A neural network study of the correlation between metabolic-cardiovasculardiseases and disability in elderly people

Numerous studies have affirmed the existence of a correlation between vario us cardiovascular diseases and functional decline in elderly people. Not mu ch information, however, is available concerning the overall effect of vari ous, possibly coexisting, cardiovascular pathologies, or metabolic conditio ns notoriously related to them, on determining disability. We wanted to ver ify if it were possible to assess: (1) The overall importance of various me tabolic and cardiovascular diseases which elderly people often suffer from contemporaneously in determining a condition of not self-sufficiency; (2) T he possibility of predicting a condition of not self-sufficiency in relatio n to the above-mentioned pathologies. In order to achieve this aim, we used an artificial neural network: a statistical-mathematical tool able to dete rmine the existence of a correlation between series of data and, once 'trai ned', to predict output data given the input data. Although artificial neur al networks have been applied in various areas of medical research, they ha ve not been previously applied in geriatrics. We have applied this method t o a sample of 179 elderly people, demonstrating that seven clinical-biologi cal variables concerning their metabolic and cardiovascular conditions are strictly related, all together, to the presence or otherwise of a functiona l impairment. When tested on a sample of 20 'unknown' elderly people, the t rained network gave the correct answer - self-sufficiency or not self-suffi ciency - in 95% of the cases. Despite the fact that the sample studied was relatively small, artificial neural networks ale undoubtedly useful in pred icting functional impairment in elderly people in relation to the presence of metabolic and cardiovascular diseases. (C) 2000 Elsevier Science ireland Ltd. All rights reserved.