Search

BACKGROUND. Influenza causes significant morbidity among children. Previous studies used indirect case ascertainment methods with little cost data. We sought to measure the burden of laboratory-confirmed influenza from hospitalized children. METHODS. We conducted a retrospective cohort study during 3 viral seasons at Primary Children's Medical Center (Salt Lake City, UT). Children [less than or equal to] 18 years of age who were hospitalized with laboratory-confirmed influenza infection were included. Outcomes included hospitalization rates, complications including intensive care unit stays, mechanical ventilation, length of stay, and total hospital costs. RESULTS. A total of 325 children had hospitalizations attributable to influenza over 3 viral seasons: 28% 2 years of age; 37% had high-risk medical conditions. Population-based rates of hospitalization for Salt Lake County residents ranged from 6.3 to 252.7 per 100 000 children. The highest rates were in children younger than 6 months, and rates decreased with increasing age. Forty-nine (15%) children had an ICU stay; 27 required mechanical ventilation, and half of these patients were >2 years of age. Total hospital cost for the cohort was $2 million; 55% was accounted for by children >2 years of age. Length of stay and total hospital costs were significantly higher in all children >2 years of age compared with children CONCLUSIONS. Proven influenza infection in children results in substantial hospital resource utilization and morbidity. Nationwide, the median hospital costs may total $55 million. Our data support the Advisory Committee on Immunization's recommendations to expand the use of influenza vaccine to children >2 years of age. Key Words influenza, children, resource utilization, complications, hospitalization, INFLUENZA IS RESPONSIBLE for seasonal epidemics of pediatric respiratory illness each year, resulting in substantial morbidity, mortality, and increased health care utilization and costs. Population-based studies in the United States [...]

Abstract Background Seasonal respiratory syncytial virus (RSV) epidemics occur annually in temperate climates and result in significant pediatric morbidity and increased health care costs. Although RSV epidemics generally occur between October and April, the size and timing vary across epidemic seasons and are difficult to predict accurately. Prediction of epidemic characteristics would support management of resources and treatment. Methods The goals of this research were to examine the empirical relationships among early exponential growth rate, total epidemic size, and timing, and the utility of specific parameters in compartmental models of transmission in accounting for variation among seasonal RSV epidemic curves. RSV testing data from Primary Children's Medical Center were collected on children under two years of age (July 2001-June 2008). Simple linear regression was used explore the relationship between three epidemic characteristics (final epidemic size, days to peak, and epidemic length) and exponential growth calculated from four weeks of daily case data. A compartmental model of transmission was fit to the data and parameter estimated used to help describe the variation among seasonal RSV epidemic curves. Results The regression results indicated that exponential growth was correlated to epidemic characteristics. The transmission modeling results indicated that start time for the epidemic and the transmission parameter co-varied with the epidemic season. Conclusions The conclusions were that exponential growth was somewhat empirically related to seasonal epidemic characteristics and that variation in epidemic start date as well as the transmission parameter over epidemic years could explain variation in seasonal epidemic size. These relationships are useful for public health, health care providers, and infectious disease researchers.

Abstract Background Respiratory Syncytial Virus (RSV), a major cause of bronchiolitis, has a large impact on the census of pediatric hospitals during outbreak seasons. Reliable prediction of the week these outbreaks will start, based on readily available data, could help pediatric hospitals better prepare for large outbreaks. Methods Naïve Bayes (NB) classifier models were constructed using weather data from 1985-2008 considering only variables that are available in real time and that could be used to forecast the week in which an RSV outbreak will occur in Salt Lake County, Utah. Outbreak start dates were determined by a panel of experts using 32,509 records with ICD-9 coded RSV and bronchiolitis diagnoses from Intermountain Healthcare hospitals and clinics for the RSV seasons from 1985 to 2008. Results NB models predicted RSV outbreaks up to 3 weeks in advance with an estimated sensitivity of up to 67% and estimated specificities as high as 94% to 100%. Temperature and wind speed were the best overall predictors, but other weather variables also showed relevance depending on how far in advance the predictions were made. The weather conditions predictive of an RSV outbreak in our study were similar to those that lead to temperature inversions in the Salt Lake Valley. Conclusions We demonstrate that Naïve Bayes (NB) classifier models based on weather data available in real time have the potential to be used as effective predictive models. These models may be able to predict the week that an RSV outbreak will occur with clinical relevance. Their clinical usefulness will be field tested during the next five years.

