The Learning Healthcare System and Cardiovascular Care: A Scientific Statement From the American Heart Association

Twenty Years of Get With The Guidelines-Stroke: Celebrating Past Successes, Lessons Learned, and Future Challenges

The Get With The Guidelines-Stroke program which, began 20 years ago, is one of the largest and most important nationally representative disease registries in the United States. Its importance to the stroke community can be gauged by its sustained growth and widespread dissemination of findings that demonstrate sustained increases in both the quality of care and patient outcomes over time. The objectives of this narrative review are to provide a brief history of Get With The Guidelines-Stroke, summarize its major successes and impact, and highlight lessons learned. Looking to the next 20 years, we discuss potential challenges and opportunities for the program.

National Trends in Hospital Performance in Guideline-Recommended Pharmacologic Treatment for Heart Failure at Discharge

BACKGROUND

The use of recommended heart failure (HF) medications has improved over time, but opportunities for improvement persist among women and at rural hospitals.

OBJECTIVES

This study aims to characterize national trends in performance in the use of guideline-recommended pharmacologic treatment for HF at U.S. Department of Veterans Affairs (VA) hospitals, at which medication copayments are modest.

METHODS

Among patients discharged from VA hospitals with HF between January 1, 2013, and December 31, 2019, receipt of all guideline-recommended HF pharmacotherapy among eligible patients was assessed, consisting of evidence-based beta-blockers; angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, or angiotensin receptor neprilysin inhibitors; mineralocorticoid receptor antagonists; and oral anticoagulation.

RESULTS

Of 55,560 patients at 122 hospitals, 32,304 (58.1%) received all guideline-recommended HF medications for which they were eligible. The proportion of patients receiving all recommended medications was higher in 2019 relative to 2013 (OR: 1.54; 95% CI: 1.44-1.65). The median of hospital performance was 59.1% (Q1-Q3: 53.2%-66.2%), improving with substantial variation across sites from 2013 (median 56.4%; Q1-Q3: 50.0%-62.0%) to 2019 (median 65.7%; Q1-Q3: 56.3%-73.5%). Women were less likely to receive recommended therapies than men (adjusted OR [aOR]: 0.84; 95% CI: 0.74-0.96). Compared with non-Hispanic White patients, non-Hispanic Black patients were less likely to receive recommended therapies (aOR: 0.83; 95% CI: 0.79-0.87). Urban hospital location was associated with lower likelihood of medication receipt (aOR: 0.73; 95% CI: 0.59-0.92).

CONCLUSIONS

Forty-two percent of patients did not receive all recommended HF medications at discharge, particularly women, minority patients, and those receiving care at urban hospitals. Rates of use increased over time, with variation in performance across hospitals.

AssistMED project: Transforming cardiology cohort characterisation from electronic health records through natural language processing - Algorithm design, preliminary results, and field prospects

INTRODUCTION

Electronic health records (EHR) are of great value for clinical research. However, EHR consists primarily of unstructured text which must be analysed by a human and coded into a database before data analysis- a time-consuming and costly process limiting research efficiency. Natural language processing (NLP) can facilitate data retrieval from unstructured text. During AssistMED project, we developed a practical, NLP tool that automatically provides comprehensive clinical characteristics of patients from EHR, that is tailored to clinical researchers needs.

MATERIAL AND METHODS

AssistMED retrieves patient characteristics regarding clinical conditions, medications with dosage, and echocardiographic parameters with clinically oriented data structure and provides researcher-friendly database output. We validate the algorithm performance against manual data retrieval and provide critical quantitative and qualitative analysis.

RESULTS

AssistMED analysed the presence of 56 clinical conditions, medications from 16 drug groups with dosage and 15 numeric echocardiographic parameters in a sample of 400 patients hospitalized in the cardiology unit. No statistically significant differences between algorithm and human retrieval were noted. Qualitative analysis revealed that disagreements with manual annotation were primarily accounted to random algorithm errors, erroneous human annotation and lack of advanced context awareness of our tool.

