Several studies found that certain more complex radiology cases, such as trauma scans or MRIs, benefited more from double reading when compared with examinations such as plain musculoskeletal radiographs. Recommendations include the use of subspecialty reinterpretation of high-risk cases, such as in patients with history of cancer or trauma, or using data from peer review to identify areas where there are more likely to be missed diagnoses and focusing peer review on those areas. Concerns over maintenance of confidentiality by the physicians and fears about the impact of peer review findings on medical malpractice litigation have been identified as a barrier to participation in peer review. One way to overcome this challenge is to deliberately design programs to ensure that all information disclosed through the process of peer review is protected under a state’s statutory peer review privilege, preventing the information from being used against a clinician in malpractice claims. All 50 States and the District of Columbia have privilege statutes that protect peer review records of medical staff members, although how the privilege is applied may vary by state. Traditional random peer review mechanisms employed to maintain compliance with accreditation requirements have not consistently been demonstrated to improve diagnostic accuracy. There is also a need to identify the root causes of discrepancies so that they can be understood and prevented. Discrepancies that are generated because of poor image or specimen quality will be addressed very differently from those that are a result of a lack of time or knowledge by the clinician. Summary of diagnostic errors The patient safety practices reviewed in this section aim to reduce diagnostic errors by targeting cognitive-based factors and systems-based factors. The evidence in support of these practices varies in depth and consistency. CDS offers solutions to address diagnostic errors through incorporation of evidence-based diagnostic protocols, and improve communication and integration with clinical workflow. This review found that CDS may improve diagnosis, although the studies tend to be exploratory in nature, validating the decision algorithms. The use of AI and machine learning has generated excitement over its potential, but they are also exploratory and lack testing during the care of actual patients. These systems need to be reassessed once fully implemented and iteratively improved in real clinical settings on patients actively undergoing diagnosis. Studies included in the review also support the notion that CDS tools are best used as adjuncts to the clinician’s decision making process and not as replacements. This was particularly true for CDS tools that assist with diagnostic study interpretation, such as ECG interpretation. The literature also identified that the diagnoses generated by CDS tools are only as good as the information that is put into the system; if the initial assessment of the patient (e.g., physical exam finding) is incorrect, it is likely that the output will be incorrect.
RNSs aim to address lapses in communication, a contributing factor to delayed diagnosis and treatment of patients in both ambulatory and inpatient settings. For both critical and non-critical CSTR of radiologic studies, lab studies, and tests pending at discharge, the use of RNS showed mixed results in the timeliness of receipt and in action on the test results. Policies and procedures that aligned with the system, mindful integration of the RNS into the existing workflow, and appropriate staffing were identified as factors supporting successful implementation of the systems. Evidence to support education and training on the diagnostic process to enhance clinical reasoning and decrease biases showed generally positive results, with study designs being strong (e.g., randomized controlled trials), although there was some lack of generalizability, as many of the studies had low numbers of subjects. Training on metacognitive skills as a way to reduce biases may improve diagnosticaccuracy, particularly as clinical experience increases. Online training, either didactic or simulation based, was shown to be successful at improving clinical reasoning skills. Studies of peer review show significant numbers of missed or misread test interpretations. However, there is a lack of evidence to show that traditional random peer review and feedback mechanisms used in radiology or pathology to maintain compliance with accreditation requirements improve diagnostic quality over time or prevent diagnostic errors from reaching the patient. For both radiology and pathology, nonrandom peer review appears to be more effective at identifying diagnostic errors than random peer review; and when peer review is conducted prospectively, there is anopportunity to identify diagnostic errors before they reach or harm the patient. Overall, there is still a relative dearth of studies focused on diagnostic error prevention and methods to improve diagnostic accuracy compared with other patient safety topics. General considerations for future research in diagnostic safety include the use of consistent measures and definitions of diagnostic error to allow comparisons of studies and aggregation of data across smaller studies (i.e., meta-analyses), moving from exploratory studies to studies conducted in real clinical settings in real time, and understanding how to best integrate technology with the current workflow to support diagnosis-related activities. Failure to rescue Failure to rescue (FTR) is failure or delay in recognizing and responding to a hospitalized patient experiencing complications from a disease process or medical intervention. As a patient safety and healthcare quality metric, FTR is typically defined as mortality following a complication, although there is no universally agreed upon definition and slight variations exist between institutions. This section reviews two patient safety practices that have been widely implemented to address FTR: patient monitoring systems (PMS) and rapid response teams (RRTs).
Failure to rescue is a well-established issue in patient safety and healthcare quality. Over the past two decades, there have been numerous studies identifying clinical antecedents to in-hospital mortality as well as strategies to respond to these events. Silber and colleagues were the first to use the term as a metric for safety and quality in their 1992 study hypothesizing that FTR might be associated more with hospital characteristics than with patient illness severity. 25 Since then, many studies have investigated the variations in patient outcomes following in-hospital complications and in 2005, the Institute of Healthcare Improvement’s 100,000 Lives campaign identified FTR as one of six key safety initiatives, estimating that implementation of rapid response systems could save 66,000 lives. 26 Because in-hospital complications can occur to any patient regardless of their diagnosis or disease process, FTR represents a ubiquitously significant problem and is therefore an important indicator of care quality. Rapid response systems (RRSs) are hospital- based systems to detect and treat deteriorating patients before adverse events occur. They have emerged as an intuitive approach to address the two core contributors to FTR: failure in adequately monitoring and identifying and failure in responding to hospitalized patients who are at high risk for rapid clinical deterioration. Patient monitoring involves assessment of various vital signs and physiological changes. Monitoring criteria are then used to help guide activation of the RRT. Although there is no universal standard, most rapid response call criteria include abnormalities in physiologic measures such as respiratory rate, heart rate, systolic blood pressure, oxygen saturation, and urine output. Additional criteria may include staff member or family member concern about the patient’s condition, mental status changes, or uncontrolled pain. Once activated by the monitoring staff, the RRT then responds to the patient to prevent avoidable morbidity and mortality. Other models exist, including medical emergency teams and critical care outreach. This section uses “RRT” as an umbrella term, as all models are conceptually united by the goal of early intervention for patients who are at high risk for clinical deterioration. The RRT is typically multidisciplinary and can consist of a nurse, physician, and respiratory therapist, although team composition may vary depending on institutional policy and guidelines. They are able to assess the patient, diagnose, provide initial treatment, and rapidly triage the patient. Patients can then transfer to a higher level of care (i.e., intensive care unit), have their care returned back to the primary medical team, or have their treatment plan revised. Specialized resources such as cardiac arrest teams or stroke teams are considered separate from the RRT and may be involved in the care of the patient, if warranted. Driven by quality and safety requirements as well as recommendations, a swift uptake in RRTs has been noted in the United States and Australia, and is increasingly seen in other developed countries.
80
Powered by FlippingBook