Over the past decade, outbreaks of new or reemergent viruses such as severe acute respiratory syndrome (SARS) virus, Middle East respiratory syndrome (MERS) virus, and Zika have claimed thousands of lives and cost governments and healthcare systems billions of dollars. Because the appearance of new or transformed diseases is likely to continue, the detection and characterization of emergent diseases is an important problem. We describe a Bayesian statistical model that can detect and characterize previously unknown and unmodeled diseases from patient-care reports and evaluate its performance on historical data. [ABSTRACT FROM AUTHOR]

Rationale: Nearly 60% of U.S. children live in counties with particulate matter less than or equal to 2.5 μm in aerodynamic diameter (PM2.5) concentrations above air quality standards. Understanding the relationship between ambient air pollution exposure and health outcomes informs actions to reduce exposure and disease risk.Objectives: To evaluate the association between ambient PM2.5 levels and healthcare encounters for acute lower respiratory infection (ALRI).Methods: Using an observational case-crossover design, subjects (n = 146,397) were studied if they had an ALRI diagnosis and resided on Utah's Wasatch Front. PM2.5 air pollution concentrations were measured using community-based air quality monitors between 1999 and 2016. Odds ratios for ALRI healthcare encounters were calculated after stratification by ages 0-2, 3-17, and 18 or more years.Measurements and Main Results: Approximately 77% (n = 112,467) of subjects were 0-2 years of age. The odds of ALRI encounter for these young children increased within 1 week of elevated PM2.5 and peaked after 3 weeks with a cumulative 28-day odds ratio of 1.15 per +10 μg/m3 (95% confidence interval, 1.12-1.19). ALRI encounters with diagnosed and laboratory-confirmed respiratory syncytial virus and influenza increased following elevated ambient PM2.5 levels. Similar elevated odds for ALRI were also observed for older children, although the number of events and precision of estimates were much lower.Conclusions: In this large sample of urban/suburban patients, short-term exposure to elevated PM2.5 air pollution was associated with greater healthcare use for ALRI in young children, older children, and adults. Further exploration is needed of causal interactions between PM2.5 and ALRI. [ABSTRACT FROM AUTHOR]

HEALTH INFORMATION TECHNOLOGY (HIT) promises to facilitate improvements in medical care. However, implementation methods are equally important, because both technology and its implementation change the culture and workflow in health [...]

Objectives: This study evaluates the accuracy and transferability of Bayesian case detection systems (BCD) that use clinical notes from emergency department (ED) to detect influenza cases. Methods: A BCD uses natural language processing (NLP) to infer the presence or absence of clinical findings from ED notes, which are fed into a Bayesain network classifier (BN) to infer patients’ diagnoses. We developed BCDs at the University of Pittsburgh Medical Center (BCDUPMC) and Intermountain Healthcare in Utah (BCDIH). At each site, we manually built a rule-based NLP and trained a Bayesain network classifier from over 40,000 ED encounters between Jan. 2008 and May. 2010 using feature selection, machine learning, and expert debiasing approach. Transferability of a BCD in this study may be impacted by seven factors: development (source) institution, development parser, application (target) institution, application parser, NLP transfer, BN transfer, and classification task. We employed an ANOVA analysis to study their impacts on BCD performance. Results: Both BCDs discriminated well between influenza and non-influenza on local test cases (AUCs > 0.92). When tested for transferability using the other institution’s cases, BCDUPMC discriminations declined minimally (AUC decreased from 0.95 to 0.94, p<0.01), and BCDIH discriminations declined more (from 0.93 to 0.87, p<0.0001). We attributed the BCDIH decline to the lower recall of the IH parser on UPMC notes. The ANOVA analysis showed five significant factors: development parser, application institution, application parser, BN transfer, and classification task. Conclusion: We demonstrated high influenza case detection performance in two large healthcare systems in two geographically separated regions, providing evidentiary support for the use of automated case detection from routinely collected electronic clinical notes in national influenza surveillance. The transferability could be improved by training Bayesian network classifier locally and increasing the accuracy of the NLP parser. [ABSTRACT FROM AUTHOR]

