Improvement in Cardiovascular Risk Prediction with Electronic Health Records

Houston Methodist cardiovascular learning health system (CVD-LHS) registry: Methods for development and implementation of an automated electronic medical record-based registry using an informatics framework approach

Objectives

To investigate the potential value and feasibility of creating a listing system-wide registry of patients with at-risk and established Atherosclerotic Cardiovascular Disease (ASCVD) within a large healthcare system using automated data extraction methods to systematically identify burden, determinants, and the spectrum of at-risk patients to inform population health management. Additionally, the Houston Methodist Cardiovascular Disease Learning Health System (HM CVD-LHS) registry intends to create high-quality data-driven analytical insights to assess, track, and promote cardiovascular research and care.

Methods

We conducted a retrospective multi-center, cohort analysis of adult patients who were seen in the outpatient settings of a large healthcare system between June 2016 - December 2022 to create an EMR-based registry. A common framework was developed to automatically extract clinical data from the EMR and then integrate it with the social determinants of health information retrieved from external sources. Microsoft's SQL Server Management Studio was used for creating multiple Extract-Transform-Load scripts and stored procedures for collecting, cleaning, storing, monitoring, reviewing, auto-updating, validating, and reporting the data based on the registry goals.

Results

A real-time, programmatically deidentified, auto-updated EMR-based HM CVD-LHS registry was developed with ∼450 variables stored in multiple tables each containing information related to patient's demographics, encounters, diagnoses, vitals, labs, medication use, and comorbidities. Out of 1,171,768 adult individuals in the registry, 113,022 (9.6%) ASCVD patients were identified between June 2016 and December 2022 (mean age was 69.2 ± 12.2 years, with 55% Men and 15% Black individuals). Further, multi-level groupings of patients with laboratory test results and medication use have been analyzed for evaluating the outcomes of interest.

Conclusions

HM CVD-LHS registry database was developed successfully providing the listing registry of patients with established ASCVD and those at risk. This approach empowers knowledge inference and provides support for efforts to move away from manual patient chart abstraction by suggesting that a common registry framework with a concurrent design of data collection tools and reporting rapidly extracting useful structured clinical data from EMRs for creating patient or specialty population registries.

A systematic review of risk factors associated with depression and anxiety in cancer patients

Depression and anxiety are common comorbid conditions associated with cancer, however the risk factors responsible for the onset of depression and anxiety in cancer patients are not fully understood. Also, there is little clarity on how these factors may vary across the cancer phases: diagnosis, treatment and depression. We aimed to systematically understand and synthesise the risk factors associated with depression and anxiety during cancer diagnosis, treatment and survivorship. We focused our review on primary and community settings as these are likely settings where longer term cancer care is provided. We conducted a systematic search on PubMed, PsychInfo, Scopus, and EThOS following the PRISMA guidelines. We included cross-sectional and longitudinal studies which assessed the risk factors for depression and anxiety in adult cancer patients. Quality assessment was undertaken using the Newcastle-Ottawa assessment checklists. The quality of each study was further rated using the Agency for Healthcare Research and Quality Standards. Our search yielded 2645 papers, 21 of these were eligible for inclusion. Studies were heterogenous in terms of their characteristics, risk factors and outcomes measured. A total of 32 risk factors were associated with depression and anxiety. We clustered these risk factors into four domains using an expanded biopsychosocial model of health: cancer-specific, biological, psychological and social risk factors. The cancer-specific risk factors domain was associated with the diagnosis, treatment and survivorship phases. Multifactorial risk factors are associated with the onset of depression and anxiety in cancer patients. These risk factors vary across cancer journey and depend on factors such as type of cancer and individual profile of the patients. Our findings have potential applications for risk stratification in primary care and highlight the need for a personalised approach to psychological care provision, as part of cancer care.

Replicability and reproducibility of predictive models for diagnosis of depression among young adults using Electronic Health Records

BACKGROUND

Recent advances in machine learning combined with the growing availability of digitized health records offer new opportunities for improving early diagnosis of depression. An emerging body of research shows that Electronic Health Records can be used to accurately predict cases of depression on the basis of individual's primary care records. The successes of these studies are undeniable, but there is a growing concern that their results may not be replicable, which could cast doubt on their clinical usefulness.

METHODS

To address this issue in the present paper, we set out to reproduce and replicate the work by Nichols et al. (2018), who trained predictive models of depression among young adults using Electronic Healthcare Records. Our contribution consists of three parts. First, we attempt to replicate the methodology used by the original authors, acquiring a more up-to-date set of primary health care records to the same specification and reproducing their data processing and analysis. Second, we test models presented in the original paper on our own data, thus providing out-of-sample prediction of the predictive models. Third, we extend past work by considering several novel machine-learning approaches in an attempt to improve the predictive accuracy achieved in the original work.

RESULTS

In summary, our results demonstrate that the work of Nichols et al. is largely reproducible and replicable. This was the case both for the replication of the original model and the out-of-sample replication applying NRCBM coefficients to our new EHRs data. Although alternative predictive models did not improve model performance over standard logistic regression, our results indicate that stepwise variable selection is not stable even in the case of large data sets.

CONCLUSION

We discuss the challenges associated with the research on mental health and Electronic Health Records, including the need to produce interpretable and robust models. We demonstrated some potential issues associated with the reliance on EHRs, including changes in the regulations and guidelines (such as the QOF guidelines in the UK) and reliance on visits to GP as a predictor of specific disorders.

Prediction and diagnosis of depression using machine learning with electronic health records data: a systematic review

BACKGROUND

Depression is one of the most significant health conditions in personal, social, and economic impact. The aim of this review is to summarize existing literature in which machine learning methods have been used in combination with Electronic Health Records for prediction of depression.

METHODS

Systematic literature searches were conducted within arXiv, PubMed, PsycINFO, Science Direct, SCOPUS and Web of Science electronic databases. Searches were restricted to information published after 2010 (from 1st January 2011 onwards) and were updated prior to the final synthesis of data (27th January 2022).

RESULTS

Following the PRISMA process, the initial 744 studies were reduced to 19 eligible for detailed evaluation. Data extraction identified machine learning methods used, types of predictors used, the definition of depression, classification performance achieved, sample size, and benchmarks used. Area Under the Curve (AUC) values more than 0.9 were claimed, though the average was around 0.8. Regression methods proved as effective as more developed machine learning techniques.

LIMITATIONS

The categorization, definition, and identification of the numbers of predictors used within models was sometimes difficult to establish, Studies were largely Western Educated Industrialised, Rich, Democratic (WEIRD) in demography.

