Application of financial analysis techniques to clinical laboratory data: a novel method of trend interpretation in the intensive care unit
Stawicki, S. Peter (2007) Application of financial analysis techniques to clinical laboratory data: a novel method of trend interpretation in the intensive care unit. OPUS 12 Scientist, 1 (2). pp. 1-4. ISSN 1940-8633
Full text available as:
Modern critical care medicine depends on constant flow of massive amounts of information. This information has limited usefulness unless it is appropriately gathered, stored, displayed, and interpreted. Despite such wealth of data in the modern intensive care units (ICU), intensivists often rely on very fragmentary ‘snapshot’ information to make important clinical decisions. In an attempt to improve the understanding of clinical data trends, application of financial analysis (FA) methods to clinical laboratory data samples was performed. Three randomly chosen, anonymized laboratory datasets of patients who spent at least 30 days in the ICU were retrospectively examined. Regularly obtained laboratory values were retrieved and recorded for each patient. Variables examined included white blood cell count, hemoglobin level, platelet count, and blood glucose levels. These variables were then entered into specialized FA software and subjected to computer-based processing. Trends in the recorded data were examined using (1) the Stochastic Oscillator (SO), (2) the Relative Strength Index (RSI) tool, (3) Price Envelope (PE) analysis and (4) Moving Average analysis. All clinical laboratory parameter analyses demonstrated that laboratory data could be successfully ‘trended’ using FA techniques. Not only was the laboratory data clearly readable and transparent when displayed in FA fashion, some trends that were not apparent on ‘gross’ inspection of the numeric data became very apparent after FA. Much like with financial patterns and vital sign data, trends noted within laboratory parameters appeared to be more significant when more than one indicator identified or ‘confirmed’ them, utilizing the concept of a confirmatory variable. Laboratory data, much like financial and vital sign data, were subject to trend reversals. Such reversals in laboratory parameters appeared to follow patterns similar to those followed by financial vehicles and markets. This report demonstrates that laboratory data can be subjected to the same manipulations as financial market data. Furthermore, FA tools appear to provide the interpreter with means to define, confirm, and possibly predict trends and trend reversals. Assumptions for use of FA methods in biologic parameter analysis are also presented.
Archive Staff Only: edit this record