Natural language processing for identification of refractory status epilepticus in children.

OBJECTIVE: Pediatric status epilepticus is one of the most frequent pediatric emergencies, with high mortality and morbidity. Utilizing electronic health records (EHRs) permits analysis of care approaches and disease outcomes at a lower cost than prospective research. However, reviewing EHR manually is time intensive. We aimed to compare refractory status epilepticus (rSE) cases identified by human EHR review with a natural language processing (NLP)-assisted rSE screen followed by a manual review. METHODS: We used the NLP screening tool Document Review Tool (DrT) to generate regular expressions, trained a bag-of-words NLP classifier on EHRs from 2017 to 2019, and then tested our algorithm on data from February to December 2012. We compared results from manual review to NLP-assisted search followed by manual review. RESULTS: Our algorithm identified 1528 notes in the test set. After removing notes pertaining to the same event by DrT, the user reviewed a total number of 400 notes to find patients with rSE. Within these 400 notes, we identified 31 rSE cases, including 12 new cases not found in manual review, and 19 of the 20 previously identified cases. The NLP-assisted model found 31 of 32 cases, with a sensitivity of 96.88% (95% CI = 82%-99.84%), whereas manual review identified 20 of 32 cases, with a sensitivity of 62.5% (95% CI = 43.75%-78.34%). SIGNIFICANCE: DrT provided a highly sensitive model compared to human review and an increase in patient identification through EHRs. The use of DrT is a suitable application of NLP for identifying patients with a history of recent rSE, which ultimately contributes to the implementation of monitoring techniques and treatments in near real time.

Bacteriology of pediatric breast abscesses beyond the neonatal period.

BACKGROUND: Limited data exist regarding the presentation and bacteriology of nonneonatal pediatric breast abscess. OBJECTIVE: To determine the bacteriology and characteristic presentation of pediatric breast abscesses in a tertiary care center. METHODS: Cross-sectional study of patients age 1 month to 21 years admitted to a pediatric Emergency Department (ED) between 1996 and 2018 with a breast abscess. Patients with pre-existing conditions were excluded. Records were reviewed to determine demographics, history, physical exam findings, wound culture results, imaging and ED disposition. We used descriptive statistics to describe prevalence of different bacteria. RESULTS: We identified 210 patients who met study criteria. Median age was 13.6 years [IQR 6.6, 17.4], and 91% (191/210) were females. Ninety-two patients (43.8%) were 'pre-treated' with antibiotics prior to ED visit, and 33/210 (16%) were febrile. Ultrasound was obtained in 85 patients (40.5%), 69 patients had a single abscess and 16 had multiple abscesses. Most patients were treated with antibiotics and 100 had a surgical intervention, of these 89 had I&D and 11 a needle aspiration. Admission rate was 45%. Culture results were available for 75 (75%). Thirty-three (44%) had a negative culture, or grew non-aureus staphylococci or other skin flora. Culture were positive for MSSA 21 (28%), MRSA 13 (17%), Proteus mirabilis 2 (2.6%), Serratia 1 (1.3%). Other organisms include Gram-negative bacilli, group A Streptococcus and enterococcus. CONCLUSIONS: Non-neonatal pediatric breast abscess bacteriology is no different than data published on other skin abscesses. MRSA coverage should be considered based on local prevalence in skin infections.

Is lymphangitic streaking associated with different pathogens?

OBJECTIVES: Little is known regarding the differences in microbiology associated with cellulitis or abscess with or without lymphangitic streaking. The objective of our study is to assess whether there are differences in the pathogens identified from wound cultures of patients with paronychia with and without associated lymphangitis. METHODS: Retrospective cross-sectional study at a tertiary pediatric emergency department over 25 years. We opted to assess patients with paronychia of the finger, assuming that these cases will have a greater variety of causative pathogens compared to other cases of cellulitis and soft tissue abscess that are associated with nail biting. Case identification was conducted using a computerized text-screening search that was refined by manual chart review. We included patients from 1 month to 20 years of age who underwent an incision and drainage (I&D) of a paronychia and had a culture obtained. The presence or absence of lymphangitis was determined from the clinical narrative in the medical record. We excluded patients treated with antibiotics prior to I&D as well as immune-compromised patients. We used descriptive statistics for prevalence and χ2 tests for categorical variables. RESULTS: Two hundred sixty-six patients met inclusion criteria. The median age was 9.7 years [IQR 4.7, 15.4] and 45.1% were female. Twenty-two patients (8.3%) had lymphangitic streaking associated with their paronychia. Patients with lymphangitis streaking were similar to those without lymphangitis in terms of age and sex (p = 0.52 and p = 0.82, respectively). Overall, the predominant bacteria was MSSA (40%) followed by MRSA (26%). No significant differences were found between the pathogens in the 22 patients with associated lymphangitis compared to the 244 patients without. CONCLUSION: Staphylococcus aureus represent the majority of pathogens in paronychia, although streptococcal species and gram-negative bacteria were also common. Among patients with paronychia of the finger, there seems to be no association between pathogen type and presence of lymphangitic streaking.

