BACKGROUND: Antiviral treatment is recommended for hospitalized patients with suspected and confirmed influenza, but evidence is limited among children. We evaluated the effect of antiviral treatment on hospital length of stay (LOS) among children hospitalized with influenza. METHODS: We included children <18 years hospitalized with laboratory-confirmed influenza in the US Influenza Hospitalization Surveillance Network. We collected data for 2 cohorts: 1 with underlying medical conditions not admitted to the ICU (n = 309, 2012-2013) and an ICU cohort (including children with and without underlying conditions; n = 299, 2010-2011 to 2012-2013). We used a Cox model with antiviral receipt as a time-dependent variable to estimate hazard of discharge and a Kaplan-Meier survival analysis to determine LOS. RESULTS: Compared with those not receiving antiviral agents, LOS was shorter for those treated ≤2 days after illness onset in both the medical conditions (adjusted hazard ratio: 1.37, P = .02) and ICU (adjusted hazard ratio: 1.46, P = .007) cohorts, corresponding to 37% and 46% increases in daily discharge probability, respectively. Treatment ≥3 days after illness onset had no significant effect in either cohort. In the medical conditions cohort, median LOS was 3 days for those not treated versus 2 days for those treated ≤2 days after symptom onset (P = .005). CONCLUSIONS: Early antiviral treatment was associated with significantly shorter hospitalizations in children with laboratory-confirmed influenza and high-risk medical conditions or children treated in the ICU. These results support Centers for Disease Control and Prevention recommendations for prompt empiric antiviral treatment in hospitalized patients with suspected or confirmed influenza.

INTRODUCTION: Several observational studies have shown decreases in measured influenza vaccine effectiveness (mVE) during influenza seasons. One study found decreases of 6%-11% per month during the 2011-12 to 2014-15 seasons. These findings could indicate waning immunity but could also occur if vaccine effectiveness is stable and vaccine provides partial protection in all vaccinees ("leaky") rather than complete protection in a subset of vaccinees. Since it is not known whether influenza vaccine is leaky, we simulated the 2011-12 to 2014-15 influenza seasons to estimate the potential contribution of leaky vaccine effect to the observed decline in mVE. METHODS: We used available data to estimate daily numbers of vaccinations and infections with A/H1N1, A/H3N2 and B viruses. We assumed that vaccine effect was leaky, calculated mVE as 1 minus the Mantel-Haenszel relative risk of vaccine on incident cases and determined the mean mVE change per 30 days since vaccination. Because change in mVE was highly dependent on infection rates, we performed simulations using low (15%) and high (31%) total (including symptomatic and asymptomatic) seasonal infection rates. RESULTS: For the low infection rate, decreases (absolute) in mVE per 30 days after vaccination were 2% for A/H1N1 and 1% for A/H3N2and B viruses. For high infection rate, decreases were 5% for A/H1N1, 4% for A/H3, and 3% for B viruses. CONCLUSIONS: The leaky vaccine bias could account for some, but probably not all of the observed intra-seasonal decreases in mVE. These results underscore the need for strategies to deal with intra-seasonal vaccine effectiveness decline.

INTRODUCTION: To evaluate the public health benefit of yearly influenza vaccinations, CDC estimates the number of influenza cases and hospitalizations averted by vaccine. Available input data on cases and vaccinations is aggregated by month and the estimation model is intentionally simple, raising concerns about the accuracy of estimates. METHODS: We created a synthetic dataset with daily counts of influenza cases and vaccinations, calculated "true" averted cases using a reference model applied to the daily data, aggregated the data by month to simulate data that would actually be available, and evaluated the month-level data with seven test methods (including the current method). Methods with averted case estimates closest to the reference model were considered most accurate. To examine their performance under varying conditions, we re-evaluated the test methods when synthetic data parameters (timing of vaccination relative to cases, vaccination coverage, infection rate, and vaccine effectiveness) were varied over wide ranges. Finally, we analyzed real (i.e., collected by surveillance) data from 2010 to 2017 comparing the current method used by CDC with the best-performing test methods. RESULTS: In the synthetic dataset (population 1 million persons, vaccination uptake 55%, seasonal infection risk without vaccination 12%, vaccine effectiveness 48%) the reference model estimated 28,768 averted cases. The current method underestimated averted cases by 9%. The two best test methods estimated averted cases with <1% error. These two methods also worked well when synthetic data parameters were varied over wide ranges (</=6.2% error). With the real data, these two methods estimated numbers of averted cases that are a median 8% higher than the currently-used method. CONCLUSIONS: We identified two methods for estimating numbers of influenza cases averted by vaccine that are more accurate than the currently-used algorithm. These methods will help us to better assess the benefits of influenza vaccination.

