Vimla L. Patel - US grants

DESCRIPTION (provided by applicant): The incidence of HIV infection among young adults continues to accelerate at alarming rates despite widespread educational efforts and dissemination of materials designed to increase awareness of risk. With young adults under the age of 25 accounting for more than half of all new HIV infections, there is an urgent need to reduce the incidence in this population. Many interventions aimed at increasing HIV knowledge have not produced substantial decreases in risky sexual behaviors, and this lack of effectiveness may be a result of differences in how young adults make decisions about abstract, prototypical situations versus specific, real-life instances. Paradoxically, young adults make safe choices when reasoning about prototypical sexual situations, yet still exhibit risky behavior in actual, real-life settings. By examining young adults' decision-making in both real and prototypical situations, we may identify differences in reasoning processes that reveal why previous interventions were ineffective and also delineate dimensions of this paradox that must be addressed in future interventions. We aim to (1) conduct qualitative research to characterize the context surrounding young adults' risky behavioral choices and apply specialized cognitive methods of discourse analysis to examine processes underlying young adults' decision-making; (2) develop and refine cognitive models of the decision-making processes used by young adults in both real-life and hypothetical situations; and (3) compare these models of decision-making to identify and separate factors or characteristics associated with healthy vs. unsafe choices in risky situations. We will recruit 30 heterosexual men and 30 heterosexual women from a local city college who will complete daily journals chronicling their sexual activity over two weeks. We will use our analyses of journal contents to guide subsequent phases of the project. These include semi-structured interviews to investigate the reasoning strategies and situational factors underlying decision-making in both (a) real-life situations as recorded in the journals and (b) a prototypical risky situation compiled from instances of unsafe sexual encounters developed from review of the journals. From the semi-structured interviews, we will implement cognitive techniques of analysis, including propositional analysis and construction of semantic networks, to model the process of decision- making in each of the two contexts. Comparison of models will reveal the factors underlying the decision-making process and how they differ in real-life and prototypical situations. We anticipate that our analysis will identify those factors that promote healthy choices and those that lead to risky behavioral choices. Our findings will provide valuable insight into the aspects of young adults' decision-making that are most critical to target in interventions.

DESCRIPTION (provided by applicant): This grant application proposes to provide web-based on-line information and knowledge resources to mental health professionals at the point of care, These resources include electronic documents and texts, databases, search engines, guidelines and any reference materials that are relevant to the decision-making process in patient care. The proliferation of errors in medical practice has been well publicized in recent years. Although the data on adverse events in psychiatry is less complete, there are numerous studies documenting the ways in which clinical practice deviates from acceptable standards. Access to timely information is often cited as on the proximal causes of error and substandard practices. Recent initiatives taken by the American Psychiatric Association, accrediting bodies, and affiliated mental health organizations to promote evidence-based medicine in the mental health profession have made available an unprecedented and rapidly growing pool of on-line resources such as clinical practice guidelines for physicians and nurses. We propose to integrate lnfobuttons manager technology with an existing electronic medical record system, Mental Health Automated Retrieval System (MHARS). lnfobuttons is a resource delivery system that has been implemented in clinical information systems at Columbia. It works by creating links within the EMR that passes the clinician's context to an application that matches the context to an information need, selects a resource for that need, formulates a query to that resource, passing the query to the resource, and displaying the results to the clinician. The result of the action is to open a browser window on the clinician's desktop with links in the forms of queries to context-sensitive Medline searches, guidelines and other pertinent documents. The specific aims are: (1) Provide mental health professionals with access to guidelines and other resources in real time at point of care. (2) Adapt the lnfobuttons information resource delivery system to the field of mental health with an initial focus in the specific area of pharmacology. (3) Employ lnfobuttons Manager application in concert with the newly implemented electronic medical record system, Mental Health Automatic Retrieval System (MHARs) (4) Implement lnfobuttons in 4 representative institutions by the patient population served by the Office of Mental Health including: adult, child, forensic, and research institutions.

DESCRIPTION (provided by applicant): The medical error report from the Institute of Medicine (1999) has greatly increased people's awareness of the frequency, magnitude, complexity, and seriousness of medical errors. As the eighth leading cause of death in US, ahead of motor vehicle accidents, breast cancer, and AIDS, medical errors need immediate attention from academic, health care, and government organizations. While medical errors can be dealt with from several perspectives, such as organizational restructuring or automation, our concern is with cognitive factors of medical errors and their clinical implications. We propose a theoretical framework in which errors are viewed as inevitable but cognitively useful phenomena that cannot be totally eliminated. In our view human errors are products of cognitive activities in people's adaptation to their complex physical, social, and cultural environments. Our cognitive approach stresses actions in conceptual understanding and thought processes during clinical problem solving. The actions reflect the level of expertise and the demands of tasks in clinical performance. In order to manage errors during clinical decision-making, it is critical to understand how decisions are made and what underlying cognitive mechanisms are used to process information during interactions with patients, colleagues, and technology in the health care environment. The broad objective of the proposed research is to develop a cognitive framework of medical errors in critical care environments (medicine, surgery and psychiatry), where decisions are often made under high stress, time pressure, and with incomplete information. Our specific aims include (1) applying and refining our initial cognitive taxonomy of errors where each category of medical error is associated with a specific cognitive mechanism, and (2) providing a theoretical explanation of why these errors occur and predicting the circumstances in which such errors would occur. Unlike the popular goal of achieving flawless performance (through development of error-free systems), the results from this study will have implications for developing adaptive systems that anticipate errors, respond to them, or substitute less serious errors that allow subsequent intervention before they result in an adverse event.