CONCLUSIONS

Current NLP approaches are feasible to acquire accurate and detailed patient characteristics tailored to clinical researchers' needs from EHR. We present an in-depth description of an algorithm development and validation process, discuss obstacles and pinpoint potential solutions, including opportunities arising with recent advancements in the field of NLP, such as large language models.

Performance of risk models to predict mortality risk for patients with heart failure: evaluation in an integrated health system

BACKGROUND

Referral of patients with heart failure (HF) who are at high mortality risk for specialist evaluation is recommended. Yet, most tools for identifying such patients are difficult to implement in electronic health record (EHR) systems.

OBJECTIVE

To assess the performance and ease of implementation of Machine learning Assessment of RisK and EaRly mortality in Heart Failure (MARKER-HF), a machine-learning model that uses structured data that is readily available in the EHR, and compare it with two commonly used risk scores: the Seattle Heart Failure Model (SHFM) and Meta-Analysis Global Group in Chronic (MAGGIC) Heart Failure Risk Score.

DESIGN

Retrospective, cohort study.

PARTICIPANTS

Data from 6764 adults with HF were abstracted from EHRs at a large integrated health system from 1/1/10 to 12/31/19.

MAIN MEASURES

One-year survival from time of first cardiology or primary care visit was estimated using MARKER-HF, SHFM, and MAGGIC. Discrimination was measured by the area under the receiver operating curve (AUC). Calibration was assessed graphically.

KEY RESULTS

Compared to MARKER-HF, both SHFM and MAGGIC required a considerably larger amount of data engineering and imputation to generate risk score estimates. MARKER-HF, SHFM, and MAGGIC exhibited similar discriminations with AUCs of 0.70 (0.69-0.73), 0.71 (0.69-0.72), and 0.71 (95% CI 0.70-0.73), respectively. All three scores showed good calibration across the full risk spectrum.

CONCLUSIONS

These findings suggest that MARKER-HF, which uses readily available clinical and lab measurements in the EHR and required less imputation and data engineering than SHFM and MAGGIC, is an easier tool to identify high-risk patients in ambulatory clinics who could benefit from referral to a HF specialist.

Effect of clinical decision support for severe hypercholesterolemia on low-density lipoprotein cholesterol levels

Severe hypercholesterolemia/possible familial hypercholesterolemia (FH) is relatively common but underdiagnosed and undertreated. We investigated whether implementing clinical decision support (CDS) was associated with lower low-density lipoprotein cholesterol (LDL-C) in patients with severe hypercholesterolemia/possible FH (LDL-C ≥ 190 mg/dL). As part of a pre-post implementation study, a CDS alert was deployed in the electronic health record (EHR) in a large health system comprising 3 main sites, 16 hospitals and 53 clinics. Data were collected for 3 months before ('silent mode') and after ('active mode') its implementation. Clinicians were only able to view the alert in the EHR during active mode. We matched individuals 1:1 in both modes, based on age, sex, and baseline lipid lowering therapy (LLT). The primary outcome was difference in LDL-C between the two groups and the secondary outcome was initiation/intensification of LLT after alert trigger. We identified 800 matched patients in each mode (mean ± SD age 56.1 ± 11.8 y vs. 55.9 ± 11.8 y; 36.0% male in both groups; mean ± SD initial LDL-C 211.3 ± 27.4 mg/dL vs. 209.8 ± 23.9 mg/dL; 11.2% on LLT at baseline in each group). LDL-C levels were 6.6 mg/dL lower (95% CI, -10.7 to -2.5; P = 0.002) in active vs. silent mode. The odds of high-intensity statin use (OR, 1.78; 95% CI, 1.41-2.23; P < 0.001) and LLT initiation/intensification (OR, 1.30, 95% CI, 1.06-1.58, P = 0.01) were higher in active vs. silent mode. Implementation of a CDS was associated with lowering of LDL-C levels in patients with severe hypercholesterolemia/possible FH, likely due to higher rates of clinician led LLT initiation/intensification.

