To estimate long-term costs and outcomes attributable to a physician–pharmacist collaborative intervention compared with physician management alone for treating essential hypertension. A Markov model cohort simulation with a 6-month cycle length to predict acute coronary syndrome, stroke, and heart failure throughout lifetime was performed. A cohort of 399 patients was obtained from two prospective, cluster randomized controlled clinical trials implementing physician–pharmacist collaborative interventions in community-based medical offices in the Midwest, USA. Framingham risk equations and other algorithms were used to predict the vascular diseases. SBP reduction due to the interventions deteriorated until 5 years. Direct medical costs using a payer perspective were adjusted to 2015 dollar value, and the main outcome was quality-adjusted life years (QALYs); both were discounted at 3%. The intervention costs were estimated from the trials, whereas the remaining parameters were from published studies. A series of sensitivity analyses including changing patient risks of vascular diseases, probabilistic sensitivity analysis, and a cost-effectiveness acceptability curve were performed. The lifetime incremental costs were $26 807.83 per QALY (QALYs gained = 0.14). The intervention provided the greatest benefit for the high-risk patients, moderate benefit for the trial patients, and the lowest benefit for the low-risk patients. If a payer is willing to pay $50 000 per QALY gained, in 48.6% of the time the intervention would be cost-effective. Team-based care such as a physician–pharmacist collaboration appears to be a cost-effective strategy for treating hypertension. The intervention is most cost-effective for high-risk patients.

Objectives: Medication monitoring is important for safe and effective medication use; however, no attitudinal measure exists for a health care provider's medication monitoring attitude. The objectives of this study were to (1) create a measure of a community pharmacist medication monitoring attitude; (2) test concurrent validity using a validated measure of medication monitoring behaviours; and (3) report community pharmacist attitudes towards medication monitoring.Methods: A mixed methods item development process was employed to generate Likert-type items from qualitative interviews. Following item review and piloting, a four-contact survey, including 20 6-point Likert-type items and the four-item Behavioral Pharmaceutical Care Scale monitoring domain, was mailed to 599 randomly sampled US community pharmacists from the state of Iowa. Exploratory factor analysis, Pearson's correlation and descriptive statistics were used to address study objectives.Key findings: There were 254 (42.4%) returned and usable surveys. Factor analysis yielded two domains, a seven-item, positively worded internal (α = 0.819) and an eight-item, negatively worded external domain (α = 0.811). Both domains were positively correlated with the monitoring domain of the Behavioral Pharmaceutical Care Scale supporting convergent validity. Overall, respondents had a positive internal monitoring attitude with a mean of 4.62 (0.68), although many practiced in an environment less conducive to monitoring, as evident by a mean of 3.13 (0.88). Pharmacists were more oriented towards medication side effect and effectiveness monitoring than nonadherence monitoring.Conclusions: The mixed methods item development process created a reliable and valid measure of a pharmacist's medication monitoring attitude. While pharmacists had an overall positive medication monitoring attitude, improvements are needed to bolster adherence monitoring and make pharmacy environments more conducive to monitoring.