Background: Pediatric asthma affects 7.1 million American children incurring an annual total direct healthcare cost around 9.3 billion dollars. Asthma control in children is suboptimal, leading to frequent asthma exacerbations, excess costs, and decreased quality of life. Successful prediction of risk for asthma control deterioration at the individual patient level would enhance self-management and enable early interventions to reduce asthma exacerbations. We developed and tested the first set of models for predicting a child's asthma control deterioration one week prior to occurrence.Methods: We previously reported validation of the Asthma Symptom Tracker, a weekly asthma self-monitoring tool. Over a period of two years, we used this tool to collect a total of 2912 weekly assessments of asthma control on 210 children. We combined the asthma control data set with patient attributes and environmental variables to develop machine learning models to predict a child's asthma control deterioration one week ahead.Results: Our best model achieved an accuracy of 71.8 %, a sensitivity of 73.8 %, a specificity of 71.4 %, and an area under the receiver operating characteristic curve of 0.757. We also identified potential improvements to our models to stimulate future research on this topic.Conclusions: Our best model successfully predicted a child's asthma control level one week ahead. With adequate accuracy, the model could be integrated into electronic asthma self-monitoring systems to provide real-time decision support and personalized early warnings of potential asthma control deteriorations. [ABSTRACT FROM AUTHOR]

Context. Poison control centers (PCCs) and emergency departments (EDs) rely upon telephone communication to collaborate. PCCs and EDs each create electronic records for the same patient during the course of collaboration, but those electronic records are not shared. Objective. The purpose of this study was to describe the current, telephone based process of PCC-ED communication as the basis for potential process improvement. Materials and methods. This study was conducted at one PCC and two tertiary care EDs. We developed workflow diagrams to depict clinician descriptions of the current process, descriptions obtained through interviews of key informants. We also analyzed transcripts of phone calls between emergency departments and the poison control center, corresponding to a random sample of 120 PCC cases occurring January 1-December 31, 2011. Results. Collaboration between the ED and PCC takes place during multiple telephone calls, and the process is unsupported by shared documentation. The process occurs in three phases: notification, collaborative care, and ongoing consultation. In the ED, multiple care providers may communicate with the PCC, but only one ED care provider communicates with the poison control center specialist at a time. Handoffs occur for both ED and PCC. Collaborative care planning is common and most cases involve some type of request for information, whether vital signs, laboratory results, or verification that a treatment was administered. We found evidence of inefficiencies and safety vulnerabilities, including the inability of PCC specialists to reach ED care providers, telephone calls routed through multiple ED staff members in an attempt to reach the appropriate care provider, and exchange of clinical information with non-clinical staff. In 55% of cases, the patient was discharged prior to any synchronous telephone communication between the ED care provider and a PCC specialist. Ambiguous communication of information was observed in 22% of cases. In 12% of cases, a PCC specialist was unable to obtain requested information from the ED. Discussion and conclusion. Inefficiencies and vulnerabilities occur in telephone-based PCC-ED communication. Prudence begs consideration of alternative processes and models of ED-PCC communication and information sharing, including a process that supports collaboration with health information exchange. [ABSTRACT FROM AUTHOR]

