Clinical Decision Support Systems (4)

Introduction Decision Support Systems Variant Tools and Frameworks

TBNHI Framework

The coordination game allows a doctor (player) to predict the most likely course of action that will be taken by each doctor based on his/her knowledge of the available options and his/her expected payment under the auditing policy.

Hence a payment audit game among doctors is used to ensure that their patterns of prescribing antibiotics do not deviate too much from the average of their peers. When a doctor contemplates the payment (or payoff) he/she will have an incentive to coordinate changes in antibiotic prescriptions.

This can be separated into five stages:

1. Determination of objectives and/or identification of problem.
2. Collection and/or analysis of information
3. Definition and/or comparison of an alternative course of action.
4. Action
5. Review of results (feedback on payoff)

this study shows that game-theory modeling, built into the decision support systems, has made significant contributions to understanding the coordinating instruments for managing the doctors practice patterns and encouraging doctors toward these patterns favored by the third-party insurance payer.

the real-world data set, gathered from the NHI program in Taiwan, shows that doctors patterns of prescribing can be reshaped through peer-comparison and the monetary payoff of MCPA procedures. This research also shows that the quality report cards help the doctor filter and evaluate information for establishing suitable patterns in their practice management. This study concluded that the DSS with game-theory modeling has made significant contributions for improving the doctors’ prescribing behavior.

GPs Tools

GPs are the decision makers responsible for choosing appropriate pathology tests for specific patients in specific situations. This research describes the specific case of designing a framework for an intelligent DSS (Knowledge discovery through data mining and text mining) in the context of pathology test ordering by general practitioners (GPs).

In doing so it illustrates the processes of discovering practical and related knowledge from pathology request data generated and stored in a professional pathology company, investigates and understands the decision makers (GPs) through a survey about their current practices in test ordering and their requirements for decision support. Our framework for the proposed intelligent DSS draws together a new operative and robust methodology that can be used to generate the required evidence to support GPs decision making and achieve more effective and appropriate pathology test ordering.

Result of research is these process and framework contribute effective guidance for practitioners as well as theoretical understanding concerning intelligent decision support in a complex environment. The process and framework developed through this case contributes effective guidance for practitioners and theoretical understanding concerning intelligent decision support in a complex environment.

Medical Database Adaptor (MEIDA)

The three separate complementary tasks in the MEIDA framework. First, the clinical domain expert embeds standard terms and units within the knowledge base of an MDSS. Second, the local DB administrator and the local terminology expert map the local database schema, terms and units into standardized schema, terms and units. Third, access is provided at runtime to the clinical local DB, using the mapping tables that were created during the mapping phase. The Medical Database Adaptor (MEIDA), for linking knowledge based medical decision-support systems (MDSSs) to multiple clinical databases, using standard medical schemata and vocabularies.

IBM Watson for Clinical Decision Support

IBM Watson is now being developed to help medical professionals with their medical decision-making by extracting information from large volumes of data, including medical literature that are relevant to the patient and the decision at hand. In health care, however, Watson would not provide a single answer, such as the only therapy.

Watson actually created a long list of possible responses, and then evaluated the appropriateness of each response to narrow down to likely responses. The rules of the game, however, required that Watson provide only a single answer. There is no such restriction in health care.

Watson performs three cognitive computing tasks. First, Watson analyzes the patient s electronic health record in its natural language form. Watson then analyzes content from many documents, and returns treatment options. Second, Watson is a discovery tool, giving access to extracts from the historical medical record and the documents used to generate the treatment options. Third, Watson is a decision support tool. It learns through repetition what kinds of information are important in making certain decisions. If it is not given that information, Watson will indicate this. When Watson gets the additional information, it can generate updated options and may indicate what other information could be relevant.

EBMeDS

The Evidence-Based Medicine electronic Decision Support (EBMeDS) is a clinical decision support service (DSS), which receives structured patient data from an electronic health record (EHR) system and returns a structured decision support response message to the EHR. The EBMeDS may be used with a variety of health care information systems, including PHRs, EMRs, EPRs etc. Only for the sake of simplicity this document refers to all of them as EHR. The request and response messages are formatted in XML and are exchanged using the HTTP protocol. The service is usually installed and maintained in the local network of an EHR.

Maryam Ghalandari