CONCLUSION

This review supports the potential use of machine learning techniques with Electronic Health Records for the prediction of depression. All the selected studies used clinically based, though sometimes broad, definitions of depression as their classification criteria. The reported performance of the studies was comparable to or even better than that found in primary care. There are concerns with generalizability and interpretability.

Smart Technologies used as Smart Tools in the Management of Cardiovascular Disease and their Future Perspective

Cardiovascular disease (CVD) is a leading cause of morbidity and mortality worldwide. The advent of smart technologies has significantly impacted the management of CVD, offering innovative tools and solutions to improve patient outcomes. Smart technologies have revolutionized and transformed the management of CVD, providing innovative tools to improve patient care, enhance diagnostics, and enable more personalized treatment approaches. These smart tools encompass a wide range of technologies, including wearable devices, mobile applications,3D printing technologies, artificial intelligence (AI), remote monitoring systems, and electronic health records (EHR). They offer numerous advantages, such as real-time monitoring, early detection of abnormalities, remote patient management, and data-driven decision-making. However, they also come with certain limitations and challenges, including data privacy concerns, technical issues, and the need for regulatory frameworks. In this review, despite these challenges, the future of smart technologies in CVD management looks promising, with advancements in AI algorithms, telemedicine platforms, and bio fabrication techniques opening new possibilities for personalized and efficient care. In this article, we also explore the role of smart technologies in CVD management, their advantages and disadvantages, limitations, current applications, and their smart future.

Evidence supporting the choice of a new cardiovascular risk equation for Australia

This article reviews the risk equations recommended for use in international cardiovascular disease (CVD) primary prevention guidelines and assesses their suitability for use in Australia against a set of a priori defined selection criteria. The review and assessment were commissioned by the National Heart Foundation of Australia on behalf of the Australian Chronic Disease Prevention Alliance to inform recommendations on CVD risk estimation as part of the 2023 update of the Australian CVD risk assessment and management guidelines. Selected international risk equations were assessed against eight selection criteria: development using contemporary data; inclusion of established cardiovascular risk factors; inclusion of ethnicity and deprivation measures; prediction of a broad selection of fatal and non-fatal CVD outcomes; population representativeness; model performance; external validation in an Australian dataset; and the ability to be recalibrated or modified. Of the ten risk prediction equations reviewed, the New Zealand PREDICT equation met seven of the eight selection criteria, and met additional usability criteria aimed at assessing the ability to apply the risk equation in practice in Australia.

The psychosis metabolic risk calculator (PsyMetRiC) for young people with psychosis: International external validation and site-specific recalibration in two independent European samples

Background

Cardiometabolic dysfunction is common in young people with psychosis. Recently, the Psychosis Metabolic Risk Calculator (PsyMetRiC) was developed and externally validated in the UK, predicting up-to six-year risk of metabolic syndrome (MetS) from routinely collected data. The full-model includes age, sex, ethnicity, body-mass index, smoking status, prescription of metabolically-active antipsychotic medication, high-density lipoprotein, and triglyceride concentrations; the partial-model excludes biochemical predictors.

Methods

To move toward a future internationally-useful tool, we externally validated PsyMetRiC in two independent European samples. We used data from the PsyMetab (Lausanne, Switzerland) and PAFIP (Cantabria, Spain) cohorts, including participants aged 16-35y without MetS at baseline who had 1-6y follow-up. Predictive performance was assessed primarily via discrimination (C-statistic), calibration (calibration plots), and decision curve analysis. Site-specific recalibration was considered.

Findings

We included 1024 participants (PsyMetab n=558, male=62%, outcome prevalence=19%, mean follow-up=2.48y; PAFIP n=466, male=65%, outcome prevalence=14%, mean follow-up=2.59y). Discrimination was better in the full- compared with partial-model (PsyMetab=full-model C=0.73, 95% C.I., 0.68-0.79, partial-model C=0.68, 95% C.I., 0.62-0.74; PAFIP=full-model C=0.72, 95% C.I., 0.66-0.78; partial-model C=0.66, 95% C.I., 0.60-0.71). As expected, calibration plots revealed varying degrees of miscalibration, which recovered following site-specific recalibration. PsyMetRiC showed net benefit in both new cohorts, more so after recalibration.

Interpretation

The study provides evidence of PsyMetRiC's generalizability in Western Europe, although further local and international validation studies are required. In future, PsyMetRiC could help clinicians internationally to identify young people with psychosis who are at higher cardiometabolic risk, so interventions can be directed effectively to reduce long-term morbidity and mortality.

Funding

NIHR Cambridge Biomedical Research Centre (BRC-1215-20014); The Wellcome Trust (201486/Z/16/Z); Swiss National Research Foundation (320030-120686, 324730- 144064, and 320030-173211); The Carlos III Health Institute (CM20/00015, FIS00/3095, PI020499, PI050427, and PI060507); IDIVAL (INT/A21/10 and INT/A20/04); The Andalusian Regional Government (A1-0055-2020 and A1-0005-2021); SENY Fundacion Research (2005-0308007); Fundacion Marques de Valdecilla (A/02/07, API07/011); Ministry of Economy and Competitiveness and the European Fund for Regional Development (SAF2016-76046-R and SAF2013-46292-R).For the Spanish and French translation of the abstract see Supplementary Materials section.

Ectopic Fat and Cardiac Health in People with HIV: Serious as a Heart Attack

PURPOSE OF REVIEW

This study aims to summarize knowledge of alterations in adipose tissue distribution among people with HIV (PWH), with a focus on the cardiac depot and how this relates to the known higher risk of cardiovascular disease in this unique population.

RECENT FINDINGS

Similar to the general population, cardiac fat depots mirror visceral adipose tissue in PWH. However, altered fat distribution, altered fat quality, and higher prevalence of enlarged epicardial adipose tissue depots are associated with increased coronary artery disease among PWH. Adipose tissue disturbances present in PWH ultimately contribute to increased risk of cardiovascular disease beyond traditional risk factors. Future research should aim to understand how regulating adipose tissue quantity and quality can modify cardiovascular risk.

