Predictors of Psychiatric Comorbidity in Medical Outpatients

Abstract

OBJECTIVE: Psychiatric comorbidity in medical outpatients is associated with personal suffering and reduced psychosocial functioning. Simple clinical indicators are needed to improve recognition and treatment of psychiatric comorbidity. This study aimed to identify predictors of psychiatric comorbidity for diagnostic use in busy medical settings and to describe their criterion validity. METHODS: The SCID was adopted as the independent criterion standard for the presence of a psychiatric comorbidity in 357 patients (68% female; mean age, 43 years) of six internal medicine outpatient clinics and 12 general practices. Potential indicators of psychiatric comorbidity were investigated by means of patient and physician questionnaires. Logistic regression analyses were used to identify independent predictors of psychiatric comorbidity, and their operating characteristics were determined. RESULTS: Of 18 indicators, the four most important predictors of psychiatric comorbidity were identified: a screening question for nervousness, anxiety, or worries (odds ratio, 11.9; p <.001), a screening question for depressed mood (odds ratio, 8.8; p <.001), the self-report of three or more bothersome physical symptoms (odds ratio, 3.2; p =.001), and feeling distressed by partner difficulties (odds ratio, 2.7; p =.006). The combined assessment of the four predictors resulted in positive predictive values as high as 100%, negative predictive values as high as 91%, sensitivities as high as 86%, and specificities as high as 100%. CONCLUSIONS: The identification of mental disorders in medical outpatients could be substantially improved by the knowledge and use of four easily accessible predictors. When the presence of one or more of these predictors can be confirmed, it is suggested that the patient undergo further evaluation to determine more precisely the presence and specific type of psychiatric disorder being identified.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 59 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.