5, range 5.5–10). Seven nursing staff reported spending less than 10 min to visit a patient and conduct the INR test. Nurses most commonly reported spending a further 2–5 min entering the result into the MedePOC system (four nurses), with a three nurses spending between 5 and 10 min. Four nurses indicated that GPs generally responded within 24 h of receiving the INR test; two nurses indicated that responses typically came within 24–48 h. The patients who responded to

the evaluation questionnaire were all satisfied with their ACF’s involvement in the study (median score 9, range 5.5–9.5). Most patients indicated they would prefer a finger-prick blood test with the portable INR monitor to the usual venous blood sampling Panobinostat in vitro taken by the pathology service (median score 8.5, range 0.5–9.5), and that they would prefer their INR to be

monitored at the nursing home rather than through pathology testing (median score 9.3, range 0.5–9.5). All patients felt that their warfarin was better controlled during the study (median score 9, range 6.5–9.5). This proof-of-concept study was designed to test the utility of ACF-based INR testing with electronic communication to GPs. Weekly POC INR monitoring conducted in ACFs with electronic communication of results to GPs and warfarin doses back to the ACFs resulted in non-significant improvements in INR control. An improvement in warfarin control was shown for the majority of patients, Romidepsin and POC monitoring was well received by nursing staff, suggesting that further research is warranted to investigate whether this strategy can improve INR control in this population. Several factors may limit the generalisability of this study. We included a selected group of participants

who may not be representative of the broader population of older people taking warfarin. It is possible that they were more motivated and more adherent GPX6 with their therapy than other patients.[23] Indeed, the intervention may be more successful in people with poorer prior INR control, as the participants had a reasonable level of control prior to the intervention. Other potential limitations include the small sample size, non-randomised design and relatively short duration of the intervention. Certainly, further investigation of the system in patients taking warfarin is warranted, including an evaluation of the costs involved and effectiveness of the approach in a larger cohort. While we have reported on the implementation of the system in a small number of ACFs, a number of challenges would need to be addressed if it was to become more widely used, including ongoing education of nursing staff, quality assurance of the testing procedure and auditing of the communication process to ensure appropriate response to INR test results.