Dataset Information

Variable selection and prediction using a nested, matched case-control study: Application to hospital acquired pneumonia in stroke patients.

ABSTRACT: Matched case-control designs are commonly used in epidemiologic studies for increased efficiency. These designs have recently been introduced to the setting of modern imaging and genomic studies, which are characterized by high-dimensional covariates. However, appropriate statistical analyses that adjust for the matching have not been widely adopted. A matched case-control study of 430 acute ischemic stroke patients was conducted at Massachusetts General Hospital (MGH) in order to identify specific brain regions of acute infarction that are associated with hospital acquired pneumonia (HAP) in these patients. There are 138 brain regions in which infarction was measured, which introduce nearly 10,000 two-way interactions, and challenge the statistical analysis. We investigate penalized conditional and unconditional logistic regression approaches to this variable selection problem that properly differentiate between selection of main effects and of interactions, and that acknowledge the matching. This neuroimaging study was nested within a larger prospective study of HAP in 1915 stroke patients at MGH, which recorded clinical variables, but did not include neuroimaging. We demonstrate how the larger study, in conjunction with the nested, matched study, affords us the capability to derive a score for prediction of HAP in future stroke patients based on imaging and clinical features. We evaluate the proposed methods in simulation studies and we apply them to the MGH HAP study.

SUBMITTER: Qian J

PROVIDER: S-EPMC3954429 | biostudies-literature | 2014 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Variable selection and prediction using a nested, matched case-control study: Application to hospital acquired pneumonia in stroke patients.

Qian Jing J Payabvash Seyedmehdi S Kemmling André A Lev Michael H MH Schwamm Lee H LH Betensky Rebecca A RA

Biometrics 20131209 1

Matched case-control designs are commonly used in epidemiologic studies for increased efficiency. These designs have recently been introduced to the setting of modern imaging and genomic studies, which are characterized by high-dimensional covariates. However, appropriate statistical analyses that adjust for the matching have not been widely adopted. A matched case-control study of 430 acute ischemic stroke patients was conducted at Massachusetts General Hospital (MGH) in order to identify speci ...[more]

PMID: 24320930

Similar Datasets

Project description:BackgroundPrognostic models of hospital-induced delirium, that include potential predisposing and precipitating factors, may be used to identify vulnerable patients and inform the implementation of tailored preventive interventions. It is recommended that, in prediction model development studies, candidate predictors are selected on the basis of existing knowledge, including knowledge from clinical practice. The purpose of this article is to describe the process of identifying and operationalizing candidate predictors of hospital-induced delirium for application in a prediction model development study using a practice-based approach.MethodsThis study is part of a larger, retrospective cohort study that is developing prognostic models of hospital-induced delirium for medical-surgical older adult patients using structured data from administrative and electronic health records. First, we conducted a review of the literature to identify clinical concepts that had been used as candidate predictors in prognostic model development-and-validation studies of hospital-induced delirium. Then, we consulted a multidisciplinary task force of nine members who independently judged whether each clinical concept was associated with hospital-induced delirium. Finally, we mapped the clinical concepts to the administrative and electronic health records and operationalized our candidate predictors.ResultsIn the review of 34 studies, we identified 504 unique clinical concepts. Two-thirds of the clinical concepts (337/504) were used as candidate predictors only once. The most common clinical concepts included age (31/34), sex (29/34), and alcohol use (22/34). 96% of the clinical concepts (484/504) were judged to be associated with the development of hospital-induced delirium by at least two members of the task force. All of the task force members agreed that 47 or 9% of the 504 clinical concepts were associated with hospital-induced delirium.ConclusionsHeterogeneity among candidate predictors of hospital-induced delirium in the literature suggests a still evolving list of factors that contribute to the development of this complex phenomenon. We demonstrated a practice-based approach to variable selection for our model development study of hospital-induced delirium. Expert judgement of variables enabled us to categorize the variables based on the amount of agreement among the experts and plan for the development of different models, including an expert-model and data-driven model.

