RESUMO
STUDY OBJECTIVE: This study aimed to (1) develop and validate a natural language processing model to identify the presence of pulmonary embolism (PE) based on real-time radiology reports and (2) identify low-risk PE patients based on previously validated risk stratification scores using variables extracted from the electronic health record at the time of diagnosis. The combination of these approaches yielded an natural language processing-based clinical decision support tool that can identify patients presenting to the emergency department (ED) with low-risk PE as candidates for outpatient management. METHODS: Data were curated from all patients who received a PE-protocol computed tomography pulmonary angiogram (PE-CTPA) imaging study in the ED of a 3-hospital academic health system between June 1, 2018 and December 31, 2020 (n=12,183). The "preliminary" radiology reports from these imaging studies made available to ED clinicians at the time of diagnosis were adjudicated as positive or negative for PE by the clinical team. The reports were then divided into development, internal validation, and temporal validation cohorts in order to train, test, and validate an natural language processing model that could identify the presence of PE based on unstructured text. For risk stratification, patient- and encounter-level data elements were curated from the electronic health record and used to compute a real-time simplified pulmonary embolism severity (sPESI) score at the time of diagnosis. Chart abstraction was performed on all low-risk PE patients admitted for inpatient management. RESULTS: When applied to the internal validation and temporal validation cohorts, the natural language processing model identified the presence of PE from radiology reports with an area under the receiver operating characteristic curve of 0.99, sensitivity of 0.86 to 0.87, and specificity of 0.99. Across cohorts, 10.5% of PE-CTPA studies were positive for PE, of which 22.2% were classified as low-risk by the sPESI score. Of all low-risk PE patients, 74.3% were admitted for inpatient management. CONCLUSION: This study demonstrates that a natural language processing-based model utilizing real-time radiology reports can accurately identify patients with PE. Further, this model, used in combination with a validated risk stratification score (sPESI), provides a clinical decision support tool that accurately identifies patients in the ED with low-risk PE as candidates for outpatient management.
Assuntos
Serviço Hospitalar de Emergência , Processamento de Linguagem Natural , Embolia Pulmonar , Humanos , Embolia Pulmonar/diagnóstico por imagem , Embolia Pulmonar/diagnóstico , Masculino , Feminino , Pessoa de Meia-Idade , Angiografia por Tomografia Computadorizada , Registros Eletrônicos de Saúde , Medição de Risco/métodos , Idoso , Assistência Ambulatorial , Sistemas de Apoio a Decisões Clínicas , Adulto , Estudos RetrospectivosRESUMO
BACKGROUND: Clinical education across the professions is challenged by a lack of recognition for faculty and pressure for patient throughput and revenue generation. These pressures may reduce direct observation of patient care provided by students, a requirement for both billing student-involved services and assessing competence. These same pressures may also limit opportunities for interprofessional education and collaboration. METHODS: An interprofessional group of faculty collaborated in a sequential quality improvement project to identify the best patients and physical location for a student teaching clinic. Patient chief complaint, use of resources, length of stay, estimated severity of illness and student participation and evaluation of the clinic was tracked. RESULTS: Clinic Optimization and Patient Care: Five hundred and thirty-two emergency department (ED) patients were seen in the first 19 months of the clinic. A clinic located near the ED allowed for patients with higher emergency severity index and greater utilization of imaging. Patients had similar or lower lengths of stay and higher satisfaction than patients who remained in the ED (p < 0.0001). In the second clinic location, from October 2016-June 2019, 644 patients were seen with a total of 667 concerns; the most common concern was musculoskeletal (50.1%). Student Interprofessional Experience: A total of 991 students participated in the clinic: 68.3% (n = 677) medical students, 10.1% (n = 100) physician assistant students, 9.7% (n = 96) undergraduate nursing students, 9.1% (n = 90) physical therapy students, and 2.8% (n = 28) nurse practitioner students. The majority (74.5%, n = 738) of student participants worked with students from other professions. More than 90% of students reported that faculty set a positive learning environment respectful of students. However, 20% of students reported that faculty could improve provision of constructive feedback. Direct Observation: Direct observation of core entrustable professional activities for medical students was possible. Senior medical students were more likely to be observed generating a differential diagnosis or management plan than first year medical students. CONCLUSIONS: Creation of a DOCENT clinic in the emergency department provided opportunities for interprofessional education and observation of student clinical skills, enriching student experience without compromising patient care.
Assuntos
Bacharelado em Enfermagem , Estudantes de Medicina , Estudantes de Enfermagem , Serviço Hospitalar de Emergência , Retroalimentação , Ocupações em Saúde , Humanos , Relações InterprofissionaisRESUMO
Medical errors are commonly multifactorial, with adverse clinical consequences often requiring the simultaneous failure of a series of protective layers, termed the Swiss Cheese model. Remedying and preventing future medical errors requires a series of steps, including detection, mitigation of patient harm, disclosure, reporting, root cause analysis, system modification, regulatory action, and engineering and manufacturing reforms. We describe this process applied to two cases of improper orientation of a Heimlich valve in a thoracostomy tube system, resulting in enlargement of an existing pneumothorax and the development of radiographic features of tension pneumothorax. We analyse elements contributing to the occurrence of the error and depict the implementation of reforms within our healthcare system and with regulatory authorities and the manufacturer. We identify features of the Heimlich valve promoting this error and suggest educational, design, and regulatory reforms for enhanced patient safety.