Utility of Blood Cultures and Empiric Antibiotics in Febrile Pediatric Hemophilia Patients With Central Venous Access Devices.

BACKGROUND: Children with hemophilia frequently require long-term central venous access devices (CVADs) for regular infusion of factor products. Hemophilia patients are not immunocompromised, but the presence and use of CVADs are associated with infections including bacteremia. Currently, the utility of blood cultures in evaluation of the febrile hemophilia patient with an indwelling CVAD is unknown, nor is optimal empiric antibiotic use. METHODS: We performed a retrospective cross-sectional study of febrile immunocompetent hemophilia patients with CVADs presenting to a large academic urban pediatric emergency department from 1995 to 2017. We used a natural language processing electronic search, followed by manual chart review to construct the cohort. We analyzed rate of pathogen recovery from cultures of blood in subgroups of hemophilia patients, the pathogen profile, and the reported pathogen susceptibilities to ceftriaxone. RESULTS: Natural language processing electronic search identified 181 visits for fever among hemophilia patients with indwelling CVADs of which 147 cases from 44 unique patients met study criteria. Cultures of blood were positive in 56 (38%) of 147 patients (95% confidence interval, 30%-47%). Seventeen different organisms were isolated (10 pathogens and 7 possible pathogens) with Staphylococcus aureus and coagulase-negative Staphylococcus species as the most common. Thirty-four percent of isolates were reported as susceptible to ceftriaxone. Positive blood cultures were more common in cases involving patients with inhibitors (n = 71) versus those without (n = 76), odds ratio, 7.4 (95% confidence interval, 3.5-15.9). This was observed irrespective of hemophilia type. CONCLUSIONS: Febrile immunocompetent hemophilia patients with indwelling CVADs have high rates of bacteremia. Empiric antimicrobial therapy should be targeted to anticipated pathogens and take into consideration local susceptibility patterns for Staphylococcus aureus.

Variation in the Presentation of Intussusception by Age.

OBJECTIVE: To compare the clinical presentation of intussusception among children younger and older than 24 months of age. DESIGN/METHODS: We performed a retrospective cross-sectional cohort study of children treated in the emergency department, aged 1 month to 6 years, who had an abdominal ultrasound to evaluate for intussusception over a 5-year period. After stratifying by an age cut-point of 24 months, univariate and multivariate analyses were performed. RESULTS: One thousand two hundred fifty-eight cases of suspected intussusception were studied; median age was 1.7 years (interquartile range, 0.8, 2.9 years), and 37% were female. Intussusception was identified in 176 children (14%); 153 (87%) were ileocolic, and 23 were ileoileal. Abdominal pain (odds ratio, 4.0; 95% confidence interval [CI], 1.5-10.5), emesis (OR, 3.5; 95% CI, 1.8-6.7), bilious emesis (OR, 2.9; 95% CI, 1.5-5.7), lethargy (OR, 2.3; 95% CI, 1.3-5.7), rectal bleeding (OR, 2.8; 95% CI, 1.4-5.7), and irritability (OR, 0.4; 95% CI, 0.2-0.8) were found to be predictors in those younger than 24 months. In children older than 24 months, male sex was the only predictor identified (OR, 2.0; 95% CI, 1.1-3.7). In cases where abdominal radiographs were obtained (n = 1212), any abnormality on abdominal radiograph was found to be predictive in both age groups (OR, 7.8; 95% CI, 3.8-25.7; and OR, 3.1; 95% CI, 1.8-5.2, respectively). CONCLUSIONS: Intussusception presents differently in children younger than 24 months compared with older children. "Traditional" clinical predictors of intussusception should be interpreted with caution when assessing children older than 2 years.

