Blockchain Papers

Tina Yi Jin Hsieh, Carl Eriksson, Garth Meckler, Matthew Hansen · 12 authors

Introduction and Objective: Traditional adverse safety events (ASE) identification relies on domain experts to manually review and annotate charts, which hinders the scalability of processing high-volume EMS data. This study explores the use of large language model (LLM) with a knowledge base to automate extraction of adverse safety events (ASE) from unstructured emergency medical service (EMS) notes for pediatric out-of-hospital cardiac arrest (OHCA) as proof of concept. Data Sources and Study Design: Pediatric OHCA records from a national EMS provider were obtained from 2017 to 2020. Leveraging the Pediatric Prehospital Adverse Safety Event Detection System (PEDS) as a foundational knowledge base, we used the LinkML framework to develop an ontology to define ASEs across six essential EMS care domains. To convert unstructured EMS narratives into structured prompts, we used the Structured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES) method, which generated schema-driven prompts to guide the GPT-3.5 model in identifying ASEs. By mapping unstructured data into structured concepts consistent with PEDS guidelines, the model produced targeted prompts that supported effective entity extraction. Results: We evaluated framework effectiveness with accuracy, recall, precision, F1 score, and specificity across 42 pediatric OHCA cases covering ASE-related entities. RescueGPT showed high accuracy in detecting common ASEs (Patient Rhythm, Age, Weight, Length) but revealed challenges in rare events (Failure to Establish IV Access, Incorrect Airway Equipment Size, Failure to Ventilate Patient) likely due to more inconsistent and complex documentation. Conclusions: RescueGPT demonstrates potential in scaling automated ASE detection, but performance varies by completeness and clarity of EMS narrative, particularly with rare events. Fragmented clinical documentation limits accuracy and highlights the need for standardized collection protocols in EMS systems. Future directions will focus on implementing rebalancing strategies for rare events, applying explainability methods to improve decision-making transparency, and refining text segmentation techniques to handle mixed outcomes to further improve performance.

Adugna Negussie, Abay Sisay, Agumas Shibabaw, Edosa Kebede · 9 authors

Abstract Background Medical laboratory professionals play vital role in healthcare. The growing demand for quality laboratory services and emerging technologies underscore the crucial need for Continuing Professional Development (CPD). However, there is limited information on CPD programs in Ethiopia. Thus, this study aimed to assess medical laboratory professionals’ perceptions, attitudes, and challenges towards CPD and improve engagement to enhance diagnostic service quality. Methods this cross-sectional study enrolled 228 medical laboratory professionals in Ethiopia from July to October 2023. Using a mixed-methods approach that combined quantitative data from an online survey and qualitative data from interviews. SPSS version 28 was used for data analysis. Results the average age of the study participants was 32.6 ±6.4 (SD) years, the majority were men (88.6%), and 44.3% have worked for more than ten years. Of the participants, 51% never had CPD training. About three-fourth of the participants perceived CPD as essential to their professional career. About 45.2% of the study participants perceived that the purpose of CPD course is to renew their license and gain knowledge and skills that are not covered in basic training. While the majority of participants had good attitudes towards CPD, about 10% of them stated that it is not important in their career growth. The majority of the study participants were not in support of the decentralized CPD system. A notable problem with finance, insufficient manpower, unsupportive employers, a lack of awareness by regulatory bodies, inadequate access to training close to their working area, were identified as significant challenges of the CPD program. Conclusion the study highlights the perception that CPD is crucial for enhanced laboratory practices and career advancement. The study highlights the need for targeted strategies to address the identified problems and increase the engagement of medical laboratory professionals in the CPD program.

