Abstract
OBJECTIVE: To test the effect of diagnosis on training an artificial neural network (ANN) to predict length of stay (LOS) for psychiatric patients involuntarily admitted to a state hospital. DESIGN: A series of ANNs were trained representing schizophrenia, affective disorders, and diagnosis-related group (DRG) 430. In addition to diagnosis, variables used in training included demographics, severity of illness, and others identified to be significant in predicting LOS. RESULTS: Depending on diagnosis, ANN-predictions compared with actual LOS indicated accuracy rates ranging from 35% to 70%. The validity of ANN predictions was determined by comparing LOS estimates with the treatment team's predictions at 72 hours following admission, with the ANN predicting as well as or better than did the treatment team in all cases. CONCLUSIONS: One problem in traditional approaches to predicting LOS is the inability of a derived predictive model to maintain accuracy in other independently derived samples. The ANN reported here was capable of maintaining the same predictive efficiency in an independently derived cross-validation sample. The results of ANNs in a cross-validation sample are discussed and the application of this tool in augmenting clinical decision is presented.
Full Text
The Full Text of this article is available as a PDF (854.4 KB).
Selected References
These references are in PubMed. This may not be the complete list of references from this article.
- Choca J. P., Peterson C. A., Shanley L. A., Richards H., Mangoubi E. Problems in using statistical models to predict psychiatric length of stay: an illustration. Hosp Community Psychiatry. 1988 Feb;39(2):195–197. doi: 10.1176/ps.39.2.195. [DOI] [PubMed] [Google Scholar]
- Davis G. E., Lowell W. E., Davis G. L. A neural network that predicts psychiatric length of stay. MD Comput. 1993 Mar-Apr;10(2):87–92. [PubMed] [Google Scholar]
- Dorwart R. A., Schlesinger M., Davidson H., Epstein S., Hoover C. A national study of psychiatric hospital care. Am J Psychiatry. 1991 Feb;148(2):204–210. doi: 10.1176/ajp.148.2.204. [DOI] [PubMed] [Google Scholar]
- Dorwart R. A., Schlesinger M. Privatization of psychiatric services. Am J Psychiatry. 1988 May;145(5):543–553. doi: 10.1176/ajp.145.5.543. [DOI] [PubMed] [Google Scholar]
- Mulsant B. H. A neural network as an approach to clinical diagnosis. MD Comput. 1990 Jan-Feb;7(1):25–36. [PubMed] [Google Scholar]
- Stoskopf C., Horn S. D. The Computerized Psychiatric Severity Index as a predictor of inpatient length of stay for psychoses. Med Care. 1991 Mar;29(3):179–195. doi: 10.1097/00005650-199103000-00001. [DOI] [PubMed] [Google Scholar]
- Taube C., Lee E. S., Forthofer R. N. DRGs in psychiatry. An empirical evaluation. Med Care. 1984 Jul;22(7):597–610. doi: 10.1097/00005650-198407000-00002. [DOI] [PubMed] [Google Scholar]
- Taube C., Lee E. S., Forthofer R. N. Diagnosis-related groups for mental disorders, alcoholism, and drug abuse: evaluation and alternatives. Hosp Community Psychiatry. 1984 May;35(5):452–455. doi: 10.1176/ps.35.5.452. [DOI] [PubMed] [Google Scholar]
- Tu J. V., Guerriere M. R. Use of a neural network as a predictive instrument for length of stay in the intensive care unit following cardiac surgery. Comput Biomed Res. 1993 Jun;26(3):220–229. doi: 10.1006/cbmr.1993.1015. [DOI] [PubMed] [Google Scholar]