Does provider adherence to a treatment guideline change clinical outcomes for patients with bipolar disorder? Results from the Texas Medication Algorithm Project
Background. Despite increasing adoption of clinical practice guidelines in psychiatry, there is little measurement of provider implementation of these recommendations, and the resulting impact on clinical outcomes. The current study describes one effort to measure these relationships in a cohort of public sector out-patients with bipolar disorder.
Method. Participants were enrolled in the algorithm intervention of the Texas Medication Algorithm Project (TMAP). Study methods and the adherence scoring algorithm have been described elsewhere. The current paper addresses the relationships between patient characteristics, provider experience with the algorithm, provider adherence, and clinical outcomes. Measurement of provider adherence includes evaluation of visit frequency, medication choice and dosing, and response to patient symptoms. An exploratory composite ‘adherence by visit’ score was developed for these analyses.
Results. A total of 1948 visits from 141 subjects were evaluated, and utilized a two-stage declining effects model. Providers with more experience using the algorithm tended to adhere less to treatment recommendations. Few patient factors significantly impacted provider adherence. Increased adherence to algorithm recommendations was associated with larger decreases in overall psychiatric symptoms and depressive symptoms over time, but did not impact either immediate or long-term reductions in manic symptoms.
Conclusions. Greater provider adherence to treatment guideline recommendations was associated with greater reductions in depressive symptoms and overall psychiatric symptoms over time. Additional research is needed to refine measurement and to further clarify these relationships.(Published Online September 29 2005)
c1 Department of Psychological Sciences, Purdue University, 703 Third St., West Lafayette, IN 47907-2081, USA. (Email: email@example.com)