Madison Milne‐Ives, Ching Lam, Najib Rehman, Raja Sharif · 5 authors

BACKGROUND: Adverse drug event reporting is critical for ensuring patient safety; however, numbers of reports have been declining. There is a need for a more user-friendly reporting system and for a means of verifying reports that have been filed. OBJECTIVE: This project has 2 main objectives: (1) to identify the perceived benefits and barriers in the current reporting of adverse events by patients and health care providers and (2) to develop a distributed ledger infrastructure and user interface to collect and collate adverse event reports to create a comprehensive and interoperable database. METHODS: A review of the literature will be conducted to identify the strengths and limitations of the current UK adverse event reporting system (the Yellow Card System). If insufficient information is found in this review, a survey will be created to collect data from system users. The results of these investigations will be incorporated into the development of a mobile and web app for adverse event reporting. A digital infrastructure will be built using distributed ledger technology to provide a means of linking reports with existing pharmaceutical tracking systems. RESULTS: The key outputs of this project will be the development of a digital infrastructure, including a backend distributed ledger system and an app-based user interface. CONCLUSIONS: This infrastructure is expected to improve the accuracy and efficiency of adverse event reporting systems by enabling the monitoring of specific medicines or medical devices over their life course while protecting patients' personal health data. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/28616.

Khaled Al–Surimi

We are pleased to publish the second issue of the Global Journal on Quality and Safety in Healthcare (JQSH). In this issue, we would like to discuss the similarities and differences between research and quality improvement (QI) projects in health care. Imagine you are working in a hospital or a department within a hospital and you want to improve an aspect of health-care quality and safety by focusing on the issue of medication errors. Given that situation, you decide to implement a “zero harm” rule because of medication errors. The question is will this be a QI or a research project? In another example, you are a resident working in an oncology department and you noticed that most patients receiving certain chemotherapeutic agents had neuropathy complications, so you decided to collaborate with the physical therapist on a project to compare patients who received chemotherapy drugs and exercise with those who did not exercise. Again, the question is will this be a research project or a QI project? Regardless of the answer, it is important to implement the project systematically. If your project is focused on QI, then you should consult the QI specialists in your hospital who can help you to use the appropriate QI methodology, which includes Plan, Do, Study, Act (PDSA) cycles. If your project qualifies as research, then you should consult a research methodologist and biostatistician regarding study design, sample size, and others and work with the institutional review board (IRB) to provide guidance and templates.Many health professionals do not know how a research project differs from a QI project and when they complement each other.[1–3] Our traditional thinking is that quality and safety improvement in health care as well as the effectiveness of an intervention can only be studied in the form of a traditional scientific research project, as it has its own well-established rigorous approach. We may be ignorant or unaware of how to use the QI scientific approach to study the performance of a health-care system.[4,5] The problem lies within our frame of thinking because we are prioritizing the proof of effectiveness over bringing about and sustaining improvement. We use the results of pre-assessment and post-assessment research as the gold standard for evidence-based policy and practice, whereas in reality, sustaining the improvement is continuous and more dynamic.[1,6]Research projects are question-driven and focus on providing proof of effectiveness. The main purpose of research is to generate new generalizable knowledge about a particular subject to a study population, where the study results often end up published in academic journals. In this case, researchers must follow a strict study protocol approved by the IRB, including obtaining the consent from study participants before starting the project and report any deviation from the protocol to the IRB, if needed.[7–9] However, QI projects are data-driven and focus on showing sustained improvement to a specific process and system or outcomes within a health-care organization using, if possible, the research evidence generated as the basis for developing the improvement interventions.[10] A QI project does not aim to generate new knowledge as a research project does, rather, it generates several learning lessons as to what actually works and does not work and why. A QI project produces empirical evidence to benefit other organizations within a similar context and setting, which are interested in replicating the change to improve a process or system using the rapid PDSA cycle approach.