CHARACTERISTICS OF STAFF-PATIENT INTERACTIONS ON A BEHAVIORAL TREATMENT UNIT FOR PATIENTS WITH SERIOUS MENTAL ILLNESS, AND THE EFFECTS OF A BEHAVIOR MANAGEMENT TRAINING PROGRAM ON STAFF PERFORMANCE

Despite the existence of effective psychological interventions for people diagnosed with schizophrenia, these continue to be underutilized. This problem is most critical in the case of so called “treatment-refractory” patients, where, paradoxically, effective interventions with among the strongest evidence base in psychiatry exist – namely – inpatient social-learning-based rehabilitation programs. Barriers to implementation and effective maintenance of such programs include a low frequency of staff-patient interactions and a relatively high frequency of non-therapeutic staff behaviors, lack of staff training and background in behavioral treatment, negative attitudes towards behavioral interventions, and tension between professional disciplines. This study examined the nature and frequency of staff-patient interactions on an inpatient behavioral psychiatric treatment unit for individuals who suffer from severe and persistent mental illness. A second aim of this study was to determine the effects on staff performance of a mandatory training program in behavior management. Effects were examined pre-post, as well as by professional discipline. At baseline, clinical staff displayed significantly more therapeutic behaviors and significantly fewer negative behaviors compared to medical staff when interacting with patients. This pattern of differences between the two groups persisted to post-training. Unexpectedly, clinical staff displayed fewer instances of instructing and staff to staff conversations compared to medical staff. From pre- to post-training, all staff demonstrated increased positive and therapeutic behaviors and decreased instances of negative behaviors when interacting with patients. Performance on both written and in-vivo staff assessment at post-training was significantly correlated with improved staff-patient interactions on the unit after training, most notably on measures of negative staff-patient interaction. Although both written and in-vivo test scores significantly predicted change in negative staff behaviors, the in-vivo test performance increased predictive ability over and above that of written test performance. Staff who disagreed with social learning and behavioral management principles displayed less improvement in negative behaviors from pre- to post-assessment compared to other staff. These data have implications for clarifying staff training needs in programs for chronically ill people with serious mental illness. Dissertation advisor: William D. Spauldin