Context. The US emergency departments and poison control centers use telephone communication to exchange information about poison exposed patients. Electronically exchanged patient information could better support care for poisoned patients by improving information availability for decision making and by decreasing unnecessary emergency department telephone interruptions. As federal initiatives push to increase clinical health information exchange (HIE), it is essential to assess the readiness of US poison control centers. We conducted a nationwide Delphi study to determine consensus on legal, operational, and clinical considerations that are important for electronic information exchange between emergency departments and poison control centers. Materials and methods. A national panel of US experts (n = 71) in emergency medicine and poison control participated in a Delphi study, September-December 2010. Panelists rated statements describing concepts related to implementation, adoption, or potential outcomes of electronic information exchange between emergency departments and poison control centers. The statements reflected panelist responses to initial open-ended questions and literature-based concepts. Results. A total of 71 panelists agreed to participate. The response rate for each round ranged from 0.73 to 0.77. Most (114/121) statements reached consensus. Seven statements failed to reach consensus. Panelists indicated that user involvement in the design of systems and tools is important. Workflow integration, safety, evidence of benefit, and outcomes are high-importance issues. Discussion/conclusions. Future research and development related to electronic information exchange should address high-importance issues: safety, patient outcomes, workflow integration, and evidence of benefit. It should also address key barriers: initial and ongoing costs associated with electronic information exchange, the absence of software and tools to facilitate exchange, and the need for training. Users should be involved in the design of an electronic information exchange process, and the process should support, not replace, verbal communication. [ABSTRACT FROM AUTHOR]

Background: Seasonal respiratory syncytial virus (RSV) epidemics occur annually in temperate climates and result in significant pediatric morbidity and increased health care costs. Although RSV epidemics generally occur between October and April, the size and timing vary across epidemic seasons and are difficult to predict accurately. Prediction of epidemic characteristics would support management of resources and treatment. Methods: The goals of this research were to examine the empirical relationships among early exponential growth rate, total epidemic size, and timing, and the utility of specific parameters in compartmental models of transmission in accounting for variation among seasonal RSV epidemic curves. RSV testing data from Primary Children's Medical Center were collected on children under two years of age (July 2001-June 2008). Simple linear regression was used explore the relationship between three epidemic characteristics (final epidemic size, days to peak, and epidemic length) and exponential growth calculated from four weeks of daily case data. A compartmental model of transmission was fit to the data and parameter estimated used to help describe the variation among seasonal RSV epidemic curves. Results: The regression results indicated that exponential growth was correlated to epidemic characteristics. The transmission modeling results indicated that start time for the epidemic and the transmission parameter co-varied with the epidemic season. Conclusions: The conclusions were that exponential growth was somewhat empirically related to seasonal epidemic characteristics and that variation in epidemic start date as well as the transmission parameter over epidemic years could explain variation in seasonal epidemic size. These relationships are useful for public health, health care providers, and infectious disease researchers. [ABSTRACT FROM AUTHOR]

Background: Respiratory Syncytial Virus (RSV), a major cause of bronchiolitis, has a large impact on the census of pediatric hospitals during outbreak seasons. Reliable prediction of the week these outbreaks will start, based on readily available data, could help pediatric hospitals better prepare for large outbreaks. Methods: Naïve Bayes (NB) classifier models were constructed using weather data from 1985-2008 considering only variables that are available in real time and that could be used to forecast the week in which an RSV outbreak will occur in Salt Lake County, Utah. Outbreak start dates were determined by a panel of experts using 32,509 records with ICD-9 coded RSV and bronchiolitis diagnoses from Intermountain Healthcare hospitals and clinics for the RSV seasons from 1985 to 2008. Results: NB models predicted RSV outbreaks up to 3 weeks in advance with an estimated sensitivity of up to 67% and estimated specificities as high as 94% to 100%. Temperature and wind speed were the best overall predictors, but other weather variables also showed relevance depending on how far in advance the predictions were made. The weather conditions predictive of an RSV outbreak in our study were similar to those that lead to temperature inversions in the Salt Lake Valley. Conclusions: We demonstrate that Naïve Bayes (NB) classifier models based on weather data available in real time have the potential to be used as effective predictive models. These models may be able to predict the week that an RSV outbreak will occur with clinical relevance. Their clinical usefulness will be field tested during the next five years. [ABSTRACT FROM AUTHOR]

