Term Paper: Decision Support System

Pages: 5 (1843 words)  ·  Bibliography Sources: 5  ·  Level: Master's  ·  Topic: Health - Nursing  ·  Buy This Paper

Decision Support Systems

Describe your understanding of the importance, challenges and barriers of terminologic systems in CDSS.

A clinical decision support system (CDSS) is a type of software designed to aid clinicians in their decision-making by matching individual patient characteristics and knowledge about the patient's disease to computerized databases "for the purpose of generating patient-specific assessments or recommendations" (CDSS, 2010, Glossary). However, implementation of CDSS has been spotty at best. Today, despite the influx of technology into most spheres of modern life, information regarding drug treatment and other vital aspects of patient care are often paper-based in many clinical environments and "even when the information is present in the form of electronic documents on the web or in clinical information systems, doctors are not realistically given adequate time to search for the information specific to a particular patient sitting in front of them" (Park 2009, p.293).

Consistent and comprehensive integration of patient data into easily-retrievable systems is essential to preserving patient safety. In one Korean hospital, with the implementation of CDSS medication error rates fell 81%: from 142 per 1,000 patients to 26.6 per 1,000 patients after the CDSS system was fully mobilized (Park 2009, p.293). However, while the shift to computer-based systems that provide individualized guidance and assistance for physicians is gradually becoming more accepted, inconsistent medical terminology remains a barrier between different hospitals -- even between different nations -- in sharing medically-related information. "The approach taken by the Unified Medical Language System (UMLS), in which disparate terminology systems are integrated, has allowed construction of an electronic thesaurus (the Metathesaurus) that avoids imposing any restrictions upon the content, structure, or semantics of the source terminologies. As such, the UMLS has served as a unifying paradigm by providing appropriate links among equivalent entities that are used in different contexts or for different purposes. It accordingly provides a vehicle through which possibly orthogonal semantic models can co-exist within a single framework. This framework provides a model for the collaborative evolution of biomedical terminology and allows a synergistic relationship between the UMLS and its source terminology systems" (Campbell 1998).

Medical terminologies tend to be broad, sprawling, and complex in their scope, and often quite regional- or discipline-specific. The goal is that CDSS systems should be able to translate a variety of meanings into the same treatment protocols. The barriers to this are partly technological -- errors in translating similar yet distinct terminologies can occur, as well as more conventional errors of translating descriptions of drugs from doctors' handwriting, or incomplete patient histories. There is also a human factor: terms can change quickly in practice, particularly specific terminology used by physicians in subspecialties. Thus, there may be incorrect labeling of cases when physicians use the database and the information itself may be mistranslated and poorly integrated.

It is unlikely that a perfectly universal system of terminology will emerge amongst practitioners in a cross-cultural, cross-disciplinary fashion and also unlikely that any computerized system, however flexible, will be able to create a seamless web of translation for users of CDSS. However, a lack of perfection is no excuse not to try, given the great benefits that can be derived in terms of patient care from the use of CDSS.

Question 2: In an evaluation study, the decision-support system ONCOCIN provided advice concerning cancer therapy that was approved by experts in only 79% of cases (Hickam et al., 1985b). Do you believe that this performance is adequate for a computation tool that is designed to help physicians to make decisions regarding patient care? What safeguards, if any, would you suggest to ensure the proper use of such a system? Would you be willing to visit a particular physician if you knew in advance that she made decisions regarding treatment were approved by expert colleagues less than 80% of the time? If you would not, what level of performance would you consider adequate? Justify your answers.

One of the unfortunate aspects of oncology is that the drugs used to treat cancer are toxic to the body in a uniform manner: they harm the body's beneficial functions as well as the cancer cells. When administering chemotherapy, the aim is to find the correct balance between administering a high enough dose to destroy the cancer cells while still managing the chemotherapy side effects (Musen et al. 1986, p.44). ONCOCIN, a computerized clinical decision support system that was developed in 1979, was designed to use artificial intelligence to offer advice to physicians on medicines, dosages, and testing relating to oncology. It united medical record keeping with decision support (ONCOCIN, 2010, Discovery Media).