[11] Through cycles of testing, we learn what is going to improve and why, without the need to generalize the results to another context, as research projects usually aim to do. Also in QI projects, the measurement framework is not about pre and post. It is about continually measuring the metric of interest that you want to improve and coming up with not just one intervention but multiple interventions based on learning from prior PDSA cycles. At the end, you reach the point of realizing sustained improvement through a series of interventions that were informed by testing in the actual system that you want to improve. The PDSA cycle is repeated, and new changes are made to continue to improve a process and, ultimately, the outcome. The essential measurements included in a QI project are process measures, outcomes measures, and balancing measures, which are used to show that the improvement occurs over time. Data from QI activities are usually aggregated and presented in run/control charts, histograms, and line graphs, whereas data from research are analyzed using statistical tests such as t-test, chi-square test, and regression analysis, and then aggregated and presented in appropriate tables and/or graphs.Typically, QI results are shared within the organization and might be implemented in other departments. The lessons learned from QI activities can be published; however, it must be clear to the readers that the project was for QI, not traditional research. Although a QI project does not require IRB approval, some organizations have QI committees that approve and coordinate QI project activities, and some organizations require articles to be approved before submitting for publication.In summary, the sustained improvement realized in a QI project can be complemented and validated with a thorough research-based assessment of effectiveness.[12] We should not consider the proof of effectiveness the same as the proof of sustained improvement, but they both are very important. I would like to emphasize that both research and QI projects use scientific and systematic approaches, albeit different, but both methods are scientific and rigorous in their own ways. The aims, methods, and outcomes in research and QI projects are quite different. Hence, understanding the differences and similarities between research and QI projects will help to determine the right approach when designing and implementing the right project for the right purpose using the right method. Table 1 is a snapshot comparison between QI and research with more focus on the project's aim and method aspects.In research projects, we can be guided by asking the following: Do we have a clear question to be investigated and answered?What do we hope to accomplish by answering the question?What is currently known about the topic?What are the risks and benefits for patients involved with the study of this topic?What type of study design will be used (observational vs. experimental)?How will the data be analyzed and presented (statistical tests, P-values, etc.)?In QI projects, we can ask the following: What is the magnitude of the quality problem based on available data?What types of quality tools have been used to measure and assess the problem?What is the measurement plan to be used during implementation of the project?What types of changes/interventions will be tested during the PDSA cycles?Has the proposed change/intervention been used in other health-care settings or reported in the literature?Will the results of this project directly improve patient-care outcomes or processes?Is the organization's management supportive of the project and willing to dedicate employee's time and supplies to do the project?What is the sustainability plan for the results?

David M. Gaba

The healthcare industry may be the largest and most expensive endeavor of the developed world, with the United States at the top of the list of per capita expenditure. Clearly, as indicated by the intense (and continuing) debate over the Affordable Care Act, the issues of the healthcare industry are of extreme interest to the public and policy makers.The biggest problems in the healthcare industry are about how to achieve its fundamental goals—how to provide adequate and equitable care to the entire populace; how to guarantee equitable access to all; how to achieve optimal population health; how to ensure efficacy, quality, and safety of patient care; how to provide choice of provider and hospital; and, most importantly, how to pay for all of these goals and how to obtain political agreement of the populace to make it happen.Fortunately, this monograph will address primarily issues of quality and safety, and will largely ignore these other very large and thorny issues. Some of the ideas in this chapter have been addressed in part by the author in prior journal publications.12From a safety standpoint, it is now well recognized that there is a significant incidence of harming patients in the course of trying to diagnose and treat them.3Many such events are known to be preventable. The incidence of minor problems is very high, but even serious events have been found in approximately 1% of all hospitalizations. It is often said that many of these adverse events are irrelevant because the patients they occur in are already very ill, and hence might well have suffered negative outcomes anyway. However, I contend that no patient “signs up for bad care,” so we should still be very concerned about such events even when they do not, in the final analysis, actually affect the final outcome. The next time, maybe they will.In addition, it is likely that many errors that occur, even serious errors, are