Document Type


Date of Degree

Fall 2013

Degree Name

PhD (Doctor of Philosophy)

Degree In


First Advisor

Doucette, William R

First Committee Member

Sorofman, Bernard

Second Committee Member

Ernst, Erika

Third Committee Member

Wesely, Pamela

Fourth Committee Member

Goedken, Amber


Medication related problems occur frequently and can be difficult to predict. Medication monitoring by health care providers allows for problems to be identified before they become severe. Medication monitoring is an an emerging role for community pharmacists. This mixed methods study examines pharmacist perspectives on medication monitoring. Study methods included 12 semi-structured interviews, a mixed methods item generation process to create a pharmacist medication motioning attitude measure, and mailed factorial survey designed to assess pharmacist decision making related to asking non-adherence, side effect, and effectiveness questions for randomly generated refill dispensing vignettes. Hierarchical linear modeling was used to identify significant vignette level and pharmacist level variables associated with likelihood to ask the three monitoring questions. The qualitative analysis showed barriers to medication monitoring including time limitations, busyness, low patient expectation, and a perceived routine nature of refills by both patients and pharmacists. Monitoring non-adherence was a challenge because workflows often do not make non-adherence apparent to the pharmacists when the patient presents to the pharmacy. Lastly, monitoring interactions often are precipitated by "gateway conversations" which begin as technical or cost issues related to the prescription but then progress to monitoring issues when the patient reflects interest. Analysis of the survey showed in general, pharmacists had positive medication monitoring attitudes and worked in pharmacies somewhat conducive to medication monitoring, although there was variation. The factorial survey showed pharmacist monitoring attitudes were significantly associated with the likelihood to ask all three monitoring question types. For the different prescriptions involved, warfarin and hydrocodone were significantly associated with asking monitoring questions whereas fluoxetine and metoprolol appeared less question-worthy. The number of days late was associated with greater question asking likelihood suggesting days late is an important activator for pharmacist medication monitoring. Number of patients waiting was a barrier. This study shows community pharmacists are oriented to monitoring, but there are significant barriers which need to be addressed when advancing this role.


Communication, Community, Factorial Survey, Medication, Monitoring, Pharmacy


ix, 207 pages


Includes bibliographical references (pages 190-207).


Copyright 2013 Matthew John Witry