Predictors of a drainable suppurative adenitis among children presenting with cervical adenopathy.

OBJECTIVES: We sought to identify predictors for a drainable suppurative adenitis [DSA] among patients presenting with acute cervical lymphadenitis. METHODS: A retrospective cross sectional study of all patients admitted to an urban pediatric tertiary care emergency department over a 15 year period. Otherwise healthy patients who underwent imaging for an evaluation of cervical lymphadenitis were included. Cases were identified using a text-search module followed by manual review. We excluded immunocompromised patients and those with lymphadenopathy felt to be not directly infected (i.e. reactive) or that was not acute (symptom duration >28 days). Data collected included: age, gender, duration of symptoms, highest recorded temperature, physical exam findings, laboratory and imaging results, and surgical findings. A DSA was defined as >1.5 cm in diameter on imaging. We performed binary logistic regression to determine independent clinical predictors of a DSA. RESULTS: Three hundred sixty-one patients met inclusion criteria. Three hundred six patients (85%) had a CT scan, 55 (15%) had an ultrasound and 33 (9%) had both. DSA was identified in 71 (20%) patients. Clinical features independently associated with a DSA included absence of clinical pharyngitis, WBC >15,000/mm3, age ≤3 years, anterior cervical chain location, largest palpable diameter on exam >3 cm and prior antibiotic treatment of >24 h. The presence of fever, skin erythema, or fluctuance on examination, was not found to be predictive of DSA. CONCLUSIONS: We identified independent predictors of DSA among children presenting with cervical adenitis. Risk can be stratified into risk groups based on these clinical features.

C-reactive protein or erythrocyte sedimentation rate results reliably exclude invasive bacterial infections.

BACKGROUND: Clinicians utilize inflammatory markers, including C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), to identify febrile children who may have an occult serious illness or infection. OBJECTIVES: Our objective was to determine the relationship between invasive bacterial infections (IBIs) and CRP and ESR in febrile children. METHODS: We performed a retrospective cross-sectional study of 1460 febrile children <21 years of age, who presented to a single Emergency Department (ED) between 2012 and 2014 for evaluation of fever of <14 days' duration, who had both CRP and ESR obtained. Our primary outcome was IBI, defined as growth of pathogenic bacteria from a culture of cerebrospinal fluid or blood. We reviewed all ED encounters that occurred within three days of the index visits for development of IBI. We examined the negative predictive value (NPV) of CRP and ESR for IBI. RESULTS: Of the 1460 eligible ED encounters, the median patient age was 5.3 years [interquartile range (IQR) 2.4-10.0 years] and 762 (50.4%) were hospitalized. The median duration of fever was 4 days (IQR 1-7 days). Overall, 20 had an IBI (20/1460; 1.4%, 95% confidence interval (CI) 0.9-2.1%). None of those with a normal CRP (NPV 273/273; 100%, 95% CI 98.6-100%) or a normal ESR (NPV 486/486; 100%, 95% CI 99.2-100%) had an IBI. CONCLUSIONS: In our cross-sectional study of febrile children, IBI was unlikely with either a normal CRP or ESR. Inflammatory markers could be used to assist clinical decision-making while awaiting results of bacterial cultures.

Should patients with complex febrile seizure be admitted for further management?