Background: The seasonal incidence of influenza is often approximated as "5% to 20%". Methods: We used two methods to estimate the seasonal incidence of symptomatic influenza in the United States. First, we made a statistical estimate extrapolated from influenza-associated hospitalization rates for 2010-11 to 2015-16, collected as part of national surveillance, covering approximately 9% of the United States, and including the existing mix of vaccinated and unvaccinated persons. Second, we performed a literature search and meta-analysis of published manuscripts that followed cohorts of subjects during 1996-2016 to detect laboratory-confirmed symptomatic influenza among unvaccinated persons; we adjusted this result to the United States median vaccination coverage and effectiveness during 2010-2016. Results: The statistical estimate of influenza incidence among all ages ranged from 3.0-11.3% among seasons, with median values of 8.3 (95% confidence interval [CI] 7.3%, 9.7%) for all ages, 9.3% (CI 8.2%, 11.1%) for children <18 years and 8.9% (CI 8.2%, 9.9%) for adults 18-64 years. Corresponding values for the meta-analysis were 7.1% (CI 6.1, 8.1) for all ages, 8.7% (6.6, 10.5) for children, and 5.1% (3.6, 6.6) for adults. Conclusions: The two approaches produced comparable results for children and persons of all ages. The statistical estimates are more versatile and permit estimation of season-to-season variation. During 2010-2016, the incidence of symptomatic influenza among vaccinated and unvaccinated United States residents, including both medically attended and non-attended infections, was approximately 8% and varied from 3% to 11% among seasons.

Hepatitis C virus (HCV) is transmitted primarily through contact with the blood of an infected person, and healthcare-associated HCV infection outbreaks are well documented. 1 The prevalence of HCV infection among patients with end-stage renal disease (ESRD) who receive hemodialysis (HD) is high (7.8%–14%).Reference Goodkin, Bieber, Gillespie, Robinson and Jadoul 2 HCV infection is an independent risk factor for death in HD patients and increases the chance of graft failure and death after renal transplantation.Reference Kalantar-Zadeh, Kilpatrick and McAllister 3 Although interferon injection therapies are poorly tolerated in HCV-infected HD patients, those who completed the therapies demonstrated improved survival.Reference Kalantar-Zadeh, Kilpatrick and McAllister 3 In 2014, all-oral curative HCV therapies became available specifically for patients with ESRD.Reference Roth, Nelson and Bruchfeld 4 | Testing is essential to detect HCV infection in high-risk populations. The Centers for Disease Control and Prevention (CDC) recommends routine screening of chronic HD patients for HCV antibody (anti-HCV) upon facility admission and every 6 months to identify possible dialysis-related transmission.Reference Alter, Lyerla and Tokars 5 Patients who test positive for anti-HCV should undergo confirmatory HCV nucleic acid testing (NAT) to detect HCV RNA.Reference Getchell, Wroblewski and DeMaria 6 HCV-infected patients should be evaluated for recommended care and treatment.Reference Alter, Lyerla and Tokars 5 | Testing has frequently identified HCV transmission and outbreaks in dialysis units that may otherwise be difficult to detect because of the typically asymptomatic nature of acute infection. Basic infection-control lapses are often identified in dialysis centers with HCV outbreaks.Reference Alter, Lyerla and Tokars 5 As a result, any case of new HCV infection in a patient undergoing HD should be reported to public health authorities to facilitate investigation and rapid correction of breaches. Rigorous adherence to recommended infection control practices is needed to protect patients and prevent outbreaks in this population. We analyzed data from outbreaks of HCV in dialysis facilities reported to the CDC for opportunities to reduce HCV prevalence in this population.