In this retrospective cohort study, a clinical and administrative database of children hospitalized at Primary Children''s Medical Center, Salt Lake City, Utah, between January 1, 2002, and December 31, 2006, was used to identify those with laboratory-confirmed rotavirus infections and at least one seizure. In all, 59 children were identified, 34 of whom (58%) had no other potential medical explanation for their seizures. Of these 34 children, 23 (68%) were afebrile at seizure onset and 11 were febrile. Electroencephalography was performed for 21 of the 34 children (62%); all findings were normal, except for a child with slowing related to cerebral edema. Twenty-six of the 34 children (76%) had neuroimaging studies; all findings were normal, except for the child with cerebral edema and a child with an incidental arachnoid cyst. Twenty of the 34 children (59%) had a lumbar puncture; again, all findings were normal. All 34 children recovered uneventfully, including the 6 children who spent at least 1 day in an intensive care unit. Follow-up data on 27 of these children identified 2 children (7%) who required chronic anticonvulsant therapy. The results indicate that seizures associated with rotavirus infection are a relatively benign neurologic condition in young children. With few exceptions, neurodiagnostic studies do not influence management or outcome. [Copyright &y& Elsevier]

The 2017-18 U.S. influenza season was notable for its high severity, with approximately 45 million illnesses and 810,000 influenza-associated hospitalizations throughout the United States (1). The purpose of the investigation reported here was to create a state-level estimate of the number of persons in Utah who became ill with influenza disease during this severe national seasonal influenza epidemic and to create a sustainable system for making timely updates in future influenza seasons. Knowing the extent of influenza-associated illness can help public health officials, policymakers, and clinicians tailor influenza messaging, planning, and responses for seasonal influenza epidemics or during pandemics. Using national methods and existing influenza surveillance and testing data, the influenza burden (number of influenza illnesses, medical visits for influenza, and influenza-associated hospitalizations) in Utah during the 2016-17 and 2017-18 influenza seasons was estimated. During the 2016-17 season, an estimated 265,000 symptomatic illnesses affecting 9% of Utah residents occurred, resulting in 125,000 medically attended illnesses and 2,700 hospitalizations. During the 2017-18 season, an estimated 338,000 symptomatic illnesses affecting 11% of Utah residents occurred, resulting in 160,000 medically attended illnesses and 3,900 hospitalizations. Other state or county health departments could adapt similar methods in their jurisdictions to estimate the burden of influenza locally and support prompt public health activities., Competing Interests: All authors have completed and submitted the International Committee of Medical Journal Editors form for disclosure of potential conflicts of interest. Mr. McCaffrey reports grants from the Council of State and Territorial Epidemiologists during the conduct of the study. Dr. Pavia reports personal fees from Antimicrobial Therapy Inc, WebMD, Genentech, Merck, and Sequirius outside the submitted work. No other potential conflicts of interest were disclosed.

The prediction and characterization of outbreaks of infectious diseases such as influenza remains an open and important problem. This paper describes a framework for detecting and characterizing outbreaks of influenza and the results of testing it on data from ten outbreaks collected from two locations over five years. We model outbreaks with compartment models and explicitly model non-influenza influenza-like illnesses., (This is an Open Access article. Authors own copyright of their articles appearing in the Journal of Public Health Informatics. Readers may copy articles without permission of the copyright owner(s), as long as the author and OJPHI are acknowledged in the copy and the copy is used for educational, not-for-profit purposes.)

