Prevalence of potential drug interactions of clinical importance in primary health care and its associated factors / Prevalência de potenciais interações medicamentosas de importância clínica na atenção primária à saúde e seus fatores associados

Le résumé fourni par la source

Introduction: older patients usually had multiple diseases and so use many medicines. The elevated risks of pharmacotherapy in this population justified the development of ratings for unsafe medicines.Objective: to estimate the prevalence of potential drug interactions of clinical importance in primary health care and its associated factors, improving prescription practices and increasing patient safety.Methods: a cross-sectional study of dependent variable “number of potential drug-drug interactions of clinical importance”, in all medicines and patients who accessed medicines via public primary health care, 2013. The independent variables were socio demographic, accessibility of health services and pharmacotherapy. Multivariate analyses were performed using the Statistical Learning Theory with Exaustive-CHAID algorithm, with test Pearson's chi-square adjusted by the Bonferroni method.Results: a total of 4,037 patients were included in this study and the patient prevalence of at least one drug-drug interaction was 36.5% with severity moderate (66.2%) or major (28.5%). The most prevalent conduct for management of them were monitor the patient (59.0%), adjust the dosage of the medicines (21.9%) and monitoring signs and symptoms (16.7%). In the multivariate analysis by the Theory of Statistical Learning when we compared the “patients who had at least one drug interaction of clinical importance” with those who did not have them at the first hierarchical level of relevance, the variable “number of drugs in use” prevailed with a p value 0.0001. The analysis also proposed 7 different risk strata to explain the distinction between having at least one interaction of clinical importance, namely: 1, 2, 3, 4, 5, 6-7 and 8 drugs. When comparing patients with 2 medications and those with 8 or more medications, the prevalence of drug interactions increases by about 80%. Using polypharmacy (5 or more drugs) as the cutoff point to make the same comparison, the increase is about 45%. Other variables with statistical relevance to explain having or not having hair were “multiple drug dispensations per month” (p = 0.003 and p = 0.01) and “being elderly” (p = 0.003). Having “multiple drug dispensations per month” reduced the prevalence of interactions by about 10% for both patients with 3 medications (p = 0.003) and those with 6 or 7 medications.Conclusions: the drug-drug interactions showed be different in primary care of hospitals and other place for health care. And the number of medications in use by the patient seems to be the main marker for patient selection for this type of analysis, with polypharmacy being a relevant cutoff point, but above all the use of 8 or more medications indicates a prevalence of more than 90% patients of at least one interaction of clinical importance. There are few studies of potential drug-drug interactions in public primary health care, especially with analysis of the severity and management of them. We recommend more studies for clarify prevalence, types and associated factors.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.