BACKGROUND AND AIMS: Children with first complex febrile seizure (CFS) are often admitted for observation. The goals of this study were 1) to assess the risk of seizure recurrence during admission, 2) to determine whether early EEG affects acute management. DESIGN/METHODS: We retrospectively reviewed a cohort of children 6-60months of age admitted from a Pediatric Emergency Department for first CFS over a 15year period. We excluded children admitted for supportive care of their febrile illness. Data extraction included age, gender, seizure features, laboratory and imaging studies, EEG, further seizures during admission, and antiepileptic drugs (AEDs) given. RESULTS: One hundred eighty three children met inclusion criteria. Seven patients had seizures during the admission (7/183 or 3.8%) Since 38 children were loaded with anti-epileptic medication during their visit, the adjusted rate is 7/145 or 4.8. Increased risk of seizure recurrence during admission was observed in children presenting with multiple seizures (P=0.005). EEG was performed in 104/183 children (57%) and led to change in management in one patient (1%, 95% C.I. 0.05-6%). Six of the 7 children with seizure had an EEG. The study was normal in 3 and findings in the other 2 did not suggest/predict further seizures during the admission. CONCLUSIONS: Children with first CFSs are at low risk for seizure recurrence during admission. Multiple seizures at presentation are associated with risk of early recurrence and may warrant an admission. EEG had limited effect on acute management and should not be an indication for admission.

Positive guaiac and bloody stool are poor predictors of intussusception.

BACKGROUND: Currant jelly stool is a late manifestation of intussusception and is rarely seen in clinical practice. Other forms of GI bleeding have not been thoroughly studied and little is known about their respective diagnostic values. OBJECTIVE: To assess the predictive value of GI bleeding (positive guaiac test, bloody stool and rectal bleeding in evaluation of intussusception. METHODS: We performed a retrospective cross-sectional study cohort of all children, ages 1month-6years of age, who had an abdominal ultrasound obtained evaluating for intussusception over 5year period. We identified intussusception if diagnosed by ultrasound, air-contrast enema or surgery. Univariate and a multivariate logistic regression analysis were performed. RESULTS: During the study period 1258 cases met the study criteria; median age was 1.7years (IQR 0.8, 2.9) and 37% were females. Overall 176 children had intussusception; 153 (87%) were ileo-colic and 23 were ileo-ileal. Univariate risk ratio and adjusted Odds ratio were 1.3 (95% CI, 0.8, 2.0) and 1.3 (0.7, 2.4) for positive guaiac test, 1.1 (0.6, 2.1) and 0.9 (0.3, 3.0) for bloody stool, and 1.7 (1.02, 2.8) and 1.3 (0.5, 3.1) for rectal bleeding . CONCLUSION: Blood in stool, whether visible or tested by guaiac test has poor diagnostic performance in the evaluation of intussusception and is not independently predictive of intussusception. If the sole purpose of a rectal exam in these patients is for guaiac testing it should be reconsidered.

Validating a Clinical Prediction Rule for Ventricular Shunt Malfunction.

OBJECTIVE: This study aims to validate a published ventricular shunt clinical prediction rule for the identification of children at low risk for ventricular shunt malfunction based on the absence of 3 high-risk clinical predictors (irritability, nausea or vomiting, and headache). METHODS: We identified children aged 21 years and younger with a ventricular shunt who presented between 2010 and 2013 to a single pediatric emergency department (ED) for evaluation of potential shunt malfunction. We defined a ventricular shunt malfunction as obstruction to cerebrospinal fluid flow requiring operative neurosurgical intervention within 72 hours of initial ED evaluation. We applied this ventricular shunt clinical prediction rule to the study population and report the test characteristics. RESULTS: We identified 755 ED visits for 294 children with potential ventricular shunt malfunction. Of these encounters, 146 (19%; 95% confidence interval [CI], 17%-22%) had a ventricular shunt malfunction. The ventricular shunt clinical prediction rule had a sensitivity of 99% (95% CI, 94%-100%), specificity of 7% (95% CI, 5%-9%), and negative predictive value of 95% (95% CI, 82%-99%). Two children with a ventricular shunt malfunction were misclassified as low risk by this clinical prediction rule. CONCLUSIONS: Ventricular shunt malfunctions were common. Although children classified as low risk by the ventricular shunt clinical prediction rule were less likely to have a shunt malfunction, routine neuroimaging may still be required because exclusion of ventricular shunt malfunction may be difficult on clinical grounds alone.

Epidemiology and Predictors of Orbital Fractures in Children.