During March-May 2014, a Middle East respiratory syndrome (MERS) outbreak occurred in Jeddah, Saudi Arabia, that included many persons who worked or received medical treatment at King Fahd General Hospital. We investigated 78 persons who had laboratory-confirmed MERS during March 2-May 10 and documented contact at this hospital. The 78 persons with MERS comprised 53 patients, 16 healthcare workers, and 9 visitors. Among the 53 patients, the most probable sites of acquisition were the emergency department (22 patients), inpatient areas (17), dialysis unit (11), and outpatient areas (3). Infection control deficiencies included limited separation of suspected MERS patients, patient crowding, and inconsistent use of infection control precautions; aggressive improvements in these deficiencies preceded a decline in cases. MERS coronavirus transmission probably was multifocal, occurring in multiple hospital settings. Continued vigilance and strict application of infection control precautions are necessary to prevent future MERS outbreaks.

Middle East respiratory syndrome (MERS) coronavirus (MERS-CoV) is a novel respiratory pathogen first reported in 2012. During September 2014-January 2015, an outbreak of 38 cases of MERS was reported from 4 healthcare facilities in Taif, Saudi Arabia; 21 of the 38 case-patients died. Clinical and public health records showed that 13 patients were healthcare personnel (HCP). Fifteen patients, including 4 HCP, were associated with 1 dialysis unit. Three additional HCP in this dialysis unit had serologic evidence of MERS-CoV infection. Viral RNA was amplified from acute-phase serum specimens of 15 patients, and full spike gene-coding sequencing was obtained from 10 patients who formed a discrete cluster; sequences from specimens of 9 patients were closely related. Similar gene sequences among patients unlinked by time or location suggest unrecognized viral transmission. Circulation persisted in multiple healthcare settings over an extended period, underscoring the importance of strengthening MERS-CoV surveillance and infection-control practices.

OBJECTIVES: We sought to assess risk of Guillain-Barre syndrome (GBS) among influenza A (H1N1) 2009 monovalent (pH1N1) vaccinated and unvaccinated populations at the end of the 2009 pandemic. METHODS: We applied GBS surveillance data from a US population catchment area of 45 million from October 15, 2009, through May 31, 2010. GBS cases meeting Brighton Collaboration criteria were included. We calculated the incidence density ratio (IDR) among pH1N1 vaccinated and unvaccinated populations. We also estimated cumulative GBS risk using life table analysis. Additionally, we used vaccine coverage data and census population estimates to calculate denominators. RESULTS: There were 392 GBS cases; 64 (16%) occurred after pH1N1vaccination. The vaccinated population had lower average risk (IDR = 0.83, 95% confidence interval = 0.63, 1.08) and lower cumulative risk (6.6 vs 9.2 cases per million persons, P = .012) of GBS. CONCLUSIONS: Our findings suggest that at the end of the influenza season cumulative GBS risk was less among the pH1N1 vaccinated than the unvaccinated population, suggesting the benefit of vaccination as it relates to GBS. The observed potential protective effect on GBS attributed to vaccination warrants further study.

In response to the 2009 H1N1 pandemic and subsequent vaccination program, the Department of Health and Human Services and collaborators developed the Post-Licensure Rapid Immunization Safety Monitoring (PRISM) Program as a demonstration project to detect rare adverse events rapidly. The program monitored three million people who had received the H1N1 vaccine by linking data from large private health plans and from public immunization registries that had originally not been designed to share data, and on a larger scale than had been previously attempted. The program generated safety data in two weeks rather than three to six months-the standard time frame achievable using health plan data. PRISM substantially contributed to the understanding of the safety of H1N1 vaccines. Its use in the case of H1N1 highlights the necessity of proactive planning, scalable infrastructure, and public-private partnerships in tracking adverse events after vaccination in epidemics. It also illustrates how data could be integrated to produce policy-relevant information for other medical products.

