COMPARTMENTAL-MODELS FOR GLYCEMIC PREDICTION AND DECISION-SUPPORT IN CLINICAL DIABETES CARE - PROMISE AND REALITY

This paper reviews and critically appraises the application of compart mental models for generating glycaemic predictions and offering clinic al decision support in diabetes care. Comparisons are made with altern ative algorithmic-based approaches. Unresolved issues raised for model -based techniques include the relative lack of input data necessary fo r generating reasonable blood glucose predictions, and the high level of uncertainty associated with such predictions which limits their use as guides for therapeutic insulin-dosage adjustments. It is concluded that compartmental model-based approaches, while not offering much be nefit for clinical/therapeutic application, will have a role to play a s research tools and for educational use. By contrast it is proposed t hat algorithmic-based approaches, especially in conjunction with telem edicine and Internet applications, are likely to see growing use for d ay-to-day therapeutic decision support. Randomised controlled clinical trials however will be required, together with other evaluation effor ts, before algorithmic-based approaches-like any other clinical techni que-can be widely adopted into routine medical practice. (C) 1998 Else vier Science Ireland Ltd. All rights reserved.