Our study suggests that measures of concordance should be revised to incorporate items that measure

the doctor’s contribution in making the decision as well as in encouraging the patient to be involved in the decision. The adapted scale, with good inter-item reliability, could be used as a concordance measure in HIV clinics. The study had limitations. First, only patients’ perspectives and characteristics were measured and there was no information on the individual doctors [35] or from independent observers. Secondly, the study did not aim to determine causality. It is therefore possible that patients in better health perceived their communication with doctors as more concordant. One study found that patients with less intense symptoms were more satisfied with their care [36], although this finding was not replicated in a later study Quizartinib mouse [37]. However, research has shown C59 wnt manufacturer that patients with better self-rated health were more likely to be consumerist [38] and thus likely to have higher expectations of medical care, which should lead to perceiving doctors as less concordant. Research using intervention trials has shown that increased patient involvement in the medical

consultation results in better health outcomes in patients with ulcers and diabetes [39,40]. Our study demonstrated that overall concordance was related to CD4 cell count 6–12 months post-study after the baseline CD4 cell count was controlled for, suggesting a potential causal link between concordance and health outcomes. Further

research is needed to determine causality and to investigate possible mechanisms such as greater adherence, greater perceived control over illness and reduced anxiety/depression. Thirdly, our Sitaxentan limited sample size and restricted geographical study locations make it difficult to generalize from our findings. White homosexual men who were university educated and born in the United Kingdom were more likely to complete the Concordance Scale. However, no relationship was found between these demographic factors and concordance. Differences between completers and noncompleters were also found in terms of CD4 cell count and VL, but these disappeared once we controlled for stoppers being less likely to complete the scale. Moreover, symptom, adherence and quality of life variables did not differ between completers and noncompleters. It should also be noted that the five participating clinics account for a large proportion of UK patients, but may not necessarily be representative of all NHS providers of HIV care in the United Kingdom, nor reflect all clinician styles. This study supports the importance of patients’ reports of concordance in terms of health and health-related outcomes within HIV care. Further research is needed to establish causality by conducting intervention studies.