BACKGROUND: After the Department of Defense implemented a mandatory anthrax vaccination program in 1998 concerns were raised about potential long-term safety effects of the current anthrax vaccine. The CDC multicenter, randomized, double-blind, placebo-controlled Anthrax Vaccine Adsorbed (AVA) Human Clinical Trial to evaluate route change and dose reduction collected data on participants' quality of life. Our objective is to assess the association between receipt of AVA and changes in health-related quality of life, as measured by the SF-36 health survey (Medical Outcomes Trust, Boston, MA), over 42 months after vaccination. METHODS: 1562 trial participants completed SF-36v2 health surveys at 0, 12, 18, 30 and 42 months. Physical and mental summary scores were obtained from the survey results. We used Generalized Estimating Equations (GEE) analyses to assess the association between physical and mental score difference from baseline and seven study groups receiving either AVA at each dose, saline placebo at each dose, or a reduced AVA schedule substituting saline placebo for some doses. RESULTS: Overall, mean physical and mental scores tended to decrease after baseline. However, we found no evidence that the score difference from baseline changed significantly differently between the seven study groups. CONCLUSIONS: These results do not favor an association between receipt of AVA and an altered health-related quality of life over a 42-month period.

OBJECTIVE: We sought to characterize reports to the Vaccine Adverse Event Reporting System (VAERS) of pregnant women who received tetanus toxoid, reduced diphtheria toxoid, and acellular pertussis vaccine (Tdap). STUDY DESIGN: We searched VAERS for reports of pregnant women who received Tdap from Jan. 1, 2005, through June 30, 2010. We conducted a clinical review of reports and available medical records. RESULTS: We identified 132 reports of Tdap administered to pregnant women; 55 (42%) described no adverse event (AE). No maternal or infant deaths were reported. The most frequent pregnancy-specific AE was spontaneous abortion in 22 (16.7%) reports. Injection site reactions were the most frequent non-pregnancy-specific AE found in 6 (4.5%) reports. One report with a major congenital anomaly (gastroschisis) was identified. CONCLUSION: During a time when Tdap was not routinely recommended in pregnancy, review of reports to VAERS in pregnant women after Tdap did not identify any concerning patterns in maternal, infant, or fetal outcomes.

PURPOSE: The Centers for Disease Control and Prevention Emerging Infections Program implemented active, population-based surveillance for Guillain-Barre syndrome (GBS) following H1N1 vaccines in 10 states/metropolitan areas. We report additional analyses of these data using self-controlled methods, which avoid potential confounding from person-level factors and co-morbidities. METHODS: Surveillance officers identified GBS cases with symptom onset during October 2009-April 2010 and ascertained receipt of H1N1 vaccines. We calculated self-controlled relative risks by comparing the number of cases with onset during a risk interval 1-42 days after vaccination with cases with onset during fixed (days 43-84) or variable (days 43-end of study period) control intervals. We calculated attributable risks by applying statistically significant relative risks to an independent estimate of GBS incidence. RESULTS: Fifty-nine GBS cases received H1N1 vaccine with or without seasonal vaccine. The relative risk was 2.1 (95%CI 1.2, 3.5) by the variable-window and 3.0 (95%CI 1.4, 6.4) by the fixed-window analyses. The corresponding attributable risks per million doses administered were 1.5 (95%CI 0.3, 3.4) and 2.8 (95%CI 0.6, 7.4). CONCLUSIONS: These attributable risks are similar to those of some previous formulations of seasonal influenza vaccine (about one to two cases per million doses administered), suggesting a low risk of GBS following the H1N1 vaccine that is not clearly higher than that of seasonal influenza vaccines. (Published 2012. This article is a US Government work and is in the public domain in the USA.)