hidden. Some are not apparent because the patient is very ill, so only detailed investigation or analysis can disclose an error. In other cases, errors can be hidden simply by failing to inform anyone of them and waiting to see what happens. Moreover, healthcare does not have the robust incident or accident investigation processes that are routine in transportation (e.g., National Transportation Safety Board). Most investigations—in the infrequent occasions that they occur—are conducted only at the local level and with varying degrees of sophistication and alacrity.Parts of healthcare (such as anesthesia and surgery) and nuclear power production are but two examples of activities of “high intrinsic hazard” (aviation is a well-known third). The hazard in these activities is inherent—it can be managed and controlled—but the hazard cannot be eliminated. Yet, the management of hazard in nuclear power and aviation has become so good that it is accepted that adverse events are not “normal.”Nuclear reactors should not unexpectedly interrupt power production, and they should never harm workers or the public, melt down, or explode. Airplanes are not supposed to crash—ever. In these arenas when one of these things happens, we know that something went horribly wrong. Yet, human beings are inherently prone to catastrophic internal failures that result in serious disability or death. Thus, adverse outcomes—not necessarily due to errors or poor care—are very common in healthcare. All of us are going to die, and most of us will die in close proximity to healthcare. It is difficult to sort out which events are the ordinary “natural history” of disease and which are due to suboptimal care. This makes efforts at measuring safety outcomes particularly difficult in healthcare.All of the high intrinsic hazard industries share the fact that they are so critical to human welfare that we cannot just shut them all down while we solve all of their problems. Certainly, we can't stop performing healthcare activities just because they are imperfect—the ravages of disease are worse. While a single nuclear reactor can be shut down whenever necessary, and a flight can be cancelled or delayed, it is sometimes impossible (and possibly unethical) to refrain from or abort an emergency medical procedure due to a significant safety risk when the patient's underlying disease processes will otherwise quickly cause serious harm or death.Decisions on a larger scale are more complicated. While it is in principle possible to abandon the use of nuclear power in some countries, this can only be done temporarily or it must be phased out over a very long period of time. Access to abundant electrical power is the lifeblood of modern societies. Similarly, the dislocations caused by even short stoppages of air travel by the 9/11 terrorist event or the Icelandic volcano's ash cloud demonstrated that air travel also cannot be stopped for long. On the other hand, in healthcare, the introduction of new, potentially lifesaving drugs and devices can be delayed pending proof that they are safe and effective.The calculus of such decisions may vary from country to country, although many aspects of healthcare (and nuclear power production) are similar everywhere. In healthcare, the practices of physicians stem originally from the roots of the “autonomous healer” who used individual, often idiosyncratic, knowledge and “skill” to diagnose and treat ailments. There were few curative or invasive therapies. While administering potions to, cupping, and bleeding patients didn't usually help them very much, and might have hastened their demise, they were not generally powerful enough to directly cause serious harm or death. Hospitals were originally organized more as “guild workshops” 4 wherein the members of the physician's guild could independently ply their trade.Now, in the early 21st century, some things have changed drastically while others have not. We have many more diagnostic and treatment interventions that can often cure. Many are very powerful and can themselves directly, and quite quickly, cause serious harm or death. I like to say that there is a high potential lethality per square meter in settings like the operating room, intensive care unit, emergency room, or chemotherapy administration unit. Wielding such interventions requires very complex care coordinated across many individuals and many work units.Since the latter half of the 20th century, it has become possible to compare many patient outcomes in response to diagnosis or treatment, a process that is still unfinished. Despite all of this change, the structure of the hospital, for example, has not changed much in hundreds of years, retaining many elements of the guild workshop. Even where an institution is the employer of physicians, the amount of autonomy of practice given to physicians is enormous, despite the grumblings of how medicine is dictated by the rules and regulations of payers and other bodies. The system also is structured around assumptions that the individual skill of the professionals will be uniform, solid, and unvarying over time, which of course is impossible to guarantee.Even the division of labor is old. I conjecture that if healthcare were to be developed now, from scratch, we would not have job types of “doctor,” “nurse,” “pharmacist,” and “respiratory therapist,” to name only a few. We would have many other job types and a vastly different organizational and work structure—hopefully based on a more rational