Health Evaluation through Logical Programming (HELP)
Health Evaluation through Logical Processing (HELP) was the first hospital information system to integrate clinical data accumulation and clincal decision support. The real addition of this system was using the computer to aid decision-making, such as recognition of cardiac tamponade, by analysis of patient results. One trial suggested the the program had a 94% success rate of choosing an appropriate antibiotic regimen compared to a 77% success rate for physicians [Desouza KC 2002]. It also supports selection of appropriate antibiotics in case of infectious diseases. Clinical decision support also predicts and prevents many adverse drug events (ADE's).
Contents
History
HELP was designed by Gardner, Pryor, and Warner [Gardner RM 1999]. The original system was developed at the LDS Hospital in Salt Lake City, Utah, and has been active since 1967. The parent organization is now known at Intermountain Health Care (IHC). The original focus of the system was better understanding of how circulation is controlled. It was used in the cardiac cath lab, and then extended into the operating room and ICU. Eventually it became a hospital wide system, and then at 9 system hospitals.
HELP was developed at the University of Utah is enhanced with HCOM, a compiler that allowed Bayesian decision support modules to be developed.
Originally, physicians were only "data reviewers," not "data enterers." Also, the system did not allow exchange of records between facilities. That changed with the introduction of the Longitudinal patient Data Repository (LDR) starting in the late 1990s.
Present Status
HELP is a knowledge-based hospital information system (HIS) [3]. In addition to the HIS routine work, HELP has a robust decision-making function which is fully incorporated into the system. The decision support function can provide alerts/reminders, data interpretation, patient diagnosis and management suggestions, etc.
Evaluation of HELP System
HELP has been evaluated in several ways. Some of the examples of these evaluations which demonstrate the applicability of this system in different contexts, follow:
User Acceptance
User questionnaires in the early 1990's from 246 physicians and 374 nurses revealed several findings that encouraged development of medical expert systems contradicting concern that these types of systems would not be appreciated or utilized [1]. These findings were:
- There was no correlation between age, specialty or general computer experience of the individuals with their attitudes about the HELP system.
- Users did not feel that the system would cause a "big brother" concern of external monitoring or sanctions.
- Computer decision support did not seem to diminish their decision making authority.
Infectious Diseases
Over a course of 13 years, this feature of the HELP, initially only a prophylactic antibiotic reminder system, evolved into a very sophisticated 'antibiotic assistant' in 1998. This has resulted in a significant improvement in antibiotic administration costs, adverse drug reactions, and total patient care.
Management of Adverse Drug Reactions
By using the HELP system, adverse drug reactions were detected more than 80 times more, compared to manual methods. The system can help detect these events quickly and prevent the effects, which in turn, results in reduced length of hospitalization and considerable cost savings.
References
- Gardner RM, Pryor TA, Warner HR. The HELP hospital information system: update 1998. International Journal of Medical Informatics 54(1999)169-182.
- Desouza KC. Managing Knowledge with Artificial Intelligence. Westport, CT. 2002. Greenwood Publishing Group. P. 84. 17 Jan 2009.
- http://www.openclinical.org/aisp_help.html