Background: Health care and public health professionals rely on accurate, real-time monitoring of infectious diseases for outbreak preparedness and response. Early detection of outbreaks is improved by systems that are comprehensive and specific with respect to the pathogen but are rapid in reporting the data. It has proven difficult to implement these requirements on a large scale while maintaining patient privacy., Objective: The aim of this study was to demonstrate the automated export, aggregation, and analysis of infectious disease diagnostic test results from clinical laboratories across the United States in a manner that protects patient confidentiality. We hypothesized that such a system could aid in monitoring the seasonal occurrence of respiratory pathogens and may have advantages with regard to scope and ease of reporting compared with existing surveillance systems., Methods: We describe a system, BioFire Syndromic Trends, for rapid disease reporting that is syndrome-based but pathogen-specific. Deidentified patient test results from the BioFire FilmArray multiplex molecular diagnostic system are sent directly to a cloud database. Summaries of these data are displayed in near real time on the Syndromic Trends public website. We studied this dataset for the prevalence, seasonality, and coinfections of the 20 respiratory pathogens detected in over 362,000 patient samples acquired as a standard-of-care testing over the last 4 years from 20 clinical laboratories in the United States., Results: The majority of pathogens show influenza-like seasonality, rhinovirus has fall and spring peaks, and adenovirus and the bacterial pathogens show constant detection over the year. The dataset can also be considered in an ecological framework; the viruses and bacteria detected by this test are parasites of a host (the human patient). Interestingly, the rate of pathogen codetections, on average 7.94% (28,741/362,101), matches predictions based on the relative abundance of organisms present., Conclusions: Syndromic Trends preserves patient privacy by removing or obfuscating patient identifiers while still collecting much useful information about the bacterial and viral pathogens that they harbor. Test results are uploaded to the database within a few hours of completion compared with delays of up to 10 days for other diagnostic-based reporting systems. This work shows that the barriers to establishing epidemiology systems are no longer scientific and technical but rather administrative, involving questions of patient privacy and data ownership. We have demonstrated here that these barriers can be overcome. This first look at the resulting data stream suggests that Syndromic Trends will be able to provide high-resolution analysis of circulating respiratory pathogens and may aid in the detection of new outbreaks., (©Lindsay Meyers, Christine C Ginocchio, Aimie N Faucett, Frederick S Nolte, Per H Gesteland, Amy Leber, Diane Janowiak, Virginia Donovan, Jennifer Dien Bard, Silvia Spitzer, Kathleen A Stellrecht, Hossein Salimnia, Rangaraj Selvarangan, Stefan Juretschko, Judy A Daly, Jeremy C Wallentine, Kristy Lindsey, Franklin Moore, Sharon L Reed, Maria Aguero-Rosenfeld, Paul D Fey, Gregory A Storch, Steve J Melnick, Christine C Robinson, Jennifer F Meredith, Camille V Cook, Robert K Nelson, Jay D Jones, Samuel V Scarpino, Benjamin M Althouse, Kirk M Ririe, Bradley A Malin, Mark A Poritz. Originally published in JMIR Public Health and Surveillance (http://publichealth.jmir.org), 06.07.2018.)

Objectives: This study evaluates the accuracy and portability of a natural language processing (NLP) tool for extracting clinical findings of influenza from clinical notes across two large healthcare systems. Effectiveness is evaluated on how well NLP supports downstream influenza case-detection for disease surveillance., Methods: We independently developed two NLP parsers, one at Intermountain Healthcare (IH) in Utah and the other at University of Pittsburgh Medical Center (UPMC) using local clinical notes from emergency department (ED) encounters of influenza. We measured NLP parser performance for the presence and absence of 70 clinical findings indicative of influenza. We then developed Bayesian network models from NLP processed reports and tested their ability to discriminate among cases of (1) influenza, (2) non-influenza influenza-like illness (NI-ILI), and (3) 'other' diagnosis., Results: On Intermountain Healthcare reports, recall and precision of the IH NLP parser were 0.71 and 0.75, respectively, and UPMC NLP parser, 0.67 and 0.79. On University of Pittsburgh Medical Center reports, recall and precision of the UPMC NLP parser were 0.73 and 0.80, respectively, and IH NLP parser, 0.53 and 0.80. Bayesian case-detection performance measured by AUROC for influenza versus non-influenza on Intermountain Healthcare cases was 0.93 (using IH NLP parser) and 0.93 (using UPMC NLP parser). Case-detection on University of Pittsburgh Medical Center cases was 0.95 (using UPMC NLP parser) and 0.83 (using IH NLP parser). For influenza versus NI-ILI on Intermountain Healthcare cases performance was 0.70 (using IH NLP parser) and 0.76 (using UPMC NLP parser). On University of Pisstburgh Medical Center cases, 0.76 (using UPMC NLP parser) and 0.65 (using IH NLP parser)., Conclusion: In all but one instance (influenza versus NI-ILI using IH cases), local parsers were more effective at supporting case-detection although performances of non-local parsers were reasonable.

