Background The prevalence of drug-related problems increases with age. One important cause is drug-drug interactions, which contribute to hospital admissions among the elderly. Elderly people with dementia are particularly vulnerable.

Purpose The aim of the present study was to assess the occurrence and characteristics of drug-drug interactions and to investigate potential risk factors among elderly people with dementia.

Material and methods Medical records of 458 people aged ≥65 years, with dementia or cognitive impairment that were admitted to two hospitals in northern Sweden between 9 January 2012 and 2 December 2014, were reviewed retrospectively. Information on medication use at the time of admission was collected. Clinically-relevant drug-drug interactions requiring either dose adjustments or avoidance of concomitant use were identified using the Janusmed interactions database. Interactions were further classified regarding pharmacological mechanism, i.e. pharmacokinetic or pharmacodynamic interactions and their subdivisions, according to Stockley’s classification system. Descriptive statistics and simple and multiple logistic regressions were used to analyse data.

Results Four hundred and one drug-drug interactions were identified and 43.2% of the persons had at least one interaction. This is in line with, or somewhat higher than, results shown in other studies. In 95.8% of cases interactions required dose adjustment and in 4.2% of cases it was considered that the drug combination should be avoided. Pharmacokinetic interactions were most frequently observed of which warfarin – acetaminophen (n=26) was most common. Among pharmacodynamic interactions, furosemide – citalopram (n=35) and acetylsalicylic acid – citalopram (n=32) were the most frequently observed. An association was found in the multivariable model between the number of medications prescribed in the individual patient and the risk of one or more drug-drug interaction (OR 1.312, 95% CI: 1.227 to 1.403). No associations were found between sex, age, MMSE score, type of accommodation or geographic location, and the risk of at least one drug-drug interaction.

Conclusion Clinically-relevant drug-drug interactions are prevalent among elderly people with dementia living in northern Sweden. To avoid drug-related problems, the risk of drug-drug interactions should be noted, especially in the present population. This is particularly important with increasing numbers of medications prescribed.

No conflict of interest