BACKGROUND: On February 20, 2010, a 23 year old male Army Reservist (index case) with symptom onset 4h after receiving inactivated monovalent pandemic 2009 (H1N1) vaccine (MIV) was hospitalized with possible Guillain-Barre syndrome (GBS). Within 1-2 days, 13 reservists from the same unit presented to the emergency department and 14 filed Vaccine Adverse Event Reporting System (VAERS) reports of nonspecific symptoms following MIV. OBJECTIVES: To describe the spectrum of adverse events (AE) among reservists in the unit after MIV and to identify factors contributing to this cluster of reports. METHODS: We reviewed the reservists' VAERS reports and hospital records for demographics, influenza vaccination status, diagnostic results and outcome. All VAERS reports after vaccination from the same MIV lot were also screened. We conducted a survey of unit reservists to identify contributing factors for this cluster. RESULTS: The presumptive diagnosis of GBS in the index case was not confirmed. All other reservists demonstrated normal exam findings and laboratory investigations. VAERS reports following vaccination from the same MIV lot revealed no consistent pattern. Our survey of factors contributing to the cluster was returned by 55 reservists (response rate 28%). AEs following MIV were significantly more often reported by female and black reservists. There was a tendency for concern about the safety of the 2010-2011 seasonal influenza vaccine to be higher for reservists that reported an AE to MIV (p=0.13) or that sought medical attention for their symptoms (p=0.08). CONCLUSIONS: This cluster represents possible stimulated reporting following receipt of inactivated pandemic 2009 (H1N1) vaccine among service personnel.

BACKGROUND: In 2002 CDC initiated the Anthrax Vaccination Program (AVP) to provide voluntary pre-exposure vaccination with Anthrax Vaccine Adsorbed (AVA) for persons at high risk of exposure to Bacillus anthracis spores. There has been concern that AVA could be associated with long term impairment of mental and/or physical health. OBJECTIVES: To ascertain whether physical and mental functional status, as measured by the SF-36v2 health survey (Medical Outcomes Trust, Boston, MA), of AVA recipients and controls changed differently over time. METHODS: We enrolled 437 exposed (received AVA) and 139 control subjects. The exposed group received AVA under then-current Advisory Committee on Immunization Practices (ACIP) recommendations. SF-36v2 surveys were completed at 0, 12, and 30 months. SF-36v2 physical and mental scores both range from 0 to 100 with an estimated national average of 50 points. RESULTS: For physical scores, the average change from baseline was -0.53 for exposed vs. -0.67 for controls at 12 months (p=0.80) and -1.09 for exposed vs. -1.97 for controls at 30 months (p=0.23). For mental scores, the average change from baseline was -1.50 for exposed vs. -1.64 for controls at 12 months (p=0.86) and -2.11 for exposed vs. -0.24 for controls at 30 months (p=0.06). In multivariable analysis, the difference in mental score change between exposed vs. controls at 30 months was less pronounced (p=0.37) but other findings were similar to univariate analyses. CONCLUSIONS: These results do not favor an association between receipt of AVA and an altered health related quality of life over a 30-month period.

Thank you for your editorial in the December 6, 2010 issue titled “Developing the Next Generation of Vaccinologists” [1]. While we support your call for expanded formal vaccinology training, we also wish to point out that the Centers for Disease Control and Prevention (CDC)’s Immunization Safety Office (ISO) currently has two programs focused on mentoring and training in vaccine safety. | The oldest training opportunity is an ISO position within CDC's Epidemic Intelligence Service (EIS) program (http://www.cdc.gov/eis/index.html). EIS is a 2-year post-graduate training program of service and on-the-job learning for health professionals that provides “hands-on” practical training in epidemiology and public health [2], [3]. Over the past 15 years, 11 EIS Officers have received training in immunization safety through assignments with ISO.

AIMS/HYPOTHESIS: To evaluate whether vaccination increases the risk of type 1 diabetes mellitus in active component U.S. military personnel. METHODS: We conducted a retrospective cohort study among active component U.S. military personnel age 17-35 years. Individuals with first time diagnoses of type 1 diabetes between January 1, 2002 and December 31, 2008 were identified using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. We used Poisson regression to estimate risk ratios between individual vaccine exposures and type 1 diabetes. Secondary analyses were performed controlling for receipt of multiple vaccines and available demographic variables. RESULTS: Our study population consisted of 2,385,102 individuals followed for approximately 7,644,098 person-years of service. This included 1074 incident type 1 diabetes cases. We observed no significant increased risk of type 1 diabetes after vaccination with anthrax vaccine adsorbed (AVA) [RR=1.00; 95% CI (0.85, 1.17)], smallpox vaccine [RR=0.84; 95% (CI 0.70, 1.01)], typhoid vaccine [RR=1.03; 95% CI (0.87, 1.22)], hepatitis B vaccine [RR=0.83; 95% CI (0.72, 0.95)], measles mumps rubella vaccine (MMR) [RR=0.71, 95% CI (0.61, 0.83)], or yellow fever vaccine [RR=0.70; 95% CI (0.59, 0.82)]. CONCLUSIONS: We did not find an increased risk of diagnosed type 1 diabetes and any of the study vaccines. We recommend that follow-up studies using medical record review to confirm case status should be considered to corroborate these findings.

