Hospital emergency care takes place in a high-volume, highly complex environment that is prone to errors and quality concerns. Patients often arrive at the emergency department (ED) in large numbers with problems ranging from heart attacks and HIV complications to burns, gunshot wounds, and domestic violence. They must be quickly triaged for surgery, hospital admission, or other types of care. ED clinicians often must make rapid, life-altering clinical decisions with little knowledge of the patient. Three recent articles are summarized here that focus on quality of care in EDs; a fourth study describes a tool for use by hospitals in assessing their domestic violence intervention programs.
The first article, by Robin M. Weinick, Ph.D., of the Center for Primary Care Research, Agency for Healthcare Research and Quality, reveals the limitations of large data sets in conveying the rich complexity of care in the ED. In the second article, Helen Burstin, M.D., M.P.H., Director of AHRQ's Center for Primary Care Research, discusses the important role of EDs in improving patient safety and reducing medical errors in the ED. The third AHRQ-supported study (HS11592) by Pat Croskerry, M.D., Ph.D., of Dalhousie University Medical School, suggests cognitive strategies to improve clinical decisionmaking in the ED. Finally, measures to evaluate the quality of ED-based domestic violence programs are described in the fourth article by Jeffrey H. Coben, M.D., formerly AHRQ's Domestic Violence Scholar-in-Residence and currently with the Allegheny-Singer Research Institute. The four articles are described here.
Weinick, R.M. (2002, November). "Things my data never told me." Academic Emergency Medicine 9(11), pp. 1071-1073.
Existing large-scale data collection efforts such as the National Hospital Ambulatory Medical Care Survey, the Medical Expenditure Panel Survey, and the Healthcare Cost and Utilization Project each provide unique information on visits to EDs. They contain data on the number of visits made, the diagnoses, procedures, and medications involved; the individual characteristics of patients; and charges and payments for visits. The challenge is how to go beyond this information to understand a broader spectrum of the ED experience, suggests Dr. Weinick.
Large-scale data sets are unable to capture the complexities of everyday experiences in the ED, including patterns of staff activity, the impact that conditions elsewhere in the hospital have on the ED, and the communication and teamwork required. Smaller efforts, whether quantitative or qualitative, often have the ability to capture this greater detail and richness, but they are limited in usefulness by their smaller sample sizes and a less representative scope of data collection. The challenge lies in the tradeoff between large-scale efforts that give an overview of many people's experiences and gathering the depth of information that brings added real-world relevance. Health services research on emergency medicine needs data from the patient, staff, and system perspectives that can be compared across institutions to increase knowledge, improve working conditions, reduce waiting times and overcrowding, and improve patient experiences and care quality.
Reprints (AHRQ Publication No. 03-R019) are available from the AHRQ Publications Clearinghouse and AHRQ InstantFax.
Burstin, H. (2002, November). "Crossing the quality chasm in emergency medicine." Academic Emergency Medicine 9(11), pp. 1074-1077.
The Institute of Medicine (IOM) report, Crossing the Quality Chasm, estimated that between 44,000 and 98,000 deaths per year are from medical injury. Dr. Burstin suggests several ways to improve ED quality of care and patient safety, such as use of electronic medication prescribing and stand-alone, hand-held decision support systems. She recommends developing effective ED reporting systems that could help reduce ED medical errors and near misses; examining areas of high risk for error within the ED, such as triage misdiagnosis or misreading of radiology films; and use of error analysis tools, such as root cause analysis, to learn from errors.
She also suggests several ways to improve emergency care effectiveness, such as focusing on underuse and overuse of diagnostic and therapeutic approaches in ED medicine. For example, it has already been recommended that ED clinicians increase their use of currently underused lifesaving thrombolytic (clot-busting) therapy for heart attack victims. Improved ED efficiency will also result from better timeliness of ED services, such as quicker administration of antibiotics to pneumonia patients and thrombolytics to heart attack patients, as well as a better understanding of factors leading to ED overcrowding and ambulance diversion.