Contextualizing National Policies Regulating Access to Low-Dose Aspirin in America and Europe Using the Full Report of a Transatlantic Patient Survey of Aspirin in Preventive Cardiology

Background Aspirin is widely administered to prevent cardiovascular disease (CVD). However, appropriate use of aspirin depends on patient understanding of its risks, benefits, and indications, especially where aspirin is available over the counter (OTC). Methods and Results We did a survey of patient-reported 10-year cardiovascular risk; aspirin therapy status; form of aspirin access (OTC versus prescription); and knowledge of the risks, benefits, and role of aspirin in CVD prevention. Consecutive adults aged ≥50 years with ≥1 cardiovascular risk factor attending outpatient clinics in America and Europe were recruited. We also systematically reviewed national policies regulating access to low-dose aspirin for CVD prevention. At each site, 150 responses were obtained (300 total). Mean±SD age was 65±10 years, 40% were women, and 41% were secondary prevention patients. More than half of the participants at both sites did not know (1) their own level of 10-year CVD risk, (2) the expected magnitude of reduction in CVD risk with aspirin, or (3) aspirin's bleeding risks. Only 62% of all participants reported that aspirin was routinely indicated for secondary prevention, whereas 47% believed it was routinely indicated for primary prevention (P=0.048). In America, 83.5% participants obtained aspirin OTC compared with 2.5% in Europe (P<0.001). Finally, our review of European national policies found only 2 countries where low-dose aspirin was available OTC. Conclusions Many patients have poor insight into their objectively calculated 10-year cardiovascular risk and do not know the risks, benefits, and role of aspirin in CVD prevention. Aspirin is mainly obtained OTC in America in contrast to Europe, where most countries restrict access to low-dose aspirin.

Accurate Prediction of Stroke for Hypertensive Patients Based on Medical Big Data and Machine Learning Algorithms: Retrospective Study

Background

Stroke risk assessment is an important means of primary prevention, but the applicability of existing stroke risk assessment scales in the Chinese population has always been controversial. A prospective study is a common method of medical research, but it is time-consuming and labor-intensive. Medical big data has been demonstrated to promote disease risk factor discovery and prognosis, attracting broad research interest.

Objective

We aimed to establish a high-precision stroke risk prediction model for hypertensive patients based on historical electronic medical record data and machine learning algorithms.

Methods

Based on the Shenzhen Health Information Big Data Platform, a total of 57,671 patients were screened from 250,788 registered patients with hypertension, of whom 9421 had stroke onset during the 3-year follow-up. In addition to baseline characteristics and historical symptoms, we constructed some trend characteristics from multitemporal medical records. Stratified sampling according to gender ratio and age stratification was implemented to balance the positive and negative cases, and the final 19,953 samples were randomly divided into a training set and test set according to a ratio of 7:3. We used 4 machine learning algorithms for modeling, and the risk prediction performance was compared with the traditional risk scales. We also analyzed the nonlinear effect of continuous characteristics on stroke onset.

Results

The tree-based integration algorithm extreme gradient boosting achieved the optimal performance with an area under the receiver operating characteristic curve of 0.9220, surpassing the other 3 traditional machine learning algorithms. Compared with 2 traditional risk scales, the Framingham stroke risk profiles and the Chinese Multiprovincial Cohort Study, our proposed model achieved better performance on the independent validation set, and the area under the receiver operating characteristic value increased by 0.17. Further nonlinear effect analysis revealed the importance of multitemporal trend characteristics in stroke risk prediction, which will benefit the standardized management of hypertensive patients.

Conclusions

A high-precision 3-year stroke risk prediction model for hypertensive patients was established, and the model's performance was verified by comparing it with the traditional risk scales. Multitemporal trend characteristics played an important role in stroke onset, and thus the model could be deployed to electronic health record systems to assist in more pervasive, preemptive stroke risk screening, enabling higher efficiency of early disease prevention and intervention.

Risk prediction of covid-19 related death and hospital admission in adults after covid-19 vaccination: national prospective cohort study

OBJECTIVES

To derive and validate risk prediction algorithms to estimate the risk of covid-19 related mortality and hospital admission in UK adults after one or two doses of covid-19 vaccination.

DESIGN

Prospective, population based cohort study using the QResearch database linked to data on covid-19 vaccination, SARS-CoV-2 results, hospital admissions, systemic anticancer treatment, radiotherapy, and the national death and cancer registries.

SETTINGS

Adults aged 19-100 years with one or two doses of covid-19 vaccination between 8 December 2020 and 15 June 2021.

MAIN OUTCOME MEASURES

Primary outcome was covid-19 related death. Secondary outcome was covid-19 related hospital admission. Outcomes were assessed from 14 days after each vaccination dose. Models were fitted in the derivation cohort to derive risk equations using a range of predictor variables. Performance was evaluated in a separate validation cohort of general practices.

RESULTS

Of 6 952 440 vaccinated patients in the derivation cohort, 5 150 310 (74.1%) had two vaccine doses. Of 2031 covid-19 deaths and 1929 covid-19 hospital admissions, 81 deaths (4.0%) and 71 admissions (3.7%) occurred 14 days or more after the second vaccine dose. The risk algorithms included age, sex, ethnic origin, deprivation, body mass index, a range of comorbidities, and SARS-CoV-2 infection rate. Incidence of covid-19 mortality increased with age and deprivation, male sex, and Indian and Pakistani ethnic origin. Cause specific hazard ratios were highest for patients with Down's syndrome (12.7-fold increase), kidney transplantation (8.1-fold), sickle cell disease (7.7-fold), care home residency (4.1-fold), chemotherapy (4.3-fold), HIV/AIDS (3.3-fold), liver cirrhosis (3.0-fold), neurological conditions (2.6-fold), recent bone marrow transplantation or a solid organ transplantation ever (2.5-fold), dementia (2.2-fold), and Parkinson's disease (2.2-fold). Other conditions with increased risk (ranging from 1.2-fold to 2.0-fold increases) included chronic kidney disease, blood cancer, epilepsy, chronic obstructive pulmonary disease, coronary heart disease, stroke, atrial fibrillation, heart failure, thromboembolism, peripheral vascular disease, and type 2 diabetes. A similar pattern of associations was seen for covid-19 related hospital admissions. No evidence indicated that associations differed after the second dose, although absolute risks were reduced. The risk algorithm explained 74.1% (95% confidence interval 71.1% to 77.0%) of the variation in time to covid-19 death in the validation cohort. Discrimination was high, with a D statistic of 3.46 (95% confidence interval 3.19 to 3.73) and C statistic of 92.5. Performance was similar after each vaccine dose. In the top 5% of patients with the highest predicted covid-19 mortality risk, sensitivity for identifying covid-19 deaths within 70 days was 78.7%.

CONCLUSION

This population based risk algorithm performed well showing high levels of discrimination for identifying those patients at highest risk of covid-19 related death and hospital admission after vaccination.

