User-related pharmaceutical care problems and factors affecting them: the importance of clinical relevance

Many studies determined the number and nature of user-related Pharmaceutical Care Problems (PCP) and factors affecting them, but none considered the inclusion of clinical relevance. The aims of this study are (i) to investigate the type, number and clinical relevance of user-related PCP self-reported by home dwelling elderly on polypharmacy and (ii) to develop a risk-model for detecting elderly drug-users at risk of user-related PCP.

The study was a cross-sectional study conducted among 286 home dwelling elderly on polypharmacy (> or =75 years, > or =4 medicines) in the Netherlands. The user-related PCP found were divided into problem categories and subsequently a pharmacist and a general practitioner classified the problems into those with low and those with (potential) clinical relevance. Factors possibly associated with PCP (both for all and relevant problems) were identified, and subsequently tested in multivariate models using logistic regression.

Three hundred and ninety-eight user-related PCP were observed in 189 patients (66% of all participants). After classification of user-related PCP only 26% appeared to be of clinical relevance (26% of all participants). When including clinical relevance a shift in predominantly present problem categories is observed. Furthermore, the risk model for problems with clinical relevance contains more factors than the model which considered all problems. Factors associated with clinically relevant PCP are emotional or physical problems interfering with social life, communication skills (vision and hearing), using tablets that have to be divided, using inhaled medicines, and the number of medicines used. This risk-model has a specificity of 92% and a sensitivity of 32%.

Although user-related PCP were seen in about two-thirds of the participants, in only one out of four participants was the PCP considered to be of clinical relevance. With inclusion of clinical relevance, other problem categories become more dominant. A specific risk model is designed to select elderly patients that are most likely to have PCP in need of more urgent intervention. Unfortunately higher specificity is accompanied by low sensitivity in the present model.