Objectives: This study evaluates the accuracy and transferability of Bayesian case detection systems (BCD) that use clinical notes from emergency department (ED) to detect influenza cases., Methods: A BCD uses natural language processing (NLP) to infer the presence or absence of clinical findings from ED notes, which are fed into a Bayesain network classifier (BN) to infer patients' diagnoses. We developed BCDs at the University of Pittsburgh Medical Center (BCDUPMC) and Intermountain Healthcare in Utah (BCDIH). At each site, we manually built a rule-based NLP and trained a Bayesain network classifier from over 40,000 ED encounters between Jan. 2008 and May. 2010 using feature selection, machine learning, and expert debiasing approach. Transferability of a BCD in this study may be impacted by seven factors: development (source) institution, development parser, application (target) institution, application parser, NLP transfer, BN transfer, and classification task. We employed an ANOVA analysis to study their impacts on BCD performance., Results: Both BCDs discriminated well between influenza and non-influenza on local test cases (AUCs > 0.92). When tested for transferability using the other institution's cases, BCDUPMC discriminations declined minimally (AUC decreased from 0.95 to 0.94, p<0.01), and BCDIH discriminations declined more (from 0.93 to 0.87, p<0.0001). We attributed the BCDIH decline to the lower recall of the IH parser on UPMC notes. The ANOVA analysis showed five significant factors: development parser, application institution, application parser, BN transfer, and classification task., Conclusion: We demonstrated high influenza case detection performance in two large healthcare systems in two geographically separated regions, providing evidentiary support for the use of automated case detection from routinely collected electronic clinical notes in national influenza surveillance. The transferability could be improved by training Bayesian network classifier locally and increasing the accuracy of the NLP parser.

Background: Pediatric asthma affects 7.1 million American children incurring an annual total direct healthcare cost around 9.3 billion dollars. Asthma control in children is suboptimal, leading to frequent asthma exacerbations, excess costs, and decreased quality of life. Successful prediction of risk for asthma control deterioration at the individual patient level would enhance self-management and enable early interventions to reduce asthma exacerbations. We developed and tested the first set of models for predicting a child's asthma control deterioration one week prior to occurrence., Methods: We previously reported validation of the Asthma Symptom Tracker, a weekly asthma self-monitoring tool. Over a period of two years, we used this tool to collect a total of 2912 weekly assessments of asthma control on 210 children. We combined the asthma control data set with patient attributes and environmental variables to develop machine learning models to predict a child's asthma control deterioration one week ahead., Results: Our best model achieved an accuracy of 71.8 %, a sensitivity of 73.8 %, a specificity of 71.4 %, and an area under the receiver operating characteristic curve of 0.757. We also identified potential improvements to our models to stimulate future research on this topic., Conclusions: Our best model successfully predicted a child's asthma control level one week ahead. With adequate accuracy, the model could be integrated into electronic asthma self-monitoring systems to provide real-time decision support and personalized early warnings of potential asthma control deteriorations.

Unnecessarily broad-spectrum antibiotic prescribing for ambulatory pediatric urinary tract infection may result from clinicians not having antibiograms specific to this population. Comparing an existing hospital-based with a proposed ambulatory uropathogen antibiogram for children in Utah, Escherichia coli accounted for a larger percentage and was more susceptible to narrower-spectrum antibiotics, demonstrating the potential need for ambulatory pediatric antibiograms., (© The Author 2012. Published by Oxford University Press on behalf of the Pediatric Infectious Diseases Society. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.)