BACKGROUND: Adverse events occurring after vaccination are routinely reported to the Vaccine Adverse Event Reporting System (VAERS). We studied serious adverse events (SAEs) of a neurologic nature reported after receipt of influenza A (H1N1) 2009 monovalent vaccine during the 2009-2010 influenza season. Investigators in the Clinical Immunization Safety Assessment (CISA) network sought to characterize these SAEs and to assess their possible causal relationship to vaccination. METHODS: Centers for Disease Control and Prevention (CDC) and Food and Drug Administration (FDA) physicians reviewed all SAE reports (as defined by the Code of Federal Regulations, 21CFR section 314.80) after receipt of H1N1 vaccine reported to VAERS between October 1, 2009 and March 31, 2010. Non-fatal SAE reports with neurologic presentation were referred to CISA investigators, who requested and reviewed additional medical records and clinical information as available. CISA investigators assessed the causal relationship between vaccination and the event using modified WHO criteria as defined. RESULTS: 212 VAERS reports of non-fatal serious neurological events were referred for CISA review. Case reports were equally distributed by gender (50.9% female) with an age range of 6 months to 83 years (median 38 years). The most frequent diagnoses reviewed were: Guillain-Barre Syndrome (37.3%), seizures (10.8%), cranial neuropathy (5.7%), and acute disseminated encephalomyelitis (3.8%). Causality assessment resulted in classification of 72 events as "possibly" related (33%), 108 as "unlikely" related (51%), and 20 as "unrelated" (9%) to H1N1 vaccination; none were classified as "probable" or "definite" and 12 were unclassifiable (6%). CONCLUSION: The absence of a specific test to indicate whether a vaccine component contributes to the pathogenesis of an event occurring within a biologically plausible time period makes assessing causality difficult. The development of standardized protocols for providers to use in evaluation of adverse events following immunization, and rapid identification and follow-up of VAERS reports could improve causality assessment.

BACKGROUND: In 2004 the Clinical Consult Case Review (CCCR) working group was formed within the CDC-funded Clinical Immunization Safety Assessment (CISA) Network to review individual cases of adverse events following immunizations (AEFI). METHODS: Cases were referred by practitioners, health departments, or CDC employees. Vaccine Adverse Event Reporting System (VAERS) searches and literature reviews for similar cases were performed prior to review. After CCCR discussion, AEFI were assessed for a causal relationship with vaccination and recommendations regarding future immunizations were relayed back to the referring physicians. In 2010, surveys were sent to referring physicians to determine the utility and effectiveness of the CCCR service. RESULTS: CISA investigators reviewed 76 cases during 68 conference calls between April 2004 and December 2009. Almost half of the cases (35/76) were neurological in nature. Similar AEFI for the specific vaccines received were discovered for 63 cases through VAERS searches and for 38 cases through PubMed searches. Causality assessment using the modified WHO criteria resulted in classifying 3 cases as definitely related to vaccine administration, 12 as probably related, 16 as possibly related, 18 as unlikely related, 10 as unrelated, and 17 had insufficient information to assign causality. The physician satisfaction survey was returned by 30 (57.7%) of those surveyed and a majority of respondents (93.3%) felt that the CCCR service was useful. CONCLUSIONS: The CCCR provides advice about AEFI to practitioners, assigns potential causality, and contributes to an improved understanding of adverse health events following immunizations.

