This study proposes to test the feasibility of using severity and quality of care indices for cross site comparisons of outcome for psychiatric emergencies treated in the ED. The study will be conducted at two test hospitals in Rhode Island, one of which has an inpatient psychiatric unit. As part of the study, we propose to define psychiatric emergencies as those patients with a psychiatric complaint and/or who receive psychotropic drugs, counseling, and/or a psychiatric discharge from the ED. An advisory panel of physicians from New England and with expertise in the area of psychiatric emergencies will be convened to 1) finalize the definitions prior to the beginning of data collection and 2) review the preliminary results after data collection is completed. In addition, we propose to test the feasibility of using mortality, follow up compliance, and unscheduled ED visits within 15 days of the initial visit to measure patient outcome. We will identify the incidence of psychiatric emergencies and determine the degree of variance existing in outcomes at each site. Two nurses with experience related to psychiatric emergencies in the ED will be hired half-time to abstract patient records. Data quality will be enhanced by abstracting records within twenty four hours of admission. We expect to collect about 2500 abstracts, 1500 at one hospital and 1000 at the other. The nurse abstactors and the physicians will validate from copies of the emergency department records the reliability and validity of the case selection and abstracting at both sites. The various indices will be correlated with the expert opinion of the physicians to evaluate the sensitivity of the indices. Logistic regression and other multivariate analysis will be used to determine if the independent and dependent variables reflect impact of EMS effectiveness on psychiatric emergencies. It is our intent to use the proposed measures to replicate this study New England wide.
| Gustafson, D H; Sainfort, F; Johnson, S W et al. (1993) Measuring quality of care in psychiatric emergencies: construction and evaluation of a Bayesian index. Health Serv Res 28:131-58 |
| Sateia, M J; Gustafson, D H; Johnson, S W (1990) Quality assurance for psychiatric emergencies. An analysis of assessment and feedback methodologies. Psychiatr Clin North Am 13:35-48 |