High-Risk Prediction of Cardiovascular Diseases via Attention-Based Deep Neural Networks

High-risk prediction of cardiovascular disease is of great significance and impendency in medical fields with the increasing phenomenon of sub-health these years. Most existing pathological methods for the prognosis prediction are either costly or prone to misjudgement. Therefore, plenty of automated models based on machine learning have been proposed to predict the onset of cardiovascular disease with the premorbid information of patients extracted from their historical Electronic Health Records (EHRs). However, it is a tough job to select proper features from longitudinal and heterogeneous EHRs, and also a great challenge to obtain accurate and robust representations for patients. In this paper, we propose an entirely end-to-end model called DeepRisk based on attention mechanism and deep neural networks, which can not only learn high-quality features automatically from EHRs, but also efficiently integrate heterogeneous and time-ordered medical data, and finally predict patients' risk of cardiovascular diseases. Experiments are carried out on a real medical dataset and results show that DeepRisk can significantly improve the high-risk prediction accuracy for cardiovascular disease compared with state-of-the-art approaches.

Electronic Medical Record-Based Case Phenotyping for the Charlson Conditions: Scoping Review

Background

Electronic medical records (EMRs) contain large amounts of rich clinical information. Developing EMR-based case definitions, also known as EMR phenotyping, is an active area of research that has implications for epidemiology, clinical care, and health services research.

Objective

This review aims to describe and assess the present landscape of EMR-based case phenotyping for the Charlson conditions.

Methods

A scoping review of EMR-based algorithms for defining the Charlson comorbidity index conditions was completed. This study covered articles published between January 2000 and April 2020, both inclusive. Embase (Excerpta Medica database) and MEDLINE (Medical Literature Analysis and Retrieval System Online) were searched using keywords developed in the following 3 domains: terms related to EMR, terms related to case finding, and disease-specific terms. The manuscript follows the Preferred Reporting Items for Systematic reviews and Meta-analyses extension for Scoping Reviews (PRISMA) guidelines.

Results

A total of 274 articles representing 299 algorithms were assessed and summarized. Most studies were undertaken in the United States (181/299, 60.5%), followed by the United Kingdom (42/299, 14.0%) and Canada (15/299, 5.0%). These algorithms were mostly developed either in primary care (103/299, 34.4%) or inpatient (168/299, 56.2%) settings. Diabetes, congestive heart failure, myocardial infarction, and rheumatology had the highest number of developed algorithms. Data-driven and clinical rule–based approaches have been identified. EMR-based phenotype and algorithm development reflect the data access allowed by respective health systems, and algorithms vary in their performance.

Conclusions

Recognizing similarities and differences in health systems, data collection strategies, extraction, data release protocols, and existing clinical pathways is critical to algorithm development strategies. Several strategies to assist with phenotype-based case definitions have been proposed.

Cardiovascular Risk Calculators and their Applicability to South Asians

BACKGROUND

Estimation of absolute cardiovascular disease (CVD) risk and tailoring therapies according to the estimated risk is a fundamental concept in the primary prevention of CVD is assessed in this study. Numerous CVD risk scores are currently available for use in various populations but unfortunately, none exist for South Asians who have much higher CVD risk as compared to their western counterparts.

METHODS

A literature search was done using PubMed and Google search engines to prepare a narrative review on this topic.

RESULTS

Various currently available CVD risk scores and their pros and cons are summarized. The studies performed in native as well as migrant South Asians evaluating the accuracy of these risk scores for estimation of CVD risk are also summarized. The findings of these studies have generally been inconsistent, but it appears that the British risk scores (e.g. QRISK versions) may be more accurate because of inclusion of migrant South Asians in the derivation of these risk scores. However, the lack of any prospective study precludes our ability to draw any firm conclusions. Finally, the potential solution to these challenges, including the role of recalibration and subclinical atherosclerosis imaging, is also discussed.

CONCLUSION

This review highlights the need to develop large, representative, prospectively followed databases of South Asians providing information on various CVD risk factors and their contribution to incident CVD. Such databases will not only allow the development of validated CVD risk scores for South Asians but will also enable application of machine-learning approaches to provide personalized solutions to CVD risk assessment and management in these populations.

Trading-Off Machine Learning Algorithms towards Data-Driven Administrative-Socio-Economic Population Health Management

Together with population ageing, the number of people suffering from multimorbidity is increasing, up to more than half of the population by 2035. This part of the population is composed by the highest-risk patients, who are, at the same time, the major users of the healthcare systems. The early identification of this sub-population can really help to improve people’s quality of life and reduce healthcare costs. In this paper, we describe a population health management tool based on state-of-the-art intelligent algorithms, starting from administrative and socio-economic data, for the early identification of high-risk patients. The study refers to the population of the Local Health Unit of Central Tuscany in 2015, which amounts to 1,670,129 residents. After a trade-off on machine learning models and on input data, Random Forest applied to 1-year of historical data achieves the best results, outperforming state-of-the-art models. The most important variables for this model, in terms of mean minimal depth, accuracy decrease and Gini decrease, result to be age and some group of drugs, such as high-ceiling diuretics. Thanks to the low inference time and reduced memory usage, the resulting model allows for real-time risk prediction updates whenever new data become available, giving General Practitioners the possibility to early adopt personalised medicine.

Cost-Effectiveness of Initiating Pharmacological Treatment in Stage One Hypertension Based on 10-Year Cardiovascular Disease Risk: A Markov Modeling Study

Supplemental Digital Content is available in the text.

Antihypertensive drug treatment is cost-effective for adults at high risk of developing cardiovascular disease (CVD). However, the cost-effectiveness in people with stage 1 hypertension (140–159 mm Hg systolic blood pressure) at lower CVD risk remains unclear. The objective was to establish the 10-year CVD risk threshold where initiating antihypertensive drug treatment for primary prevention in adults, with stage 1 hypertension, becomes cost-effective. A lifetime horizon Markov model compared antihypertensive drug versus no treatment, using a UK National Health Service perspective. Analyses were conducted for groups ranging between 5% and 20% 10-year CVD risk. Health states included no CVD event, CVD and non-CVD death, and 6 nonfatal CVD morbidities. Interventions were compared using cost-per-quality-adjusted life-years. The base-case age was 60, with analyses repeated between ages 40 and 75. The model was run separately for men and women, and threshold CVD risk assessed against the minimum plausible risk for each group. Treatment was cost-effective at 10% CVD risk for both sexes (incremental cost-effectiveness ratio £10 017/quality-adjusted life-year [$14 542] men, £8635/QALY [$12 536] women) in the base-case. The result was robust in probabilistic and deterministic sensitivity analyses but was sensitive to treatment effects. Treatment was cost-effective for men regardless of age and women aged >60. Initiating treatment in stage 1 hypertension for people aged 60 is cost-effective regardless of 10-year CVD risk. For other age groups, it is also cost-effective to treat regardless of risk, except in younger women.

