OBJECTIVE: The University of Colorado developed and piloted a team-based primary care delivery model called ambulatory process excellence (APEX) in a family medicine residency in 2015. We evaluated its impact on depression screening using found data and tools readily available to practice-based evaluators. METHOD: The APEX model calls for 5 medical assistants (MAs) supporting 2 providers. MAs have dedicated time to provide delegated care, including depression screening with 2 versions of the Patient Health Questionnaire: PHQ-2 and PHQ-9. Using an interrupted time-series-with-control design, we created longitudinal profiles of the pilot and a control practice using statistical process control charts. We obtained data from preexisting dashboards derived from the electronic medical record. Outcomes included PHQ-2 screening rates, patients screening positive, and the proportion of them completing a PHQ-9. Covariates included monthly visits and new-patient appointments. Using Microsoft Excel, we transformed all data into modified z scores, plotted them on a multivariate control chart for each practice, and assessed them for evidence of special cause variation. Key informants provided information about potentially confounding concurrent events. RESULTS: Compared with baseline, the intervention practice significantly increased primary care medical visits and new-patient appointments, increased positive PHQ-2 patients, and improved PHQ-9 completion. High screening rates remained stable. In the control practice, new-patient appointments increased and PHQ2 screening improved. DISCUSSION: APEX may contribute to better depression-screening processes. We have provided a detailed description of a real-world, practice-based, quasi-experimental evaluation model using common spreadsheet software (Microsoft Excel) to transform and analyze found data with multivariate statistical process-control charts. (PsycINFO Database Record (c) 2017 APA, all rights reserved).
OBJECTIVE: The University of Colorado developed and piloted a team-based primary care delivery model called ambulatory process excellence (APEX) in a family medicine residency in 2015. We evaluated its impact on depression screening using found data and tools readily available to practice-based evaluators. METHOD: The APEX model calls for 5 medical assistants (MAs) supporting 2 providers. MAs have dedicated time to provide delegated care, including depression screening with 2 versions of the Patient Health Questionnaire: PHQ-2 and PHQ-9. Using an interrupted time-series-with-control design, we created longitudinal profiles of the pilot and a control practice using statistical process control charts. We obtained data from preexisting dashboards derived from the electronic medical record. Outcomes included PHQ-2 screening rates, patients screening positive, and the proportion of them completing a PHQ-9. Covariates included monthly visits and new-patient appointments. Using Microsoft Excel, we transformed all data into modified z scores, plotted them on a multivariate control chart for each practice, and assessed them for evidence of special cause variation. Key informants provided information about potentially confounding concurrent events. RESULTS: Compared with baseline, the intervention practice significantly increased primary care medical visits and new-patient appointments, increased positive PHQ-2 patients, and improved PHQ-9 completion. High screening rates remained stable. In the control practice, new-patient appointments increased and PHQ2 screening improved. DISCUSSION: APEX may contribute to better depression-screening processes. We have provided a detailed description of a real-world, practice-based, quasi-experimental evaluation model using common spreadsheet software (Microsoft Excel) to transform and analyze found data with multivariate statistical process-control charts. (PsycINFO Database Record (c) 2017 APA, all rights reserved).
Authors: Bethany M Kwan; Mika K Hamer; Austin Bailey; Kathy Cebuhar; Colleen Conry; Peter C Smith Journal: J Gen Intern Med Date: 2022-01-07 Impact factor: 6.473