assessment of how best, and how safely, to achieve the goals of the work in the first place.A fundamental difference in healthcare versus other industries is that “we” do not design or construct the units we work on: human beings—nor are we given an instruction manual for them. We do not understand a great deal of how the human body works, how it fails, or why and how it gets sick or recovers from illness. Yes, great strides have been made and more discoveries are happening every day, but we are mostly working empirically by trial and error.In my own field of anesthesia, we do not know many of the fundamental mechanisms by which our drugs can render patients unconscious, unaware, resistant to pain, immobile, and (fortunately) unable to recall what has transpired during surgery. Yet, by trial and error, we have worked out the methods to do these things—which clearly evolution never really intended for human beings—on a regular basis with low, but not low enough, rates of serious problems.In healthcare, the public is very concerned with personal and intimate aspects of the work, and such individual, societal, and ethical issues are commonplace. They also care very deeply about choosing and seeing “their” doctor. This is not the case for other industries where the public doesn't care specifically who exactly is doing the work (pilots and nuclear power plant operators interact with the public minimally, if at all). However, for nuclear power, the public has great concerns over the long-term impact of accidents, and also a hard to grasp “dread” factor of radiation that does not come into play in healthcare.56Organizationally, the nuclear power industry and healthcare are very different. There are just over 100 nuclear power reactors in the United States, owned and operated by 30–40 firms and under significant scrutiny by the federal regulator, the U.S. Nuclear Regulatory Commission (NRC).Healthcare is a vastly more decentralized and massive undertaking. There are 4,000– 6,000 hospitals, owned by 1,000–2,000 firms. There are roughly the same number of stand-alone surgicenters. There are more than 200,000 physician offices. More than 20 million surgical operations with anesthesia are performed, just under one billion doctor visits occur, and about three billion prescriptions are written every year in the United States. Yet, there is no federal regulatory agency of the practice of healthcare. That comes under the jurisdictions of the 50 states and the federal health systems (e.g., Department of Defense, Department of Veterans Affairs, and the Indian Health Service).The federal U.S. Food and Drug Administration regulates the approval and sale of drugs and devices. The federal Centers for Medicare & Medicaid Services (CMS) controls the criteria for federal payment for medical services. CMS may act as an indirect regulator of practice—if you won't get paid for it, you probably won't do it—and there are other indirect regulators by accreditation (e.g., The Joint Commission) or by voluntary participation (e.g., Institute for Healthcare Improvement and the Leapfrog Group). However, indirect regulation is generally not comparable to direct regulation, as in the NRC's direct oversight of nuclear power, or the Federal Aviation Administration's direct oversight of aviation.Of note, in aviation and nuclear power, the firms themselves (individual airlines or individual power utility companies) impose strong safety control over the day-to-day work of personnel, often over and above the requirements of the regulator. This is only partially true for healthcare. The work of nurses, pharmacists, and allied health personnel comes under the direct purview of the employing institution, although the degree to which actual practices at the front line reflect the stated goals or policies of the institution varies greatly.The practices of physicians have less direct oversight by the firm; the majority of physicians are independent (fee-for-service, not salaried) members of the hospital's medical staff. As such, though not under direct line authority of the hospital, they must apply for clinical privileges and their actions can be scrutinized by the institution. Other influences on physician practices come from specialty board certification and professional society practice guidelines.However, when guidelines are well articulated, strongly evidence based, and widely agreed upon by the medical community, it typically takes a decade until these practices are consistently adopted and executed. Regardless of whether physicians are actual employees of the hospital or are independent medical staff members, in practice they have nearly unlimited discretion as to how they manage individual patients. Local standardized operating procedures are occasionally imposed, but even then their authority and compliance may be minimal, especially without specific incentives for compliance or disincentives for noncompliance.In fact, all of the hazardous industries suffer from a phenomenon in which what is articulated for safety on paper does not always correspond to the reality at the front line or even to a plausible reality that could be implemented at the front line. One aspect of this has been described by the sociologist Lee Clarke as “fantasy documents,” such as policies, procedures, or plans that are created to satisfy a regulatory, internal, or public relations need, but are known by most participants to be infeasible. They “sound good” and make people feel better, but it is