Living risk prediction algorithm (QCOVID) for risk of hospital admission and mortality from coronavirus 19 in adults: national derivation and validation cohort study

OBJECTIVE

To derive and validate a risk prediction algorithm to estimate hospital admission and mortality outcomes from coronavirus disease 2019 (covid-19) in adults.

DESIGN

Population based cohort study.

SETTING AND PARTICIPANTS

QResearch database, comprising 1205 general practices in England with linkage to covid-19 test results, Hospital Episode Statistics, and death registry data. 6.08 million adults aged 19-100 years were included in the derivation dataset and 2.17 million in the validation dataset. The derivation and first validation cohort period was 24 January 2020 to 30 April 2020. The second temporal validation cohort covered the period 1 May 2020 to 30 June 2020.

MAIN OUTCOME MEASURES

The primary outcome was time to death from covid-19, defined as death due to confirmed or suspected covid-19 as per the death certification or death occurring in a person with confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in the period 24 January to 30 April 2020. The secondary outcome was time to hospital admission with confirmed SARS-CoV-2 infection. Models were fitted in the derivation cohort to derive risk equations using a range of predictor variables. Performance, including measures of discrimination and calibration, was evaluated in each validation time period.

RESULTS

4384 deaths from covid-19 occurred in the derivation cohort during follow-up and 1722 in the first validation cohort period and 621 in the second validation cohort period. The final risk algorithms included age, ethnicity, deprivation, body mass index, and a range of comorbidities. The algorithm had good calibration in the first validation cohort. For deaths from covid-19 in men, it explained 73.1% (95% confidence interval 71.9% to 74.3%) of the variation in time to death (R²); the D statistic was 3.37 (95% confidence interval 3.27 to 3.47), and Harrell's C was 0.928 (0.919 to 0.938). Similar results were obtained for women, for both outcomes, and in both time periods. In the top 5% of patients with the highest predicted risks of death, the sensitivity for identifying deaths within 97 days was 75.7%. People in the top 20% of predicted risk of death accounted for 94% of all deaths from covid-19.

CONCLUSION

The QCOVID population based risk algorithm performed well, showing very high levels of discrimination for deaths and hospital admissions due to covid-19. The absolute risks presented, however, will change over time in line with the prevailing SARS-C0V-2 infection rate and the extent of social distancing measures in place, so they should be interpreted with caution. The model can be recalibrated for different time periods, however, and has the potential to be dynamically updated as the pandemic evolves.

Chronic Non-Communicable Disease Risk Calculators - An Overview, Part II

The aim of this review was to identify the different risk assessment tools that stratify the individual's risk of four of the eight leading causes of death: stroke, ischaemic heart diseases, type 2 diabetes mellitus, and dementia. It follows part I, which summarized the risk assessment tools for the other four leading causes of death (breast cancer, lung cancer, colorectal cancer and osteoporosis). As in part I, the different tools were compared by their variables and validation criteria and an overview table was designed for each illness. The tables facilitate the choice of the adequate risk assessment tool for the individual patient in order to estimate the risk of developing an NCD. This could guide treating physicians in the decision-making process about completing diagnostics for early detection and, if necessary, treatment, such that the patient's quality of life can be preserved and costs to the health care system are minimal.

Enhancing ICD-Code-Based Case Definition for Heart Failure Using Electronic Medical Record Data

BACKGROUND

Surveillance and outcome studies for heart failure (HF) require accurate identification of patients with HF. Algorithms based on International Classification of Diseases (ICD) codes to identify HF from administrative data are inadequate owing to their relatively low sensitivity. Detailed clinical information from electronic medical records (EMRs) is potentially useful for improving ICD algorithms. This study aimed to enhance the ICD algorithm for HF definition by incorporating comprehensive information from EMRs.

METHODS

The study included 2106 inpatients in Calgary, Alberta, Canada. Medical chart review was used as the reference gold standard for evaluating developed algorithms. The commonly used ICD codes for defining HF were used (namely, the ICD algorithm). The performance of different algorithms using the free text discharge summaries from a population-based EMR were compared with the ICD algorithm. These algorithms included a keyword search algorithm looking for HF-specific terms, a machine learning-based HF concept (HFC) algorithm, an EMR structured data based algorithm, and combined algorithms (the ICD and HFC combined algorithm).

RESULTS

Of 2106 patients, 296 (14.1%) were patients with HF as determined by chart review. The ICD algorithm had 92.4% positive predictive value (PPV) but low sensitivity (57.4%). The EMR keyword search algorithm achieved a higher sensitivity (65.5%) than the ICD algorithm, but with a lower PPV (77.6%). The HFC algorithm achieved a better sensitivity (80.0%) and maintained a reasonable PPV (88.9%) compared with the ICD algorithm and the keyword algorithm. An even higher sensitivity (83.3%) was reached by combining the HFC and ICD algorithms, with a lower PPV (83.3%). The structured EMR data algorithm reached a sensitivity of 78% and a PPV of 54.2%. The combined EMR structured data and ICD algorithm had a higher sensitivity (82.4%), but the PPV remained low at 54.8%. All algorithms had a specificity ranging from 87.5% to 99.2%.

CONCLUSIONS

Applying natural language processing and machine learning on the discharge summaries of inpatient EMR data can improve the capture of cases of HF compared with the widely used ICD algorithm. The utility of the HFC algorithm is straightforward, making it easily applied for HF case identification.

Performance of the Framingham risk models and pooled cohort equations for predicting 10-year risk of cardiovascular disease: a systematic review and meta-analysis

BACKGROUND

The Framingham risk models and pooled cohort equations (PCE) are widely used and advocated in guidelines for predicting 10-year risk of developing coronary heart disease (CHD) and cardiovascular disease (CVD) in the general population. Over the past few decades, these models have been extensively validated within different populations, which provided mounting evidence that local tailoring is often necessary to obtain accurate predictions. The objective is to systematically review and summarize the predictive performance of three widely advocated cardiovascular risk prediction models (Framingham Wilson 1998, Framingham ATP III 2002 and PCE 2013) in men and women separately, to assess the generalizability of performance across different subgroups and geographical regions, and to determine sources of heterogeneity in the findings across studies.

