European journal of operational research vol:105 issue:2 pages:332-345
Psychiatric patients and psychiatric care are more difficult to measure than acute care patients. The classification schemes are always multi-variate and the measurement highly subjective. Wide ranges of problem severity and intensity of care prevail. Highly skewed distributions are common. Once the data are in, the issue becomes how to use that data to manage the care of psychiatric patients. This paper investigates and illustrates how both problems can be dealt with. The paper first concentrates on the measurement problems and indicates how good scales can be differentiated from bad scales. Reliable scales allow in a second step to trace precisely the evolution of the patient during his care process. This provides an opportunity to manage the care programme of those patients more accurately than possible so far. The case-study concerns a group of 113 patients in one psychiatric ward of a general hospital. Depression is the patient problem of choice. The methodology involves multi-variate data analysis and multi-dimensional scaling approaches. The approach is highly visual. This improves access to information, enables insight and shortens the feedback loop. These foster the exchange of information and facilitate communication about the patients within the therapeutic team. Unfortunately, understanding and experience is necessary. This work contributed to the start of a national Psychiatric Minimum Basic Data Set in Belgium as of July 1, 1996.