Signal detection for Thai traditional medicine: examination of national pharmacovigilance data using reporting odds ratio and reported population attributable risk

Herbal containing medicine consumption has increased while the awareness of adverse drug reaction (ADR) was less than conventional medicine. Early detection of unexpected numbers of ADRs from herbal medicines' reports which are abnormal from the whole database needs quantification. Disproportionality analysis has been performed for signal detection by using reporting odds ratio (ROR) as measurement. The impact of having medicine as exposures in each ADR should be measured by using reported population attributable risks (RPAR). This study aimed to quantify the contribution of Thai traditional medicine (TTM) to ADR reports and to assess the association between TTMs and serious adverse drug reactions. Data were retrieved from the adverse drug reaction surveillance database, Thai-Food and Drug Administration from 2002 to 2013. Crude and adjusted RORs for each drug-ADR pair and RPARs were computed. TTM contributed only 0.001% of all serious ADRs reported. Out of 4208 TTM-ADR pairs were examined, three had the statistically significant RORs, namely Andrographis paniculata and anaphylactic shock (ROR 2.32, 95% CI 1.03, 5.21); green traditional medicine and Stevens-Johnson syndrome (ROR 13.04, 95% CI 5.4-31.51) and Derris scandens Benth and angioedema (ROR 2.71, 95% CI 1.05-6.95). Their RPARs ranged from 0.05% to 0.16%. We conclude that TTMs need more intensive surveillance.