OBJECTIVES: Data regarding clinical predictors of orbital fractures in children are limited. We sought to describe the epidemiology of pediatric orbital fractures and identify the signs and symptoms of orbital fractures in children. METHODS: We performed a retrospective cohort study of children younger than 22 years who underwent a facial or orbital computed tomography (CT) scan to evaluate for orbital fracture. We included patients presenting to an emergency department of a tertiary care children's hospital between January 2009 and May 2013. The presence of an orbital fracture was assessed using the final interpretation of the CT by an attending radiologist in the electronic medical record. RESULTS: Among 326 children who underwent facial or orbital CT during the study period, 133 (41%) had an orbital fracture. The presence of nausea or vomiting, orbital tenderness, swelling, or ecchymosis and limitation of extraocular movement were each associated with the presence of an orbital fracture (P < 0.05 for each). Twenty-two children (6.7% of cohort and 16.5% of children with orbital fractures) underwent surgical repair for an orbital fracture. In the absence of orbital tenderness, swelling, or ecchymosis, there was 1 child with an orbital fracture who underwent operative intervention. CONCLUSIONS: Fewer than half of children undergoing CT have an orbital fracture identified, and fewer than 7% have a fracture that requires operative intervention. In the absence of orbital tenderness, swelling, or ecchymosis, only 1 child had an orbital fracture requiring surgical repair.

Utility of Lumbar Puncture in Children Presenting With Status Epilepticus.

OBJECTIVES: Because meningitis may trigger seizures, we sought to determine its frequency in children with first-time status epilepticus (SE). METHODS: We performed a retrospective cross-sectional study of children aged 1 month to 21 years who presented to a single pediatric emergency department between 1995 and 2012 with SE and who had a lumbar puncture (LP) performed as part of the diagnostic evaluation. We defined bacterial meningitis as a cerebrospinal fluid (CSF) culture positive for a bacterial pathogen or CSF pleocytosis (CSF white blood cells ≥10 cells/mm) with a blood culture positive for a bacterial pathogen. We defined viral meningitis or encephalitis using a positive enterovirus or herpes simplex virus polymerase chain reaction test. RESULTS: Among 126 children with SE who had an LP performed, 8 (6%) had CSF pleocytosis. Of these, 5 had received antibiotics before performance of a diagnostic LP. One child in the cohort was proven to have bacterial meningitis (0.8%; 95% confidence interval [CI], 0%-6%). Two other children had enteroviral meningitis (2/13 tested, 15%; 95% CI, 3%-51%), and 1 had a herpes simplex virus infection (1/47, 2%; 95% CI, 0%-15%). CONCLUSIONS: Bacterial meningitis is an uncommon cause of SE.

Febrile seizures: emergency medicine perspective.

PURPOSE OF REVIEW: The review describes current evidence on the evaluation of febrile seizures in the acute setting, the need for further outpatient assessment, and predictors regarding long-term outcomes of these patients. RECENT FINDINGS: New evidence has been added in support of limited assessment and intervention: evidence on low utility of lumbar puncture, emergent neuroimaging, and follow-up electroencephalography, as well as low yield for antipyretic prophylaxis and intermittent use of antiepileptic drugs. Finally, there is growing evidence regarding the genetic basis of both febrile seizures and vaccine-related seizures/febrile seizures. SUMMARY: Routine diagnostic testing for simple febrile seizures is being discouraged, and clear evidence-based guidelines regarding complex febrile seizures are lacking. Thus, clinical acumen remains the most important tool for identifying children with seizures who are candidates for a more elaborate diagnostic evaluation. Similarly, evidence and guidelines regarding candidates for an emergent out-of-hospital diazepam treatment are lacking.

Utility of initial EEG in first complex febrile seizure.

