Background The problem of overprescribing and failure to provide appropriate medicines in elderly people has been associated with hospitalisation.

Purpose To analyse Inappropriate Prescribing (IP) and Prescribing Omissions (PO) in elderly patients in an attempt to find a link with said patients’ hospitalisation.

Materials and methods Retrospective observational study from June to August 2013 with patients admitted to the Internal Medicine Unit in a tertiary hospital. We included patients ≥65 with at least four medicines in their home treatment. We used the STOPP/START criteria to detect IP/PO and compared them to the hospital’s diagnosis. We analysed IP/PO resolved after the hospitalisation process (treatment prescribed at discharge). Exclusion criteria: patients transferred to other units and patients who died. Data collected: sex, age, high comorbidity (Charlson Index ≥2), home treatment, diagnosis, hospitalisation duration, STOPP/START criteria.

Results 93 patients were included: 52% male, 48% female; Average age was 79 (range = 66–92), 100% high comorbidity; average of 9 medicines (range = 3–17) prescribed per patient; average of 7 days’ (range = 1–27) hospitalisation. We identified 34 IP and 72 PO in 62% of patients (1.8 criteria/ patient). 71% of criteria related to the diagnosis and 21% of criteria unrelated to the diagnosis were solved after the hospitalisation process.

Table 1 lists the IP or OP related to the diagnosis, found in 21 patients (26%).

Conclusions Most under- or overprescribing related to the reason for admission was detected and solved, but the rest is often unnoticed and continues unresolved in the treatment prescribed at discharge. STOPP/START criteria should be incorporated into primary care practice to improve prescription accuracy in older people and prevent morbidities.

No conflict of interest.