In addition, developing, testing, and evaluating ED information systems may help reduce diagnostic testing and get information to the primary care providers who need to continue the care of patients when they leave the ED. Making emergency care more patient-centered, for example, by not moving patients in pain from place to place and by involving families in ED care, will improve quality of care. Finally, Dr. Burstin calls for reducing disparities in care by ensuring that the patients with the highest need get high-technology, potentially lifesaving therapies, regardless of their race, ethnicity, language, or ability to pay. EDs need to provide more culturally and linguistically appropriate care. Trained interpreters must be available to patients when they need them in order to get adequate patient histories and avoid medical errors.
Reprints (AHRQ Publication No. 03-R011) are available from the AHRQ Publications Clearinghouse and AHRQ InstantFax.
Croskerry, P. (2002, November). "Achieving quality in clinical decision making: Cognitive strategies and detection of bias." Academic Emergency Medicine 9(11), pp. 1184-1204.
Decisions by ED physicians often serve as a barometer of good emergency care. Yet they typically make them for patients whom they do not know and whose illnesses they see through only small windows of focus and time. Attending ED physicians are often responsible for ten or more patients at a time, with one shift involving thousands of individual decisions about diagnoses, tests, and treatments. This cognitive overload is further complicated by ED resource limitations, interruptions, distractions, and shift changes. It is important that emergency physicians understand how to detect the weaknesses and biases in each of their cognitive strategies in order to make good clinical decisions, notes the author.
Dr. Croskerry attributes medical errors arising from ED physician decisions to one or more cognitive biases or dispositions to respond in a particular way. For example, ruling out worse-case scenario (ROWS) is a cognitive strategy of safety that errs on the side of caution, with the physician matching a patients' chest pain symptoms against the worst cases of unstable angina or heart attack. However, this approach depends on a physician's experience, and idiosyncratic applications of ROWS may lead to overuse of resources. Another cognitive bias is the tendency for a particular diagnosis to become established without adequate evidence, so-called diagnosis momentum. Decisions may also be based on male/female bias, for example, when ED caregivers are more vigilant for signs of domestic violence in females than males.
Cognitive errors due to such biases or dispositions often underlie delayed or missed diagnoses, a frequent cause of medical error. The increasing use of clinical decision rules, as well as other aids that reduce uncertainty and cognitive load, for example, computerized clinical decision support, will improve certain aspects of clinical decisionmaking. However, flesh-and-blood clinical decisionmaking will remain, and there will always be a place for intuition and clinical acumen. Instructing physicians in training about typical errors and how to avoid them, so-called de-biasing, may improve their decisions while waiting for the better judgment that comes from experience.
Coben, J.H. (2002, November). "Measuring the quality of hospital-based domestic violence programs." Academic Emergency Medicine 9(11), pp. 1176-1183.
Emergency medicine has been at the forefront of many health care initiatives to treat the growing number of victims of domestic violence. Prior research shows that domestic violence victims have perceived health care providers as uncaring and uninterested in their problem. Providers have also reported discomfort with the topic of domestic violence, inadequate training on the topic, and little knowledge of community resources. This author developed a consensus among a panel of 18 experts, including domestic violence researchers, program planners, and advocates, on measures useful for evaluating the quality of hospital-based domestic violence programs. Following several rounds of scoring on the usefulness of various measures for evaluating hospital-based domestic violence programs, the experts agreed on a total of 37 measures.
These measures fell within nine different domains of domestic violence program activities: policies and procedures, hospital physical environment, hospital cultural environment, training of providers, screening and safety assessment, documentation, intervention services, evaluation activities, and collaboration with local programs. For example, hospital performance would be rated highly if a hospital had written policies on domestic violence, a domestic violence task force, a standardized screening instrument, good collaboration with local domestic violence programs, an intervention checklist for staff use when victims are identified, available inpatient beds for victims who cannot go home or cannot get to a shelter, transportation for victims if needed, routine psychological assessments performed within the context of the program, followup victim contact and counseling, on-site legal counseling, and services for children of victims.
The measures also evaluated a hospital program's preventive outreach and public education activities; qualifications of the domestic violence coordinator; availability of forensic photography; coordination with local police/prosecutor; staff and administration knowledge and attitudes about domestic violence as a health care issue; and client satisfaction and community feedback on the program. These performance measures require refinement but should aid hospital efforts to implement domestic violence programs, monitor progress, and improve quality.