Dataset Information

Use of Multiple Imputation to Estimate the Proportion of Respiratory Virus Detections Among Patients Hospitalized With Community-Acquired Pneumonia.

ABSTRACT:

Background

Real-time polymerase chain reaction (PCR) on respiratory specimens and serology on paired blood specimens are used to determine the etiology of respiratory illnesses for research studies. However, convalescent serology is often not collected. We used multiple imputation to assign values for missing serology results to estimate virus-specific prevalence among pediatric and adult community-acquired pneumonia hospitalizations using data from an active population-based surveillance study.

Methods

Presence of adenoviruses, human metapneumovirus, influenza viruses, parainfluenza virus types 1-3, and respiratory syncytial virus was defined by positive PCR on nasopharyngeal/oropharyngeal specimens or a 4-fold rise in paired serology. We performed multiple imputation by developing a multivariable regression model for each virus using data from patients with available serology results. We calculated absolute and relative differences in the proportion of each virus detected comparing the imputed to observed (nonimputed) results.

Results

Among 2222 children and 2259 adults, 98.8% and 99.5% had nasopharyngeal/oropharyngeal specimens and 43.2% and 37.5% had paired serum specimens, respectively. Imputed results increased viral etiology assignments by an absolute difference of 1.6%-4.4% and 0.8%-2.8% in children and adults, respectively; relative differences were 1.1-3.0 times higher.

Conclusions

Multiple imputation can be used when serology results are missing, to refine virus-specific prevalence estimates, and these will likely increase estimates.

SUBMITTER: Bozio CH

PROVIDER: S-EPMC5890478 | biostudies-literature | 2018 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Use of Multiple Imputation to Estimate the Proportion of Respiratory Virus Detections Among Patients Hospitalized With Community-Acquired Pneumonia.

Bozio Catherine H CH Flanders W Dana WD Finelli Lyn L Bramley Anna M AM Reed Carrie C Gandhi Neel R NR Vidal Jorge E JE Erdman Dean D Levine Min Z MZ Lindstrom Stephen S Ampofo Krow K Arnold Sandra R SR Self Wesley H WH Williams Derek J DJ Grijalva Carlos G CG Anderson Evan J EJ McCullers Jonathan A JA Edwards Kathryn M KM Pavia Andrew T AT Wunderink Richard G RG Jain Seema S

Open forum infectious diseases 20180316 4

<h4>Background</h4>Real-time polymerase chain reaction (PCR) on respiratory specimens and serology on paired blood specimens are used to determine the etiology of respiratory illnesses for research studies. However, convalescent serology is often not collected. We used multiple imputation to assign values for missing serology results to estimate virus-specific prevalence among pediatric and adult community-acquired pneumonia hospitalizations using data from an active population-based surveillanc ...[more]

PMID: 29946553

Similar Datasets

Project description:ObjectivesSARS-CoV-2 may cause acute lung injury, and secondary infections are thus relevant complications in patients with COVID-19 pneumonia. However, detailed information on community- and hospital-acquired infections among patients with COVID-19 pneumonia is scarce.MethodsWe identified 220 SARS-CoV-2-positive patients hospitalized at the University Hospital Basel, Switzerland (between 25 February and 31 May 2020). We excluded patients who declined the general consent (n = 12), patients without clinical evidence of pneumonia (n = 29), and patients hospitalized for < 24 h (n = 17). We evaluated the frequency of community- and hospital-acquired infections using respiratory and blood culture materials with antigen, culture-based, and molecular diagnostics. For ICU patients, all clinical and microbial findings were re-evaluated interdisciplinary (intensive care, infectious disease, and clinical microbiology), and agreement reached to classify patients with infections.ResultsIn the final cohort of 162 hospitalized patients (median age 64.4 years (IQR, 50.4-74.2); 61.1% male), 41 (25.3%) patients were admitted to the intensive care unit, 34/41 (82.9%) required mechanical ventilation, and 17 (10.5%) of all hospitalized patients died. In total, 31 infections were diagnosed including five viral co-infections, 24 bacterial infections, and three fungal infections (ventilator-associated pneumonia, n = 5; tracheobronchitis, n = 13; pneumonia, n = 1; and bloodstream infection, n = 6). Median time to respiratory tract infection was 12.5 days (IQR, 8-18) and time to bloodstream infection 14 days (IQR, 6-30). Hospital-acquired bacterial and fungal infections were more frequent among ICU patients than other patients (36.6% vs. 1.7%). Antibiotic or antifungal treatment was administered in 71 (43.8%) patients.ConclusionsCommunity-acquired viral and bacterial infections were rare among COVID-19 pneumonia patients. By contrast, hospital-acquired bacterial or fungal infections were frequently complicating the course among ICU patients.

