Critical care computing: outcomes, confidentiality, and appropriate use

Area of Science:

• Medical Informatics
• Health Services Research
• Bioethics

Background:

• Computerized outcome predictor models and severity scoring systems are increasingly used in healthcare.
• Existing models have limitations in accurately predicting patient outcomes and guiding resource allocation.
• Societal attitudes towards death and resource scarcity pose challenges to healthcare systems.

Purpose of the Study:

• To review the limitations of current computerized outcome predictor models and severity scoring systems.
• To propose a "computational futility metric" as a logical extension of predictor models.
• To discuss the potential uses, abuses, and ethical implications of electronic surveillance in healthcare.

Main Methods:

• Literature review of existing predictor models and scoring systems.
• Conceptual development and discussion of a "computational futility metric."
• Analysis of ethical considerations and societal impacts of electronic epidemiologic surveillance.

Main Results:

• Current predictor models and scoring systems have significant limitations.
• A "computational futility metric" offers a potential new approach but requires careful consideration of its application.
• Electronic surveillance, while offering data, does not address fundamental societal issues like death denial or resource allocation.

Conclusions:

• Computerized models and metrics are tools, not solutions, for complex healthcare challenges.
• The proposed "computational futility metric" requires careful ethical evaluation regarding its uses and potential abuses.
• Electronic epidemiologic surveillance raises important issues concerning patient and physician protection.