National Healthcare Quality and Disparities Report
Latest available findings on quality of and access to health care
Data
- Data Infographics
- Data Visualizations
- Data Tools
- Data Innovations
- All-Payer Claims Database
- Healthcare Cost and Utilization Project (HCUP)
- Medical Expenditure Panel Survey (MEPS)
- AHRQ Quality Indicator Tools for Data Analytics
- State Snapshots
- United States Health Information Knowledgebase (USHIK)
- Data Sources Available from AHRQ
Search All Research Studies
AHRQ Research Studies Date
Topics
- Adverse Events (1)
- Cardiovascular Conditions (2)
- Clinical Decision Support (CDS) (1)
- Diagnostic Safety and Quality (1)
- Electronic Health Records (EHRs) (2)
- Healthcare-Associated Infections (HAIs) (1)
- (-) Health Information Technology (HIT) (6)
- Heart Disease and Health (2)
- Hospitals (1)
- Imaging (1)
- Mortality (2)
- Outcomes (1)
- (-) Risk (6)
- Sepsis (1)
- Surgery (2)
AHRQ Research Studies
Sign up: AHRQ Research Studies Email updates
Research Studies is a compilation of published research articles funded by AHRQ or authored by AHRQ researchers.
Results
1 to 6 of 6 Research Studies DisplayedLevy AE, Shah NR, Matheny ME
Determining post-test risk in a national sample of stress nuclear myocardial perfusion imaging reports: implications for natural language processing tools.
The authors investigated whether Natural Language Processing (NLP) tools could potentially help estimate myocardial perfusion imaging (MPI) risk. Subjects were VA patients who underwent stress MPI and coronary angiography 2009-11; stress test reports were randomly selected for analysis. The authors found that post-test ischemic risk was determinable but rarely reported in this sample of stress MPI reports. They conclude that this supports the potential use of NLP to help clarify risk and recommend further study of NLP in this context.
AHRQ-funded; HS022998.
Citation: Levy AE, Shah NR, Matheny ME .
Determining post-test risk in a national sample of stress nuclear myocardial perfusion imaging reports: implications for natural language processing tools.
J Nucl Cardiol 2019 Dec;26(6):1878-85. doi: 10.1007/s12350-018-1275-y..
Keywords: Imaging, Risk, Clinical Decision Support (CDS), Health Information Technology (HIT), Diagnostic Safety and Quality, Cardiovascular Conditions, Heart Disease and Health
Yang H, Tourani R, Zhu Y
Strategies for building robust prediction models using data unavailable at prediction time.
Risk prediction models based on pre- and intraoperative data have been proposed to assess the risk of HAIs at the end of the surgery, but the performance of these models lag behind HAI detection models based on postoperative data. Postoperative data are more predictive than pre- or interoperative data but it is unavailable when the risk models are applied (end of surgery). The objective of this study was to examine whether such data can be used to improve the performance of the risk model.
AHRQ-funded; HS024532.
Citation: Yang H, Tourani R, Zhu Y .
Strategies for building robust prediction models using data unavailable at prediction time.
J Am Med Inform Assoc 2021 Dec 28;29(1):72-79. doi: 10.1093/jamia/ocab229..
Keywords: Healthcare-Associated Infections (HAIs), Risk, Health Information Technology (HIT)
Fritz BA, Cui Z, Zhang M
Deep-learning model for predicting 30-day postoperative mortality.
The currently available prediction tools using summaries of intraoperative data are limited by their inability to reflect shifting risk associated with intraoperative physiological perturbations. In this study the investigators sought to compare similar benchmarks to a deep-learning algorithm predicting postoperative 30-day mortality. They concluded that a deep-learning time-series model improved prediction compared with models with simple summaries of intraoperative data.
AHRQ-funded; HS024581.
Citation: Fritz BA, Cui Z, Zhang M .
Deep-learning model for predicting 30-day postoperative mortality.
Br J Anaesth 2019 Nov;123(5):688-95. doi: 10.1016/j.bja.2019.07.025..
Keywords: Adverse Events, Health Information Technology (HIT), Mortality, Risk, Surgery
Hannan EL, Barrett SC, Samadashvili Z
Retooling of paper-based outcome measures to electronic format: comparison of the NY State public risk model and EHR-derived risk models for CABG mortality.
This study assessed the feasibility of retooling the paper-based New York State coronary artery bypass graft (CABG) surgery statistical model for mortality and readmission into a model for electronic health records (EHRs). Researchers found that only 6 data elements could be extracted from the EHR, and outlier hospitals differed for readmission but was usable for mortality. They concluded that the EHR model was inferior to the NYS model, and that simplifying the EHR risk model couldn’t capture most of the risk factors in the NYS model.
AHRQ-funded; HS022647.
Citation: Hannan EL, Barrett SC, Samadashvili Z .
Retooling of paper-based outcome measures to electronic format: comparison of the NY State public risk model and EHR-derived risk models for CABG mortality.
Med Care 2019 May;57(5):377-84. doi: 10.1097/mlr.0000000000001104..
Keywords: Surgery, Electronic Health Records (EHRs), Health Information Technology (HIT), Mortality, Outcomes, Risk, Cardiovascular Conditions
Delahanty RJ, Alvarez J, Flynn LM
Development and evaluation of a machine learning model for the early identification of patients at risk for sepsis.
In this study, the investigators aimed to use machine learning to develop a new sepsis screening tool, the Risk of Sepsis (RoS) score, and compare it with a slate of benchmark sepsis-screening tools, including the Systemic Inflammatory Response Syndrome, Sequential Organ Failure Assessment (SOFA), qSOFA, Modified Early Warning Score, and National Early Warning Score. The investigators concluded that in this retrospective study, RoS was more timely and discriminant than benchmark screening tools, including those recommend by the Sepsis-3 Task Force.
AHRQ-funded; HS024750.
Citation: Delahanty RJ, Alvarez J, Flynn LM .
Development and evaluation of a machine learning model for the early identification of patients at risk for sepsis.
Ann Emerg Med 2019 Apr;73(4):334-44. doi: 10.1016/j.annemergmed.2018.11.036..
Keywords: Health Information Technology (HIT), Hospitals, Risk, Sepsis
Taslimitehrani V, Dong G, Pereira NL
Developing EHR-driven heart failure risk prediction models using CPXR(Log) with the probabilistic loss function.
The authors proposed to apply a classification algorithm, Contrast Pattern Aided Logistic Regression (CPXR(Log)) with the probabilistic loss function, to develop and validate prognostic risk models to predict 1, 2, and 5 year survival in heart failure (HF). They found that the new loss function used in the algorithm outperforms other functions used in previous studies and that HF is a highly heterogeneous disease (different subgroups of patients require different types of considerations with their diagnosis and treatment). They concluded that logistic risk models often make systematic prediction errors and that it is prudent to use subgroup based prediction models such as those given by CPXR(Log) when investigating heterogeneous diseases.
AHRQ-funded; HS023077.
Citation: Taslimitehrani V, Dong G, Pereira NL .
Developing EHR-driven heart failure risk prediction models using CPXR(Log) with the probabilistic loss function.
J Biomed Inform 2016 Apr;60:260-9. doi: 10.1016/j.jbi.2016.01.009.
.
.
Keywords: Electronic Health Records (EHRs), Heart Disease and Health, Health Information Technology (HIT), Risk