BACKGROUND: Automated surveillance systems require statistical methods to recognize increases in visit counts that might indicate an outbreak. In prior work we presented methods to enhance the sensitivity of C2, a commonly used time series method. In this study, we compared the enhanced C2 method with five regression models. METHODS: We used emergency department chief complaint data from US CDC BioSense surveillance system, aggregated by city (total of 206 hospitals, 16 cities) during 5/2008-4/2009. Data for six syndromes (asthma, gastrointestinal, nausea and vomiting, rash, respiratory, and influenza-like illness) was used and was stratified by mean count (1-19, 20-49, 50 per day) into 14 syndrome-count categories. We compared the sensitivity for detecting single-day artificially-added increases in syndrome counts. Four modifications of the C2 time series method, and five regression models (two linear and three Poisson), were tested. A constant alert rate of 1% was used for all methods. RESULTS: Among the regression models tested, we found that a Poisson model controlling for the logarithm of total visits (i.e., visits both meeting and not meeting a syndrome definition), day of week, and 14-day time period was best. Among 14 syndrome-count categories, time series and regression methods produced approximately the same sensitivity (<5% difference) in 6; in six categories, the regression method had higher sensitivity (range 6-14% improvement), and in two categories the time series method had higher sensitivity. DISCUSSION: When automated data are aggregated to the city level, a Poisson regression model that controls for total visits produces the best overall sensitivity for detecting artificially added visit counts. This improvement was achieved without increasing the alert rate, which was held constant at 1% for all methods. These findings will improve our ability to detect outbreaks in automated surveillance system data.

OBJECTIVE: Near real-time disease detection using electronic data sources is a public health priority. Detecting pneumonia is particularly important because it is the manifesting disease of several bioterrorism agents as well as a complication of influenza, including avian and novel H1N1 strains. Text radiology reports are available earlier than physician diagnoses and so could be integral to rapid detection of pneumonia. We performed a pilot study to determine which keywords present in text radiology reports are most highly associated with pneumonia diagnosis. DESIGN: Electronic radiology text reports from 11 hospitals from February 1, 2006 through December 31, 2007 were used. We created a computerized algorithm that searched for selected keywords ("airspace disease", "consolidation", "density", "infiltrate", "opacity", and "pneumonia"), differentiated between clinical history and radiographic findings, and accounted for negations and double negations; this algorithm was tested on a sample of 350 radiology reports. We used the algorithm to study 189,246 chest radiographs, searching for the keywords and determining their association with a final International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis of pneumonia. MEASUREMENTS: Performance of the search algorithm in finding keywords, and association of the keywords with a pneumonia diagnosis. RESULTS: In the sample of 350 radiographs, the search algorithm was highly successful in identifying the selected keywords (sensitivity 98.5%, specificity 100%). Analysis of the 189,246 radiographs showed that the keyword "pneumonia" was the strongest predictor of an ICD-9-CM diagnosis of pneumonia (adjusted odds ratio 11.8) while "density" was the weakest (adjusted odds ratio 1.5). In general, the most highly associated keyword present in the report, regardless of whether a less highly associated keyword was also present, was the best predictor of a diagnosis of pneumonia. CONCLUSION: Empirical methods may assist in finding radiology report keywords that are most highly predictive of a pneumonia diagnosis.

OBJECTIVE: To determine the feasibility of using electronic laboratory and admission-discharge-transfer data from BioSense, a national automated surveillance system, to apply new modified Clostridium difficile infection (CDI) surveillance definitions and calculate overall and facility-specific rates of disease. DESIGN: Retrospective, multicenter cohort study. SETTING: Thirty-four hospitals sending inpatient, emergency department, and/or outpatient data to BioSense. METHODS: Laboratory codes and text-parsing methods were used to extract C. difficile-positive toxin assay results from laboratory data sent to BioSense during the period from January 1, 2007, through June 30, 2008; these were merged with administrative records to determine whether cases were community associated or healthcare onset, as well as patient-day data for rate calculations. A patient was classified as having hospital-onset CDI if he or she had a C. difficile toxin-positive result on a stool sample collected 3 or more days after admission and community-onset CDI if the specimen was collected less than 3 days after admission or the patient was not hospitalized. RESULTS: A total of 4,585 patients from 34 hospitals in 12 states had C. difficile-positive assay results. More than half (53.0%) of the cases were community-onset, and 30.8% of these occurred in patients who were recently hospitalized. The overall rate of healthcare-onset CDI was 7.8 cases per 10,000 patient-days, with a range among facilities of 1.5-27.8 cases per 10,000 patient-days. CONCLUSIONS: Electronic laboratory data sent to the BioSense surveillance system were successfully used to produce disease rates of CDI comparable to those of other studies, which shows the feasibility of using electronic laboratory data to track a disease of public health importance.

