Most decision-support systems in medicine have been developed in hospital environments, but only few are designed for being used by general practitioners. The present work aims to design an expert system for practitioners in chronic heart failure (CHF) treatment. It provides assistance in defining the therapy relying on CHF aetiology, gravity, physiopathological conditions, and discriminates if other coexistent diseases and/or drugs taken by the patient could interact with CHF management. It warns the physician about the possible interactions of the considered CHF therapy. In case of contraindications, the system suggests another alternative therapy. It also advices about the control tests to follow-up the prescribed therapy, and about the indicated hygienic-dietetic suggestions. To assess its internal consistency, we examined the behaviour of the system with 20 CHF patients, by comparing the suggested therapy with the prescriptions of cardiologists. In 9 cases the suggested therapeutic schemes contained all the "n" drugs administered by the cardiologists. In 5 cases the concordance was on at least two thirds of the prescribed medications, in 5 between one half and two thirds, while in 1 case there was no concordance at all. In none of the 10 cases with partial concordance, were there major discrepancies (i.e. potentially deleterious for the patient) between the expert system's suggestions and the cardiologists' prescriptions. In conclusion, the advices of the expert system were similar to those of the cardiologists, suggesting the feasibility of such a computer support to CHF management.

Treatment of chronic heart failure: an expert system advisor for general practitioners.

PERLINI, STEFANO;BERNARDI, LUCIANO
1990-01-01

Abstract

Most decision-support systems in medicine have been developed in hospital environments, but only few are designed for being used by general practitioners. The present work aims to design an expert system for practitioners in chronic heart failure (CHF) treatment. It provides assistance in defining the therapy relying on CHF aetiology, gravity, physiopathological conditions, and discriminates if other coexistent diseases and/or drugs taken by the patient could interact with CHF management. It warns the physician about the possible interactions of the considered CHF therapy. In case of contraindications, the system suggests another alternative therapy. It also advices about the control tests to follow-up the prescribed therapy, and about the indicated hygienic-dietetic suggestions. To assess its internal consistency, we examined the behaviour of the system with 20 CHF patients, by comparing the suggested therapy with the prescriptions of cardiologists. In 9 cases the suggested therapeutic schemes contained all the "n" drugs administered by the cardiologists. In 5 cases the concordance was on at least two thirds of the prescribed medications, in 5 between one half and two thirds, while in 1 case there was no concordance at all. In none of the 10 cases with partial concordance, were there major discrepancies (i.e. potentially deleterious for the patient) between the expert system's suggestions and the cardiologists' prescriptions. In conclusion, the advices of the expert system were similar to those of the cardiologists, suggesting the feasibility of such a computer support to CHF management.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11571/112232
Citazioni
  • ???jsp.display-item.citation.pmc??? 1
  • Scopus ND
  • ???jsp.display-item.citation.isi??? 7
social impact