METHODS

A search was performed in October 2017 to identify studies investigating the predictive performance of the aforementioned models. Studies were included if they externally validated one or more of the original models in the general population for the same outcome as the original model. We assessed risk of bias for each validation and extracted data on population characteristics and model performance. Performance estimates (observed versus expected (OE) ratio and c-statistic) were summarized using a random effects models and sources of heterogeneity were explored with meta-regression.

RESULTS

The search identified 1585 studies, of which 38 were included, describing a total of 112 external validations. Results indicate that, on average, all models overestimate the 10-year risk of CHD and CVD (pooled OE ratio ranged from 0.58 (95% CI 0.43-0.73; Wilson men) to 0.79 (95% CI 0.60-0.97; ATP III women)). Overestimation was most pronounced for high-risk individuals and European populations. Further, discriminative performance was better in women for all models. There was considerable heterogeneity in the c-statistic between studies, likely due to differences in population characteristics.

CONCLUSIONS

The Framingham Wilson, ATP III and PCE discriminate comparably well but all overestimate the risk of developing CVD, especially in higher risk populations. Because the extent of miscalibration substantially varied across settings, we highly recommend that researchers further explore reasons for overprediction and that the models be updated for specific populations.

Development of an intelligent decision support system for ischemic stroke risk assessment in a population-based electronic health record database

BACKGROUND

Intelligent decision support systems (IDSS) have been applied to tasks of disease management. Deep neural networks (DNNs) are artificial intelligent techniques to achieve high modeling power. The application of DNNs to large-scale data for estimating stroke risk needs to be assessed and validated. This study aims to apply a DNN for deriving a stroke predictive model using a big electronic health record database.

METHODS AND RESULTS

The Taiwan National Health Insurance Research Database was used to conduct a retrospective population-based study. The database was divided into one development dataset for model training (~70% of total patients for training and ~10% for parameter tuning) and two testing datasets (each ~10%). A total of 11,192,916 claim records from 840,487 patients were used. The primary outcome was defined as any ischemic stroke in inpatient records within 3 years after study enrollment. The DNN was evaluated using the area under the receiver operating characteristic curve (AUC or c-statistic). The development dataset included 672,214 patients (a total of 8,952,000 records) of whom 2,060 patients had stroke events. The mean age of the population was 35.5±20.2 years, with 48.5% men. The model achieved AUC values of 0.920 (95% confidence interval [CI], 0.908-0.932) in testing dataset 1 and 0.925 (95% CI, 0.914-0.937) in testing dataset 2. Under a high sensitivity operating point, the sensitivity and specificity were 92.5% and 79.8% for testing dataset 1; 91.8% and 79.9% for testing dataset 2. Under a high specificity operating point, the sensitivity and specificity were 80.3% and 87.5% for testing dataset 1; 83.7% and 87.5% for testing dataset 2. The DNN model maintained high predictability 5 years after being developed. The model achieved similar performance to other clinical risk assessment scores.

CONCLUSIONS

Using a DNN algorithm on this large electronic health record database is capable of obtaining a high performing model for assessment of ischemic stroke risk. Further research is needed to determine whether such a DNN-based IDSS could lead to an improvement in clinical practice.

Population-level changes in lifestyle risk factors for cardiovascular disease in the Heart of New Ulm Project

Lifestyle significantly influences development of cardiovascular disease (CVD), but limited data exists demonstrating lifestyle improvements in community-based interventions. This study aims to document how lifestyle risk factors changed at the population level in the context of Heart of New Ulm (HONU), a community-based CVD prevention initiative in Minnesota.

HONU intervened across worksites, healthcare and the community/environment to reduce CVD risk factors. HONU collected behavioral measures including smoking, physical activity, fruit/vegetable consumption, alcohol use and stress at heart health screenings from 2009 to 2014. All screenings were documented in the electronic health record (EHR). Changes at the community level for the target population (age 40–79) were estimated using weights created from EHR data and modeled using generalized estimating equation models.

Screening participants were similar to the larger patient population with regard to age, race, and marital status, but were slightly healthier in regards to BMI, LDL cholesterol, blood pressure, and less likely to smoke. Community-level improvements were significant for physical activity (62.8% to 70.5%, p < 0.001) and 5+ daily fruit/vegetable servings (16.9% to 28.1%, p < 0.001), with no significant change in smoking, stress, alcohol or BMI.

By leveraging local EHR data and integrating it with patient-reported outcomes, improvements in nutrition and physical activity were identified in the HONU population, but limited changes were noted for smoking, alcohol consumption and stress. Systematically documenting behaviors in the EHR will help healthcare systems impact the health of the communities they serve, both at the individual and population level.

Factors associated with exacerbations among adults with asthma according to electronic health record data

BACKGROUND

Asthma is a chronic inflammatory lung disease that affects 18.7 million U.S. adults. Electronic health records (EHRs) are a unique source of information that can be leveraged to understand factors associated with asthma in real-life populations. In this study, we identify demographic factors and comorbidities associated with asthma exacerbations among adults according to EHR-derived data and compare these findings to those of epidemiological studies.

METHODS

We obtained University of Pennsylvania Hospital System EHR-derived data for asthma encounters occurring between 2011 and 2014. Regression analyses were performed to model asthma exacerbation frequency as explained by age, sex, race/ethnicity, health insurance type, smoking status, body mass index (BMI) and various comorbidities. We analyzed data from the National Health and Nutrition Examination Survey (NHANES) from 2001 to 2012 to compare findings with those from the EHR-derived data.

RESULTS

Based on data from 9068 adult patients with asthma, 33.37% had at least one exacerbation over the four-year study period. In a proportional odds logistic regression predicting number of exacerbations during the study period (levels: 0, 1-2, 3-4, 5+ exacerbations), after controlling for age, race/ethnicity, sex, health insurance type, and smoking status, the highest odds ratios (ORs) of significantly associated factors were: chronic bronchitis (2.70), sinusitis (1.50), emphysema (1.39), fluid and electrolyte disorders (1.35), class 3 obesity (1.32), and diabetes (1.28). An analysis of NHANES data showed associations for class 3 obesity, anemia and chronic bronchitis with exacerbation frequency in an adjusted model controlling for age, race/ethnicity, sex, financial class and smoking status.

CONCLUSIONS

EHR-derived data is helpful to understand exacerbations in real-life asthma patients, facilitating design of detailed studies and interventions tailored for specific populations.

