Accuracy of oncologist assessments of psychiatric problems in cancer inpatients

Abstract

ABSTRACTObjective:Our objective was to examine the accuracy of non-psychiatrist assessments of psychiatric problems in cancer patients.Method:We conducted a retrospective chart review of cancer patients who were admitted and referred to the consultation–liaison (C–L) team between January of 2011 and December of 2012. The agreement between non-psychiatrist assessments and final diagnoses by attending C–L psychiatrists was estimated for every category of referral assessment using codes from the International Classification of Mental and Behavioral Disorders (10th revision). The data were obtained from the consultation records of 240 cancer inpatients who were referred to the C–L service at a tertiary care center in Tokyo.Results:The agreement ratio between referring oncologists and psychiatrists differed according to the evaluation categories. The degrees of agreement for the categories of “delirious,” “depressive,” “dyssomnia,” “anxious,” “demented,” “psychotic,” and “other” were 0.87, 0.43, 0.51, 0.50, 0.27, 0.55, and 0.57, respectively. The agreement for all patients was 0.65. Significant differences were observed among seven categories (chi-squared value = 42.454 at p < 0.001 and df = 6). The analysis of means for proportions showed that the degree of agreement for the “delirious” category was significantly higher and that that for the “depressive” category was lower than that for all patients, while for the “demented” category it was close to the lower decision limit but barely significant. One half of the 20 cases who were referred as depressive were diagnosed with delirium, with one quarter of those having continuously impaired consciousness. Some 7 of the 11 cases who were referred as demented were diagnosed as having delirium.Significance of Results:The accuracy of non-psychiatrist assessments for psychiatric problems in cancer patients differs by presumed diagnosis. Oncologists should consider unrecognized delirium in cancer inpatients who appear depressed or demented.