سیستم پشتیبانی تصمیم گیری به اشتراک گذاشته شده در مورد ترمیم دندان
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|5784||2012||7 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 39, Issue 14, 15 October 2012, Pages 11775–11781
Shared decision making (SDM) is an approach in which doctor–patient communication regarding available evidence and patient preferences is facilitated to enable the patient to participate in treatment decisions. SDM affords not only the inclusion of the ethical diversities involved in patient-centered care, but also the quality improvements in decision-making process. Though SDM has been studied extensively, there have been few practical implementations in real clinical environments. In this paper, we propose a shared decision-making system with its focus on dental restorative treatment planning. In our system, restorative treatment alternatives for SDM were generated by employing an ontology that had captured the clinical knowledge required for treatments. We considered patient preferences for treatment as an important support for mutual agreements between the patient and the doctor on healthcare decisions. We constructed a consistent and robust hierarchy of preferences using the analytic hierarchy process (AHP) method, to help determine treatment priorities. On the basis of our proposed system, we developed a Web-based application for the visualization of evidence-based treatment recommendations with preference-based weights. We tested our system using a scenario to illustrate how doctors and patients can make shared decisions. The application is of high value in supporting SDM between doctors and patients, and expedites effective treatments and enhanced patient satisfaction.
A clinical decision support system (CDSS) is an application designed to assist health professionals in decision-making tasks as in regard to diagnosis and treatment planning (Shortliffe, Perreault, Wiederhold, & Fagan, 2001). Focusing on dentistry, there are only a few relevant studies exploring the potentials of CDSS. For instance, Brickley et al. (Brickley & Shepherd, 1996) developed a neural network application to provide decision support for lower third molar treatment-planning. Other CDSS examples for dentistry are caries management (Benn, 2002) and intelligent dental treatment planning (Finkeissen et al., 2003). However, none of these CDSSs were designed to consider patient preferences for enhancing the quality of patient-centred, or personalized, services and thereby improving patient satisfaction. Restorative treatment decision making by dentists and patients often exhibit wide variations (Bader & Shugars, 1995). This appears to be due to differences between patient preferences and clinical judgments. Typically, dentists and patients exhibit differences in their preferences for dental restorative materials. For example, dentists tend to choose one with high longevity, while young patients are sensitive to aesthetics (Espelid et al., 2006). According to (Vidnes-Kopperud, Tveit, Gaarden, Sandvik, & Espelid, 2009), dentists use tooth-coloured restorative materials more often than dental amalgams for restorations in stress-bearing areas in young patients. In general, it is recommended that dentists consider patient preferences for dental restorative treatment alternatives prior to making a treatment decision, if the physical characteristics of the materials are not critical. Patients also tend to participate actively in the restorative treatment planning process when options abound and when they have material preferences (Oates, Fitzgerald, & Alexander, 1995). Furthermore, increased levels of patient participation tend to promote their satisfaction with treatment outcomes. Shared decision making (SDM) is an alternative to the paternalistic care model (Charles, Gafni, & Whelan, 1997). It enables both patients and clinicians to reach mutual agreements on appropriate health care and treatment decisions (Frosch & Kaplan, 1999). In SDM, the patient is provided with all available evidences and information about a given medical problem, and the suitability of each alternative treatment plan is measured in relation to the patient preferences. SDM is a relatively new concept with a few implementations in health care settings. Implementation obstacles of SDM include the task complexity, lack of time (Gravel, Légaré, & Graham, 2006), and missing information (Elwyn, Edwards, Gwyn, & Grol, 1999). In dentistry, Johnson et al. (Johnson, Schwartz, Goldberg, & Koerber, 2006) devised the Endodontic Decision Board (EndoDB) as a decision aid, a popular SDM approach. Though EndoDB facilitates the communication between patients and clinicians through the articulation of alternatives between dental treatment processes, it does not effectively identify or discuss preferences from patients’ perspectives and is not suitable for complex, busy clinical settings. The analytic hierarchy process (AHP) is one of the most well-known and widely used multi-criteria decision making (MCDM) methods. It is specifically designed for dealing with complex decisions that require the integration of quantitative data and qualitative considerations. AHP has been successfully applied to decision making processes across a wide variety of fields including health care (Liberatore & Nydick, 2008), resource allocation, and quality management. It provides both a patient and a doctor with a simple and robust mathematical method for measuring or visualizing preferences in the form of a psychological hierarchy tree with pairwise comparisons (Dolan, 2000), and patients find AHP as a suitable tool for sharing their preferences with their doctors (Singpurwalla, Forman, & Zalkind, 1999). A knowledge-based system that supports CDSSs in SDM requires a considerable amount of domain knowledge (Musen, 2001). A reusable and application-independent domain ontology, or structural framework, would be able to minimize the efforts required for knowledge attainment (GRUBER, 1993). Though CDSSs are developed using ontologies, few are applied to dental restorative treatment planning systems. In our previous study (Park, Lee, Kim, & Kim, 2010), we developed an ontology to represent the concepts related to the decision-making processes involving the choice of restoration types. The ontology specified the tacit knowledge of doctors participating in SDM, thereby facilitating communication and knowledge sharing between the doctors and the patients. The paper proposes a shared dental restorative treatment decision support system that manages the knowledge required for problem-solving and converts it into a machine-understandable form, an ontology. The system computes priorities for treatment alternatives based on patient preferences for AHP methodologies. It offers doctors and patients common understanding on which shared decisions are reached based on patient preferences and accumulated clinical knowledge. The remainder of the paper will describe the system designs and illustrate a practical usage scenario.
نتیجه گیری انگلیسی
In this study, we have presented our ontology-based SDM system for dental restorative treatment planning. Using an ontology, domain expert knowledge such as a CPG is organized, structured, and implemented to construct a CDSS. Using the AHP methodology in our system, we built a consistent and robust hierarchy of preferences and performed preference measurements. A preference hierarchy with ontology-generated, evidence-based alternatives is an effective means by which to reach a shared decision between the doctor and the patient. A Web-based interactive CDSS application is implemented to support shared decision making. The system is useful in various clinical cases in restorative treatment by providing effective treatments and enhanced patient satisfaction.