Discrimination between Obstructive Coronary Artery Disease and Cardiac Syndrome X in Women with Typical Angina and Positive Exercise Test; Utility of Cardiovascular Risk Calculators

Introduction: Nearly 40% of women with typical angina and a positive exercise tolerance test (ETT) have normal or near normal coronary angiography (CAG) labeled as cardiac syndrome X (CSX). Objective: We performed this study to evaluate the power of common cardiovascular risk calculators to distinguish patients with CSX from those with coronary artery disease (CAD). Methods: 559 women participated in the study. Three risk scores, including (1) newly pooled cohort equation of American College of Cardiology/American Heart Association (ACC/AHA) to predict 10 years risk of first atherosclerotic cardiovascular hard event (ASCVD), (2) Framingham risk score (FRS) for the prediction of 10 years coronary heart disease, and (3) the SCORE tool to estimate 10-year risk of cardiovascular mortality (SCORE), were applied. Results: CAD was diagnosed in 51.5% of the patients. 11.6% of the population had ASCVD < 2.5%, and only 13.8% of these patients had CAD on their CAG. By choosing FRS, 14.4% of patients had FRS < 7.5%, and only 11.3% of these patients had recorded CAD on CAG, while the rest of the patients were diagnosed as CSX. Using the SCORE model, 13.8% of patients had the least value (<0.5%) in whom the prevalence of CAD was 19.9%. The area under receiver operating characteristic curve (AUROC) to discriminate CSX from CAD was calculated for each scoring system, being 0.750 for ASCVD, 0.745 for FRS, and 0.728 for SCORE (p value for all AUROCs < 0.001). The Hosmer⁻Lemeshow chi squares (df, p value) for calibration were 8.787 (8, 0.361), 11.125 (8, 0.195), and 10.618 (8, 0.224) for ASCVD, FRS, and SCORE, respectively. Conclusions: Patients who have ASCVD < 2.5% or FRS < 7.5% may be appropriate cases for noninvasive imaging (Such as coronary CT angiography). CAG is indicated for patients with ASCVD ≥ 7.5% and FRS ≥ 15%, whereas the patients with intermediate risk need comprehensive patient⁻physician shared decision-making.

Association of epistaxis with atherosclerotic cardiovascular disease

OBJECTIVES/HYPOTHESIS

To investigate the association between epistaxis and atherosclerotic cardiovascular disease.

STUDY DESIGN

Case-control cohort study.

METHODS

This study included patients from the tertiary-care ear, nose, and throat department at the University Hospital of Zurich between December 1, 2016 and June 1, 2017. We assessed the cardiovascular risk profiles in a group of 41 patients presenting with epistaxis, and a group of 41 matched controls, focusing on a surrogate parameter for atherosclerosis: the carotid intima-media thickness (CIMT).

RESULTS

With a mean of 1.06 mm (standard deviation [SD] = 0.17), CIMT values were on average 26% higher in epistaxis patients than in their controls, with a mean of 0.84 mm (SD = 0.14; P < .001). Occurrence of severe epistaxis was also associated with lower ankle-brachial index values at 0.96 (SD = 0.12) versus 1.05 (SD = 0.17) (P < .001) and significantly higher QRISK2 relative risks (an algorithm for predicting cardiovascular risk) than found in the control group (1.81, SD = 0.97 vs. 1.35, SD = 0.28; P = .028). A binary logistic regression model, adjusted for possible confounders, showed an odds ratio of 2.5 for the occurrence of epistaxis per increase in CIMT of 0.1 mm in the study population (95% confidence interval: 1.56-4.11; P < .001).

CONCLUSIONS

The occurrence of severe epistaxis was shown to be closely associated with the prevalence of atherosclerotic cardiovascular disease. Accordingly, patients affected by epistaxis should be regarded as at an elevated cardiovascular risk, which indicates the need for appropriate further medical assessment and preventive measures.

LEVEL OF EVIDENCE

3b TRIAL REGISTRATION: Clinical trials NCT03092973 Laryngoscope, 129:783-787, 2019.

Subclinical parameters of arterial stiffness and arteriosclerosis correlate with QRISK3 in systemic lupus erythematosus

It is well known that cardiovascular diseases (CVD) are a major contributor of death in systemic lupus erythematosus (SLE) as well in other rheumatic illness. In the last decades, there has been a growing development of different methodologies with the purpose of early detection of CVD. Objective: The aim of this study is to correlate the usefulness of subclinical parameters of vascular aging and QRISK 3–2017 score for early detection of CVD in SLE. Methods: Clinical assessment including systemic lupus erythematosus disease activity index (SLEDAI) and systemic lupus international collaborating clinics / american college of rheumatology damage index (SLICC/ACR DI), laboratory measurements, carotid ultrasound examination, carotid intima media thickness (cIMT) measurement, carotid distention and diameter analysis, arterial stiffness measurement measured by tonometry and QRISK 3–2017 were done. All results were analyzed by SPSS 24 software. Results: We observed correlation between QRISK3 and mean cIMT (rs = 0.534, P < 0.001), PWV (rs = 0.474, P < 0.001), cfPWV (rs = 0.569, P < 0.001) and distensibility (rs = -0.420, P = 0.006). Consistent with above, SLE patients in middle and high risk QRISK 3–2017 showed increased arterial stiffness versus low risk group. Conclusions: We encourage to the rheumatology community to assess cardiovascular risk in SLE patients with QRISK 3–2017 risk calculator as an alternative method at the outpatient clinic along a complete cardiovascular evaluation when appropriate.

Approaches to Link Geospatially Varying Social, Economic, and Environmental Factors with Electronic Health Record Data to Better Understand Asthma Exacerbations

Electronic health record (EHR)-derived data has become an invaluable resource for biomedical research, but is seldom used for the study of the health impacts of social and environmental factors due in part to the unavailability of relevant variables. We describe how EHR-derived data can be enhanced via linking of external sources of social, economic and environmental data when patient-related geospatial information is available, and we illustrate an approach to better understand the geospatial patterns of asthma exacerbation rates in Philadelphia. Specifically, we relate the spatial distribution of asthma exacerbations observed in EHR-derived data to that of known and potential risk factors (i.e., economic deprivation, crime, vehicular traffic, tree cover). Areas of highest risk based on integrated social and environmental data were consistent with an area with increased asthma exacerbations, demonstrating that data external to the EHR can enhance our understanding of negative health-related outcomes.

