Despite increases in the reliability of psychiatric diagnosis resulting from enhanced specificity of diagnostic criteria and the use of standardized clinical interviews, substantial clinical judgement is needed to determine if a patient's symptoms on each of the diagnostic criteria for a disorder are sufficiently severe to be above threshold. Although anchoring, halo, and expectancy (AHE) effects have been documented in many judgment tasks, the impact of these information processing biases on psychiatric diagnosis has not been investigated. AHE effects are most likely to affect the diagnosis in cases where the severity of most criterional symptoms is near threshold, but early or salient information is relatively extreme. Study 1 varies the degree of context in which symptom judgments are made by psychiatrists, clinical psychologists, and trainees. Severity ratings made on the diagnostic criteria for major depressive episode (MDE) using verbatim transcripts of relevant sections of a structured diagnostic interview (the SCID) previously administered to 10 patients. Severity ratings made under high context are predicted to be more highly correlated than ratings made under low context (an AHE effect). In Study 2, using videotapes of actors simulating portions of the SCID interview for two patients, AHE effects are teased apart in a factorial experiment. The halo manipulation is the presence of depressed vs. nondepressed nonverbal communication in the simulated patient (SP) created by an actor. The anchoring manipulation is the order in which information is presented to the subject. The expectancy manipulation is the presence vs. absence of a previous MDE. AHE effects are predicted to influence judgments of symptom severity and decisions of symptom presence or absence for each of the non-manipulated diagnostic criteria for MDE, as well as final diagnosis and treatment recommendations. These effects are expected to be more pronounced in more highly trained clinicians. Finally, the influence of exposure to more information on these effects is investigated via comparison to a second SP in which additional data for each of the criteria are shown using supplementary questions. The results of the latter manipulation as well as training in symptom severity judgement calibration for trainees in Study 1 should suggest directions for future research aimed at increasing the accuracy of clinical judgment in psychiatric diagnosis.
| Mumma, Gregory H (2002) Effects of three types of potentially biasing information on symptom severity judgments for major depressive episode. J Clin Psychol 58:1327-45 |
| Mumma, G H; Wilson, S B (1995) Procedural debiasing of primacy/anchoring effects in clinical-like judgments. J Clin Psychol 51:841-53 |