OBJECTIVES: To identify and characterise clusters of emergency department (ED) visits for fall injuries during the 2007-2008 winter season. METHODS: Hospital ED chief complaints and diagnoses from hospitals reporting to the Centers for Disease Control and Prevention BioSense system were analysed. The authors performed descriptive analyses, used time series charts on data aggregated by metropolitan statistical areas (MSAs), and used SaTScan to find spatial-temporal clusters of visits from falls. RESULTS: In 2007-2008, 17 clusters of falls in 13 MSAs were found; the median number of excess ED visits for falls was 71 per day. SaTScan identified 11 clusters of falls, of which seven corresponded to MSA clusters found by time series and five included more than one state/district. Most clusters coincided with known periods of snowfall or freezing rain. CONCLUSION: The results show the role that a national automated system can play in tracking widespread injuries. Such a system could be harnessed to assist with prevention strategies.

Since its emergence early this year, the H1N1 2009 influenza A virus has achieved pandemic proportions. Vaccines have been produced by several manufacturers and mass-vaccination campaigns are underway in the USA and several other countries. Scrutiny of the safety of H1N1 vaccines is expected to be intense. With potentially hundreds of millions of people being vaccinated, adverse events will inevitably occur in some recently vaccinated people and the question will arise as to whether the vaccine was causative. | In The Lancet today, Steven Black and colleagues1 estimate the expected background rates of occurrence of several health conditions. The authors obtained rates from the literature and also from computerised health databases in several countries. They show that when millions of people are vaccinated, even rare health conditions can be expected to occur coincidentally shortly after vaccination. For example, among 10 million vaccinees, it can be predicted that about 22 will develop Guillain-Barré syndrome within 6 weeks of vaccination even if the vaccine does not increase the risk of the syndrome. Knowledge that this number of events would have occurred regardless of whether or not the individuals were vaccinated will be extremely important in providing an appropriate context in which to interpret adverse events after immunisation.

BACKGROUND: The Centers for Disease Control and Prevention's (CDC's) BioSense system provides near-real time situational awareness for public health monitoring through analysis of electronic health data. Determination of anomalous spatial and temporal disease clusters is a crucial part of the daily disease monitoring task. Our study focused on finding useful anomalies at manageable alert rates according to available BioSense data history. METHODS: The study dataset included more than 3 years of daily counts of military outpatient clinic visits for respiratory and rash syndrome groupings. We applied four spatial estimation methods in implementations of space-time scan statistics cross-checked in Matlab and C. We compared the utility of these methods according to the resultant background cluster rate (a false alarm surrogate) and sensitivity to injected cluster signals. The comparison runs used a spatial resolution based on the facility zip code in the patient record and a finer resolution based on the residence zip code. RESULTS: Simple estimation methods that account for day-of-week (DOW) data patterns yielded a clear advantage both in background cluster rate and in signal sensitivity. A 28-day baseline gave the most robust results for this estimation; the preferred baseline is long enough to remove daily fluctuations but short enough to reflect recent disease trends and data representation. Background cluster rates were lower for the rash syndrome counts than for the respiratory counts, likely because of seasonality and the large scale of the respiratory counts. CONCLUSION: The spatial estimation method should be chosen according to characteristics of the selected data streams. In this dataset with strong day-of-week effects, the overall best detection performance was achieved using subregion averages over a 28-day baseline stratified by weekday or weekend/holiday behavior. Changing the estimation method for particular scenarios involving different spatial resolution or other syndromes can yield further improvement.