OBJECTIVE: The risk of developing epilepsy following febrile seizures (FS) varies between 2% and 10%, with complex febrile seizures (CFS) having a higher risk. We examined the utility of detected epileptiform abnormalities on the initial EEG following a first CFS in predicting subsequent epilepsy. METHODS: This was a retrospective study of consecutive patients (ages 6-60 months) who were neurologically healthy or mildly delayed, seen in the ED following a first CFS and had both an EEG and minimum of 2-year follow-up. Data regarding clinical characteristics, EEG report, development of subsequent epilepsy, and type of epilepsy were collected. Established clinical predictors for subsequent epilepsy in children with FS and EEG status were evaluated for potential correlation with the development of subsequent epilepsy. Sensitivity, specificity, and positive and negative predictive values of an abnormal EEG (epileptiform EEG) were calculated. RESULTS: A group of 154 children met our inclusion criteria. Overall, 20 (13%) children developed epilepsy. The prevalence of epilepsy was 13% (CI 8.3-19.6%). Epileptiform abnormalities were noted in 21 patients (13.6%), EEG slowing in 23 patients (14.9%), and focal asymmetry in six (3.8%). Epileptiform EEGs were noted in 20% (4/20) of patients with epilepsy and 13% (17/134) of patients without epilepsy (p=0.48). At an estimated risk of subsequent epilepsy of 10% (from population-based studies of children with FS), we determined that the PPV of an epileptiform EEG for subsequent epilepsy was 15%. None of the clinical variables (presence of more than 1 complex feature, family history of epilepsy, or status epilepticus) predicted epilepsy. CONCLUSIONS: An epileptiform EEG was not a sensitive measure and had a poor positive predictive value for the development of epilepsy among neurologically healthy or mildly delayed children with a first complex febrile seizure. The practice of obtaining routine EEG for predicting epilepsy after the first CFS needs clarification by well-defined prospective studies.

An Introduction to Natural Language Processing: How You Can Get More From Those Electronic Notes You Are Generating.

Electronically stored clinical documents may contain both structured data and unstructured data. The use of structured clinical data varies by facility, but clinicians are familiar with coded data such as International Classification of Diseases, Ninth Revision, Systematized Nomenclature of Medicine-Clinical Terms codes, and commonly other data including patient chief complaints or laboratory results. Most electronic health records have much more clinical information stored as unstructured data, for example, clinical narrative such as history of present illness, procedure notes, and clinical decision making are stored as unstructured data. Despite the importance of this information, electronic capture or retrieval of unstructured clinical data has been challenging. The field of natural language processing (NLP) is undergoing rapid development, and existing tools can be successfully used for quality improvement, research, healthcare coding, and even billing compliance. In this brief review, we provide examples of successful uses of NLP using emergency medicine physician visit notes for various projects and the challenges of retrieving specific data and finally present practical methods that can run on a standard personal computer as well as high-end state-of-the-art funded processes run by leading NLP informatics researchers.

Synovial fluid findings in children with knee monoarthritis in lyme disease endemic areas.

BACKGROUND: Although Lyme and septic arthritis of the knee may have similar clinical presentations, septic arthritis requires prompt identification and treatment to avoid joint destruction. We sought to determine whether synovial fluid cell counts alone can discriminate between Lyme, septic, and other inflammatory arthritis. METHODS: We conducted a retrospective cohort study of children aged 1 to 18 years with knee monoarthritis who presented to 1 of 2 pediatric emergency departments located in Lyme endemic areas. We included children who had both a synovial fluid culture and an evaluation for Lyme disease. Septic arthritis was defined as a positive synovial fluid culture or synovial fluid pleocytosis (white blood cell [WBC] ≥40,000 cells/µL) with a positive blood culture. Lyme arthritis was defined as positive Lyme serology without a positive bacterial culture. All other children were considered to have other inflammatory arthritis. We compared the synovial fluid counts by arthritis type. RESULTS: We identified 384 children with knee monoarthritis, of whom 19 (5%) had septic arthritis, 257 (67%) had Lyme arthritis and 108 (28%) had other inflammatory arthritis. Children with other inflammatory arthritis had lower synovial WBC and absolute neutrophil count, as well as percent neutrophils, than those with either Lyme or septic arthritis. There were no significant differences in the synovial fluid WBC, absolute neutrophil count, and percent neutrophils for children with Lyme and septic arthritis. CONCLUSIONS: In Lyme endemic areas, synovial fluid results alone do not differentiate septic from Lyme arthritis. Therefore, other clinical or laboratory indicators are needed to direct the care of patients with knee monoarthritis.

Enhancing Pressure Injury Surveillance Using Natural Language Processing.