widely known—at least by frontline staff—that they cannot really work as described.78One factor about the aftermath of accidents that affects other industries in a profound way that doesn't happen in healthcare is that a severe accident in nuclear power, in oil refining, or even in aviation, can seriously harm the “means of production.” That is, not only may the accident hurt workers or the public, it also takes out of service the facilities (power plants, refineries, or airplanes) that are used to do the work. Even ignoring cleanup or repair costs (if relevant), this means that there is a huge financial and operational loss from the lost means of production.As indicated above, for nuclear power, this can expand all of the way to long-term plans to abandon this method of generating electricity. None of these effects is seen in healthcare. If we harm a patient in the operating room, that may be very sad, may generate litigation, and may (rarely) garner bad publicity for the hospital, but we just “send for the next patient.”I cynically suggest that if the aftermath of medical errors or preventably suboptimal care events in an OR, ICU room, or emergency department bay would be to take that room out of service for days or months, that would generate a much more aggressive response for improvement by the healthcare institution than we currently see.It is true that healthcare cannot strive for the same level of standardization within a facility, or especially between facilities having the same basic technology, as is achieved in nuclear power or the aviation industry. Human beings are not reactors or airplanes and diseases are not understood at fundamental levels, hence healthcare personnel need more flexibility to respond to unanticipated situations. However, as for many things in healthcare, the pendulum is currently too far to the side of insufficient standardization.On the equipment and procurement side, the decentralization and huge number of sites of care raise all sorts of issues. Unlike the 106 nuclear power plants of perhaps a few dozen designs, the hundreds of thousands of patient rooms, ORs, ICU bays, etc., in the 8,000 institutions each needs outfitting with various devices such as monitors and infusion pumps.Rather than being purchased as large, integrated, preconfigured units, such devices are often purchased one at a time, or, at best, in periodic bundles of hundreds. The combinatorics of all of the devices makes it impossible for vendors to test them in use all together. And, until fairly recently, there was little demand on vendors—either from regulators or the marketplace—for serious human factors testing of either prototypes or actual devices.The decision to purchase equipment is often made by small committees or single influential individuals based on idiosyncratic assessments of features. Purchase decisions are strongly affected by the purchase cost of the equipment and disposable supplies, and only rarely by total life cycle or systems cost. One area where both nuclear power and healthcare can benefit is to achieve and maintain a high degree of user-centered human factors testing of concepts, prototypes, and actual equipment during the design, premarketing, marketing, and postmarketing phases of product life.Issues of design are compounded in healthcare by the current variability in the preparation and training of personnel on the use of the equipment, even that which is life critical. Nursing and allied health disciplines generally have more structured mechanisms for providing training to personnel before they use advanced equipment via “in-services” and checkoffs of competency.Even so, experience suggests that such checkoffs can be “fantasy activities”—showing that immediately after training, and in a quiet environment, a clinician can demonstrate performance of specific tasks doesn't necessarily correlate with skill with the device during actual use in challenging real-life conditions. Fortunately, most of the time, personnel do rapidly learn to use the essential aspects of equipment in their routine bedside activities.However, problems may arise especially for devices that are used only rarely (e.g., defibrillators), in situations requiring the use of advanced and complex device features, or when it is necessary to deal with unexpected glitches or faults (e.g., when something isn't hooked up quite right or the wrong button is accidentally pressed) in a stressful in physicians have been more resistant to to training, which is rarely made Thus, it is not for a physician to a device a anesthesia in patient having never or seen or used the healthcare, there is like the in aviation, of how much experience has as an they cannot an they have been specifically and as on that of In nuclear power, each plant has a of the control room on so it is that plant operators would be to control the reactor and systems if they are not with the this suggests that perhaps healthcare nuclear power have the optimal structure for In healthcare, it is and with little devices and systems are In nuclear power, there is strong control and little risk of by but at the cost of extreme and to especially in safety critical in so many there may be a in the Clearly, to its and physician autonomy and control by firms or but has to up to its for very high Nuclear power has an safety at least in the United States, but is, to a in its not of and other the two in many there are many of where of and may each industry to a that is more and at cost to the