Project description:BackgroundSocial and hospital environmental factors that may be associated with hospital-acquired pneumonia (HAP) have not been evaluated. Comprehensive risk assessment for the incidence of HAP including sociodemographic, clinical, and hospital environmental factors was conducted using national health insurance claims data.MethodsThis is a population-based retrospective cohort study of adult patients who were hospitalized for more than 3 days from the Health Insurance Review and Assessment Service-National Inpatient Sample data between January 1, 2016 and December 31, 2018 in South Korea. Multivariable logistic regression analyses were conducted to identify the factors associated with the incidence of HAP.ResultsAmong the 512,278 hospitalizations, we identified 25,369 (5.0%) HAP cases. In multivariable analysis, well-known risk factors associated with HAP such as older age (over 70 vs. 20-29; adjusted odds ratio [aOR], 3.66; 95% confidence interval [CI] 3.36-3.99), male sex (aOR, 1.35; 95% CI 1.32-1.39), pre-existing lung diseases (asthma [aOR, 1.73; 95% CI 1.66-1.80]; chronic obstructive pulmonary disease [aOR, 1.62; 95% CI 1.53-1.71]; chronic lower airway disease [aOR, 1.79; 95% CI 1.73-1.85]), tube feeding (aOR, 3.32; 95% CI 3.16-3.50), suctioning (aOR, 2.34; 95% CI 2.23-2.47), positioning (aOR, 1.63; 95% CI 1.55-1.72), use of mechanical ventilation (aOR, 2.31; 95% CI 2.15-2.47), and intensive care unit admission (aOR, 1.29; 95% CI 1.22-1.36) were associated with the incidence of HAP. In addition, poverty (aOR, 1.08; 95% CI 1.04-1.13), general hospitals (aOR, 1.54; 95% CI 1.39-1.70), higher bed-to-nurse ratio (Grade ≥ 5; aOR, 1.45; 95% CI 1.32-1.59), higher number of beds per hospital room (6 beds; aOR, 3.08; 95% CI 2.77-3.42), and ward with caregiver (aOR, 1.19; 95% CI 1.12-1.26) were related to the incidence of HAP.ConclusionsThe incidence of HAP was associated with various sociodemographic, clinical, and hospital environmental factors. Thus, taking a comprehensive approach to prevent and treat HAP is important.

Project description:IntroductionCommunity-acquired pneumonia (CAP) and hospital-acquired pneumonia (HAP) are common complications in idiopathic inflammatory myopathy (IIM) patients, and are frequently associated with unfavorable outcome as well as prolonged antibiotic therapy. In this study, we intended to clarify whether clinical pulmonary infection score (CPIS) and multiple serum biomarkers are valuable in predicting unfavorable outcomes and prolonged antibiotic therapy in adult IIM patients complicated with CAP or HAP.MethodsData of IIM patients with CAP or HAP who were admitted to three tertiary centers from December 2010 to November 2019 were retrospectively collected. Cox proportional hazards regression analysis and logistic regression analysis were adopted to identify risk factors for unfavorable outcomes and prolonged antibiotic therapy in these patients. The predictive values of potential predictors were assessed using receiver operating characteristic analysis.ResultsThe mortality rate was 60.6% in 109 IIM patients complicated with CAP or HAP. Myositis Disease Activity Assessment Visual Analogue Scales (MYOACT) score, CPIS and timely adjustment to antibiotics based on drug susceptibility test (DST-based antibiotic) were significantly associated with long-term outcome in these patients. With an optimal cutoff value of 6.5 and area under the curve (AUC) of 0.813, CPIS was a more satisfying predictor compared with MYOACT score. The peak C-reactive protein (CRP) level, DST-based antibiotics, and complication of interstitial lung disease (ILD) were also significantly correlated with prolonged antibiotic therapy.ConclusionsIIM patients complicated with CAP or HAP frequently suffer from unfavorable outcomes. Compared with IIM disease activity, CPIS worked as a better predictor of outcome in these patients. Also, the peak CRP level during hospitalization might be valuable in predicting prolonged antibiotic therapy. The existence of ILD might impede early discontinuation of antibiotics. Timely adjustment to antibiotics based on drug susceptibility testing would decrease the mortality rate and reduce the incidence of prolonged antibiotic therapy.

Dataset Information

Variable selection and prediction using a nested, matched case-control study: Application to hospital acquired pneumonia in stroke patients.

Publications

Variable selection and prediction using a nested, matched case-control study: Application to hospital acquired pneumonia in stroke patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets