95% CI 1.3 to 6.3), hydrochlorothiazide(ROR 2.9; 95% CI 1.3 to 6.3) and candesartan (ROR 4.7; 95% CI 1.3 to 16.6).