Project description:BackgroundCommunity acquired pneumonia (CAP) is a high incidence disease resulting in about 260,000 hospital admissions per year in Germany, more than myocardial infarction or stroke. Worldwide, CAP is the most frequent infectious disease with high lethality ranging from 1.2 % in those 20-29 years old to over 10 % in patients older than 70 years, even in industrial nations. CAP poses numerous medical challenges, which the PROGRESS (Pneumonia Research Network on Genetic Resistance and Susceptibility for the Evolution of Severe Sepsis) network aims to tackle: Operationalization of disease severity throughout the course of disease, outcome prediction for hospitalized patients and prediction of transitions from uncomplicated CAP to severe CAP, and finally, to CAP with sepsis and organ failure as a life-threatening condition. It is a major aim of PROGRESS to understand and predict patient heterogeneity regarding outcome in the hospital and to develop novel treatment concepts.MethodsPROGRESS was designed as a clinical, observational, multi-center study of patients with CAP requiring hospitalization. More than 1600 patients selected for low burden of co-morbidities have been enrolled, aiming at a total of 3000. Course of disease, along with therapy, was closely monitored by daily assessments and long-term follow-up. Daily blood samples allow in depth molecular-genetic characterization of patients. We established a well-organized workflow for sample logistics and a comprehensive data management system to collect and manage data from more than 50 study centers in Germany and Austria. Samples are stored in a central biobank and clinical data are stored in a central data base which also integrates all data from molecular assessments.DiscussionWith the PROGRESS study, we established a comprehensive data base of high quality clinical and molecular data allowing investigation of pressing research questions regarding CAP. In-depth molecular characterization will contribute to the discovery of disease mechanisms and establishment of diagnostic and predictive biomarkers. A strength of PROGRESS is the focus on younger patients with low burden of co-morbidities, allowing a more direct look at host biology with less confounding. As a resulting limitation, insights from PROGRESS will require validation in representative patient cohorts to assess clinical utility.Trial registrationThe PROGRESS study was retrospectively registered on May 24(th), 2016 with ClinicalTrials.gov: NCT02782013.

Project description:BackgroundRespiratory syncytial virus (RSV) is one of the most common virus causing community-acquired pneumonia (CAP) in children. To guide the prevention, diagnosis and treatment of RSV, we aimed to analyze the epidemiology of RSV in hospitalized children with CAP.MethodsA total of 9,837 hospitalized children (≤14 years old) with CAP from January 2010 to December 2019 were reviewed. Using the real-time polymerase chain reaction (RT-PCR), the oropharyngeal swab specimens were collected and tested for RSV, influenza virus A (INFA), influenza virus B (INFB), parainfluenza virus (PIV), enterovirus (EV), coronavirus (CoV), human metapneumovirus (HMPV), human bocavirus (HBoV), human rhinovirus (HRV), and adenovirus (ADV) for each patient.ResultsThe detection rate of RSV was 15.3% (1,507/9,837). From 2010 to 2019, the RSV detection rate showed a wavy change (χ2=166.982, P<0.001), with the highest detection rate in 2011 (158/636, 24.8%). RSV can be detected throughout the year, with the highest detection rate in February (123/482, 25.5%). Children younger than 0.5 years old had the highest detection rate (410/1,671, 24.5%). The detection rate of RSV in male children (1,024/6,226, 16.4%) was higher than that in female children (483/3,611, 13.4%) (P<0.001). A proportion of 17.7% (266/1,507) of RSV positive cases were also co-infected with other viruses, and INFA (41/266, 15.4%) was the most common coinfection virus. After adjusting for potential confounders, the RSV-positive children were associated with increased risk of severe pneumonia [odds ratio (OR) 1.26, 95% confidence interval (CI): 1.04 to 1.53, P=0.019]. Moreover, children with severe pneumonia had significantly lower cycle threshold (CT) values of RSV than those without severe pneumonia (28.88±3.89 vs. 30.42±3.33, P<0.01). Patients with coinfection (38/266, 14.3%) had a higher risk of severe pneumonia than those without coinfection (142/1,241, 11.4%), but the difference was not statistically significant (OR 1.39, 95% CI: 0.94 to 2.05, P=0.101).ConclusionsThe detection rate of RSV in CAP hospitalized children changed by years, months, ages, and sexes. CAP hospitalized children with RSV are more likely to develop severe pneumonia than those without RSV. Policy makers and doctors should make timely adjustments to prevention measures, medical resources and treatment options based on these epidemiological characteristics.

Dataset Information

Use of Multiple Imputation to Estimate the Proportion of Respiratory Virus Detections Among Patients Hospitalized With Community-Acquired Pneumonia.

Background

Methods

Results

Conclusions

Publications

Use of Multiple Imputation to Estimate the Proportion of Respiratory Virus Detections Among Patients Hospitalized With Community-Acquired Pneumonia.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets