Needing to access a separate computer workstation for patient-specific treatment recommendations was seen as time consuming and a barrier to the use of the CDSS.[19,25] Pharmacists could simply ignore the care suggestions by not accessing the computer[19] or pressing the Escape key on their keyboard.[23] Similarly, CDSSs for physicians have been noted to be less effective if not integrated into the clinical workflow.

Integration also allows the development of systems whereby pharmacists Selleck PI3K inhibitor cannot bypass alerts and recommendations without providing a ‘response’ or annotation that the suggestion was acted upon or overridden. Chabot et al.[18] reported that many aspects of the CDSS software were not accessed in a QUM intervention to improve hypertension management and blood-pressure control in community pharmacy. A lack of patient interest and pharmacist time were cited as major barriers in this study. Notably, there were only two recommendations to MG 132 physicians to increase doses of antihypertensives, but 205 pharmacist contacts with 91 patients (most interventions were encouragement of patients). Tierney et

al.[23,24] noted in their two QUM studies of care suggestions for asthma, COPD, ischaemic heart disease and heart failure that contacts between pharmacists and physicians were very limited. The effectiveness of any intermediary role for pharmacists depends on the effectiveness of the communication channels. These observations on the QUM studies suggest there may be a degree of reluctance on behalf of the pharmacist to ‘meddle’ with the decisions of doctors[26] when the discussion is about the choice of medicine. This reluctance was not manifest in the CDSSs addressing safety issues (critical drug interactions, drugs in pregnancy and the like), where studies were strongly in favour of CDSSs. This is familiar territory for pharmacists and a more clearly delineated professional role. Although based on a larger number of studies than the Calabretto et al.[10] review (21 compared with four studies), the evidence provides limited practical

guidance on pharmacy CDSSs. With only one study conducted outside of the USA and three in community pharmacy settings, the generalisability check and applicability of the findings are limited. The remaining studies were conducted in a small number of facilities in the USA, with two research groups accounting for six of 10 QUM studies[16,17,19,20,23,24] and four of 11 of the drug-safety interventions.[33–36] The methods used by the groups were similar in their studies, the differences mainly related to clinical target, and to a lesser extent the setting for the intervention. This provides little evidence on the impact of factors such as system design and usability on the effectiveness of the CDSSs.