Contextual decision making and the implementation of clinical guidelines: an example from mental health

Clinical decision making plays a crucial role in the transformation of science to service. Treatment decisions typically are evaluated by comparing them against norms, such as practice guidelines. An adherence standard has been criticized as inappropriate, but no measurable alternative has been proposed to date. This study develops a new standard of incorporation and a companion matching test, and addresses two questions: (1) Do clinicians incorporate a treatment guideline even when they do not endorse it? (2) If so, do they incorporate the guideline consistently?

The study uses the clinical paradigm of treatment-resistant schizophrenia and a published guideline developed at the Yale University Department of Psychiatry that has been soundly rejected in clinical practice. A vignette study was developed, using a four-factor, fully crossed and within-subject design, then administered to 21 volunteer paid psychiatry residents.

The endorsement pattern showed a low concurrence rate and significant apparent inconsistency within subjects. However, the matching test showed a clear relationship between endorsement of the guideline and features of individual vignettes. The matching test demonstrated significant within-subject consistency and accounted for 65% of the endorsement variance.

Implications are preliminary, given limitations that pertain to the subject population and use of vignettes, the clinical paradigm, and treatment guideline. However, the study's concepts, procedures, and findings may play a valuable role in future transformative initiatives, including training clinicians in the use of clinical guidelines and evaluating the appropriateness of guidelines before their implementation.