OBJECTIVE: This study assessed the feasibility of nursing handoff notes to identify underreported hospital-acquired pressure injury (HAPI) events. METHODS: We have established a natural language processing-assisted manual review process and workflow for data extraction from a corpus of nursing notes across all medical inpatient and intensive care units in a tertiary care pediatric center. This system is trained by 2 domain experts. Our workflow started with keywords around HAPI and treatments, then regular expressions, distributive semantics, and finally a document classifier. We generated 3 models: a tri-gram classifier, binary logistic regression model using the regular expressions as predictors, and a random forest model using both models together. Our final output presented to the event screener was generated using a random forest model validated using derivation and validation sets. RESULTS: Our initial corpus involved 70,981 notes during a 1-year period from 5484 unique admissions for 4220 patients. Our interrater human reviewer agreement on identifying HAPI was high ( κ = 0.67; 95% confidence interval [CI], 0.58-0.75). Our random forest model had 95% sensitivity (95% CI, 90.6%-99.3%), 71.2% specificity (95% CI, 65.1%-77.2%), and 78.7% accuracy (95% CI, 74.1%-83.2%). A total of 264 notes from 148 unique admissions (2.7% of all admissions) were identified describing likely HAPI. Sixty-one described new injuries, and 64 describe known yet possibly evolving injuries. Relative to the total patient population during our study period, HAPI incidence was 11.9 per 1000 discharges, and incidence rate was 1.2 per 1000 bed-days. CONCLUSIONS: Natural language processing-based surveillance is proven to be feasible and high yield using nursing handoff notes.

Surveillance of Health Care-Associated Violence Using Natural Language Processing.

BACKGROUND AND OBJECTIVES: Patient and family violent outbursts toward staff, caregivers, or through self-harm, have increased during the ongoing behavioral health crisis. These health care-associated violence (HAV) episodes are likely under-reported. We sought to assess the feasibility of using nursing notes to identify under-reported HAV episodes. METHODS: We extracted nursing notes across inpatient units at 2 hospitals for 2019: a pediatric tertiary care center and a community-based hospital. We used a workflow for narrative data processing using a natural language processing (NLP) assisted manual review process performed by domain experts (a nurse and a physician). We trained the NLP models on the tertiary care center data and validated it on the community hospital data. Finally, we applied these surveillance methods to real-time data for 2022 to assess reporting completeness of new cases. RESULTS: We used 70 981 notes from the tertiary care center for model building and internal validation and 19 332 notes from the community hospital for external validation. The final community hospital model sensitivity was 96.8% (95% CI 90.6% to 100%) and a specificity of 47.1% (39.6% to 54.6%) compared with manual review. We identified 31 HAV episodes in July to December 2022, of which 26 were reportable in accordance with the hospital internal criteria. Only 7 of 26 cases were reported by employees using the self-reporting system, all of which were identified by our surveillance process. CONCLUSIONS: NLP-assisted review is a feasible method for surveillance of under-reported HAV episodes, with implementation and usability that can be achieved even at a low information technology-resourced hospital setting.

Magnet-related injury rates in children: a single hospital experience.

BACKGROUND AND OBJECTIVE: The ingestion of multiple magnets simultaneously or the placement of magnets in both nares can lead to serious injury resulting from the attraction of the magnets across the tissues. The impact of mandatory standards for toys containing magnets has not been thoroughly investigated. The aim of the present study was to describe the emergency department (ED) visit rate for magnet-related injuries. METHODS: We performed a retrospective study of children evaluated for magnet-related injuries from 1995 to 2012 in an urban tertiary care pediatric ED. We identified cases using a computerized text-search methodology followed by manual chart review. We included children evaluated for magnet ingestion or impaction in the ears, nose, vagina, or rectum. We assessed the type and number of magnets as well as management and required interventions. A Poisson regression model was used to analyze rates of injury over time. RESULTS: We identified 112 cases of magnet injuries. The median patient age was 6 years (IQR 3.5, 10), and 54% were male. Compared to before 2006, the rate for all magnet-related injuries in 2007-2012 (incidence rate ratio 3.44; 95% confidence interval 2.3-5.11) as well as multiple magnet-related injuries (incidence rate ratio 7.54; 95% confidence interval 3.51-16.19) increased. Swallowed magnets accounted for 86% of the injuries. Thirteen patients had endoscopy performed for magnet removal (12%), and 4 (4%) had a surgical intervention. Magnets from toys account for the majority of the injuries. CONCLUSIONS: The number of ED visits for magnet-related injuries in children may be rising and are underreported, with an increase in the proportion of multiple magnets involvement. In our case series, mandatory standard for toys had no mitigating effect.