RESUMEN
OBJECTIVES: Prevalence of potentially inappropriate medications (PIMs) and potential prescribing omissions (PPOs). DESIGN: Cross-sectional study through review of residents' records. SETTING: Three nursing homes of Puerto Rico's metropolitan area. PARTICIPANTS: Nursing home residents. INTERVENTION: Records of residents 65 years of age and older with documented medications and chronic diseases were reviewed. Hospitalized and hospice/palliative care residents were excluded. Beers criteria-2015 and START-STOPP criteria-2014 were applied to identify PIMs and PPOs. Beers criteria versus STOPP criteria findings were compared. MAIN OUTCOME MEASURE: Prevalence of PIMs and PPOs in relation to residents' baseline characteristics. RESULTS: One hundred four records were reviewed. Categorical variables were analyzed by Fisher's exact test and chi square test; t-tests established a relationship between variables and the prevalence of PIMs and PPOs. The quantity of PIMs was analyzed through frequency and percentage values. A statistically significant correlation was found between prescribed medications and PIMs detected using Beers criteria (P-value = 0.031) and STOPP modified criteria (P-value = 0.0002). No statistically significant data were obtained from the START criteria. CONCLUSION: Beers criteria were more effective identifying PIMs compared with STOPP criteria. The number of prescribed medications correlated directly to the number of identified PIMs. Additional studies where these tools are applied, and clinical pharmacist interventions are performed, should confirm an improved management of nursing home residents' drug therapy in Puerto Rico.