Enhancing Electronic Health Record Data with Geospatial Information

Electronic Health Record (EHR)-derived data is a valuable resource for research, and efforts are underway to overcome some of its limitations by using data from external sources to gain a fuller picture of patient characteristics, symptoms, and exposures. Our goal was to assess the utility of augmenting EHR data with geocoded patient addresses to identify geospatial variation of disease that is not explained by EHR-derived demographic factors. Using 2011-2014 encounter data from 27,604 University of Pennsylvania Hospital System asthma patients, we identified factors associated with asthma exacerbations: risk was higher in female, black, middle aged to elderly, and obese patients, as well as those with positive smoking history and with Medicare or Medicaid vs. private insurance. Significant geospatial variability of asthma exacerbations was found using generalized additive models, even after adjusting for demographic factors. Our work shows that geospatial data can be used to cost-effectively enhance EHR data.

Cardiovascular Risk Assessment

Cardiovascular risk assessment is fundamental to prevention of cardiovascular disease, because it helps determine the size of the potential benefits that might accrue to individual patients from use of statins, aspirin, and other preventive interventions. Current guidelines recommend specific algorithms for cardiovascular risk assessment that combine information from traditional risk factors including blood pressure, lipids, and smoking, along with age and sex and other factors. These algorithms are the subject of active research and controversy. This article addresses the rationale, current guidelines and use, and potential future directions of cardiovascular risk assessment.

Individual housing-based socioeconomic status predicts risk of accidental falls among adults

PURPOSE

Accidental falls are a major public health concern among people of all ages. Little is known about whether an individual-level housing-based socioeconomic status measure is associated with the risk of accidental falls.

METHODS

Among 12,286 Mayo Clinic Biobank participants residing in Olmsted County, Minnesota, subjects who experienced accidental falls between the biobank enrollment and September 2014 were identified using ICD-9 codes evaluated at emergency departments. HOUSES (HOUsing-based Index of SocioEconomic Status), a socioeconomic status measure based on individual housing features, was also calculated. Cox regression models were utilized to assess the association of the HOUSES (in quartiles) with accidental fall risk.

RESULTS

Seven hundred eleven (5.8%) participants had at least one emergency room visit due to an accidental fall during the study period. Subjects with higher HOUSES were less likely to experience falls in a dose-response manner (hazard ratio: 0.58; 95% confidence interval: 0.44-0.76 for comparing the highest to the lowest quartile). In addition, the HOUSES was positively associated with better health behaviors, social support, and functional status.

CONCLUSIONS

The HOUSES is inversely associated with accidental fall risk requiring emergency care in a dose-response manner. The HOUSES may capture falls-related risk factors through housing features and socioeconomic status-related psychosocial factors.

Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study

Objectives To develop and validate updated QRISK3 prediction algorithms to estimate the 10 year risk of cardiovascular disease in women and men accounting for potential new risk factors.Design Prospective open cohort study.Setting General practices in England providing data for the QResearch database.Participants 1309 QResearch general practices in England: 981 practices were used to develop the scores and a separate set of 328 practices were used to validate the scores. 7.89 million patients aged 25-84 years were in the derivation cohort and 2.67 million patients in the validation cohort. Patients were free of cardiovascular disease and not prescribed statins at baseline.Methods Cox proportional hazards models in the derivation cohort to derive separate risk equations in men and women for evaluation at 10 years. Risk factors considered included those already in QRISK2 (age, ethnicity, deprivation, systolic blood pressure, body mass index, total cholesterol: high density lipoprotein cholesterol ratio, smoking, family history of coronary heart disease in a first degree relative aged less than 60 years, type 1 diabetes, type 2 diabetes, treated hypertension, rheumatoid arthritis, atrial fibrillation, chronic kidney disease (stage 4 or 5)) and new risk factors (chronic kidney disease (stage 3, 4, or 5), a measure of systolic blood pressure variability (standard deviation of repeated measures), migraine, corticosteroids, systemic lupus erythematosus (SLE), atypical antipsychotics, severe mental illness, and HIV/AIDs). We also considered erectile dysfunction diagnosis or treatment in men. Measures of calibration and discrimination were determined in the validation cohort for men and women separately and for individual subgroups by age group, ethnicity, and baseline disease status.Main outcome measures Incident cardiovascular disease recorded on any of the following three linked data sources: general practice, mortality, or hospital admission records.Results 363 565 incident cases of cardiovascular disease were identified in the derivation cohort during follow-up arising from 50.8 million person years of observation. All new risk factors considered met the model inclusion criteria except for HIV/AIDS, which was not statistically significant. The models had good calibration and high levels of explained variation and discrimination. In women, the algorithm explained 59.6% of the variation in time to diagnosis of cardiovascular disease (R², with higher values indicating more variation), and the D statistic was 2.48 and Harrell's C statistic was 0.88 (both measures of discrimination, with higher values indicating better discrimination). The corresponding values for men were 54.8%, 2.26, and 0.86. Overall performance of the updated QRISK3 algorithms was similar to the QRISK2 algorithms.Conclusion Updated QRISK3 risk prediction models were developed and validated. The inclusion of additional clinical variables in QRISK3 (chronic kidney disease, a measure of systolic blood pressure variability (standard deviation of repeated measures), migraine, corticosteroids, SLE, atypical antipsychotics, severe mental illness, and erectile dysfunction) can help enable doctors to identify those at most risk of heart disease and stroke.

Corrected ROC analysis for misclassified binary outcomes

Creating accurate risk prediction models from Big Data resources such as Electronic Health Records (EHRs) is a critical step toward achieving precision medicine. A major challenge in developing these tools is accounting for imperfect aspects of EHR data, particularly the potential for misclassified outcomes. Misclassification, the swapping of case and control outcome labels, is well known to bias effect size estimates for regression prediction models. In this paper, we study the effect of misclassification on accuracy assessment for risk prediction models and find that it leads to bias in the area under the curve (AUC) metric from standard ROC analysis. The extent of the bias is determined by the false positive and false negative misclassification rates as well as disease prevalence. Notably, we show that simply correcting for misclassification while building the prediction model is not sufficient to remove the bias in AUC. We therefore introduce an intuitive misclassification-adjusted ROC procedure that accounts for uncertainty in observed outcomes and produces bias-corrected estimates of the true AUC. The method requires that misclassification rates are either known or can be estimated, quantities typically required for the modeling step. The computational simplicity of our method is a key advantage, making it ideal for efficiently comparing multiple prediction models on very large datasets. Finally, we apply the correction method to a hospitalization prediction model from a cohort of over 1 million patients from the Veterans Health Administrations EHR. Implementations of the ROC correction are provided for Stata and R. Published 2017. This article is a U.S. Government work and is in the public domain in the USA.