ONCOCIN was used to suggest potential treatments based upon two databases of oncology protocols, one procedural and one inferential. Procedural knowledge defined a patient's progression between the various states of the treatment plan. After consulting ONCOCIN for the general treatment plan, ONCOCIN used the inferential component of the knowledge base to "refine" the plan, based upon the patient's individual needs and medical history (Musen et al. 1986, p.44).

Before ONCOCIN was developed, computer-based treatment programs were subject to frequent criticism. Computers must not simply be able to store data, but must be able to select and analyze pertinent data and until ONCOCIN, most computerized support systems were merely data warehouses. ONCOCIN was considered an improvement because it provided information about a patient's past as well as his or her immediate health status, hence the system's more intuitive and responsive nature (Kahn et al. 1985, p.172). By being able to weigh two aspects of the building-blocks of the patient's care -- treatment and medical history -- ONCOCIN was hoped to revolutionize oncology. For example, ONCOCIN could determine drug doses on the basis of time schedule, toxicity, and blood counts, based upon clinical and patient data as well as chemotherapy protocol guidelines and the recommendations of oncologists who had participated in the construction of the computerized advice system (ONCOCIN, 2010, Discovery Media).

However, there were a number of problems with ONCOCIN's implementation. The first was that it was not clear what type of training was required to intelligently review ONCOCIN's results. Typical users of ONCOCIN were new residents and even non-physician clinical assistants rather than experienced doctors (ONCOCIN, 2010, Discovery Media). The justification for the system was that in oncology, so much knowledge regarding toxicity had been studied and formally documented it was easily searchable and fairly 'idiot proof' to use. However, regardless of the statistical level of 'confidence' in the results, it is hard to imagine any patient feeling comfortable going for treatment with a new resident using a computerized program to create a prefabricated treatment plan.

Unsurprisingly it soon was found ONCOCIN was no substitute for actual, trained physicians using their clinical judgment. "In some cases the situations at hand did not fit into the rules known by the system, also it took about six weeks to enter the rules for a new protocol and to test them. Although the knowledge was documented it was not all-inclusive; new protocols were being found and used all the time, there was no way of ever getting a complete set of the protocols. The advice provided by ONCOCIN was approved by experts in only 79% of the cases" (ONCOCIN, 2010, Discovery Media).

When ONCOCIN was developed, implicit in its construction was that information regarding oncology protocols would remain stable. However, that was far from the case. To some degree the system became a victim of the success of cancer treatment. Cancer treatment is a continual and evolving body of research. Additionally, even if the patient's entire history was entered into the system, patient differences were often far more subtle and complex than could be processed through a system like ONCOCIN, such as a patient's willingness to tolerate certain levels of stress and pain, emotional stability, and family support. It was far more difficult to mathematically predict different patient variables and deviations from norms than expected, and even this does not take into consideration the full picture of the patient's unique situation.

Physicians, at the time of ONCOCIN's conception, were wary of the computerized decision-making system's approach, particularly given the fact that the results were only approved by experts in about 80% of the cases. When considering advice in regards to routine treatment, physicians were more apt to trust their nearby colleagues and their own past decisions for support. For experimental treatments, physicians felt they needed to review the patient's individual needs. 80% is not a meaningful statistic of agreement. "For these reasons, physicians used ONCOCIN as a critique to their own work. ONCOCIN was converted into an embedded critic: rather than use the system primarily to generate treatment plans, doctors were intended to routinely enter their own plans into ONCOCIN and the system offered criticism as a side benefit. They would input the information and look at the significant differences between the plan proposed by ONCOCIN and their own" (ONCOCIN, 2010, Discovery Media). ONCOCIN was a support system, like an additional colleague giving advice, or consulting peer-reviewed journals. However, unlike real… [END OF PREVIEW]

Decision Support Systems Term Paper

Decision Support Systems Research Proposal

Business Decision Support System Thesis

Information Technology and Decision Support Computer Essay

Management Systems What Are the Pros Essay

View 1,000+ other related papers  >>

Cite This Term Paper:

APA Format

Decision Support System.  (2010, June 25).  Retrieved September 20, 2019, from https://www.essaytown.com/subjects/paper/decision-support-system/7988745

MLA Format

"Decision Support System."  25 June 2010.  Web.  20 September 2019. <https://www.essaytown.com/subjects/paper/decision-support-system/7988745>.

Chicago Format

"Decision Support System."  Essaytown.com.  June 25, 2010.  Accessed September 20, 2019.