Improving Venous Thromboembolism Prophylaxis Through Service Integration, Policy Enhancement, and Health Informatics

Introduction

Venous thromboembolism (VTE) prevention and management are susceptible issues that require specific rules to sustain and oversee their functioning, as preventing VTE is a vital patient safety priority. This paper aims to investigate and provide recommendations for VTE assessment and reassessment through policy enhancement and development.

Methods

We reviewed different papers and policies to propose recommendations and theme analysis for policy modifications and enhancements to improve VTE prophylaxis and management.

Results

Recommendations were set to enhance the overall work of VTE prophylaxis, where the current VTE protocols and policies must ensure high levels of patient safety and satisfaction. The recommendations included working through a well-organized multidisciplinary team and staff engagement to support and enhance VTE's work. Nurses', pharmacists', and physical therapists' involvement in setting up the plan and prevention is the way to share the knowledge and paradigm of experience to standardize the management. Promoting policies regarding VTE prophylaxis assessment and reassessment using electronic modules as a part of the digital health process was proposed. A deep understanding of the underlying issues and the incorporation of generic policy recommendations were set.

Conclusion

This article presents recommendations for stakeholders, social media platforms, and healthcare practitioners to enhance VTE prophylaxis and management.

Advancing the care of individuals with cancer through innovation & technology: Proceedings from the cardiology oncology innovation summit 2020 and 2021

As cancer therapies increase in effectiveness and patients' life expectancies improve, balancing oncologic efficacy while reducing acute and long-term cardiovascular toxicities has become of paramount importance. To address this pressing need, the Cardiology Oncology Innovation Network (COIN) was formed to bring together domain experts with the overarching goal of collaboratively investigating, applying, and educating widely on various forms of innovation to improve the quality of life and cardiovascular healthcare of patients undergoing and surviving cancer therapies. The COIN mission pillars of innovation, collaboration, and education have been implemented with cross-collaboration among academic institutions, private and public establishments, and industry and technology companies. In this report, we summarize proceedings from the first two annual COIN summits (inaugural in 2020 and subsequent in 2021) including educational sessions on technological innovations for establishing best practices and aligning resources. Herein, we highlight emerging areas for innovation and defining unmet needs to further improve the outcome for cancer patients and survivors of all ages. Additionally, we provide actionable suggestions for advancing innovation, collaboration, and education in cardio-oncology in the digital era.

A Real-World Analysis of New-Onset Heart Failure After Anterior Wall ST-Elevation Acute Myocardial Infarction in the United States

The 1-year incidence of heart failure (HF) after anterior wall ST-elevation acute myocardial infarction (STEMI) remains difficult to determine because of inconsistencies in reporting, definitions, and adjudication. The objective of this study was to evaluate the 1-year incidence of HF after anterior wall STEMI in a real-world data set using a variety of potential criteria and composite definitions. In a retrospective cohort study, anonymized patient data was accessed through a federated health research network (TriNetX Limited Liability Company (LLC)) of 56 US healthcare organizations (US Collaborative Network). Patients were identified based on the International Classification of Diseases, Tenth Revision criteria for anterior wall STEMI during the 10-year period from 2013 to 2022 and the absence of prespecified signs or symptoms of HF. Values for 1-year incidence were calculated as 1 minus Kaplan-Meier survival at 12 months after anterior wall STEMI. Univariate Cox proportional hazard ratio was calculated to compare risk associated with potential risk factors. The analysis utilized 5 different types of definition criteria for HF: Diagnosis codes, Signs and symptoms, Laboratory/imaging, Medications, and Composites. A total of 34,395 patients from the US Collaborative Network met eligibility criteria and were included in the analysis. The 1-year incidence of HF varied from 2% to 30% depending upon the definition criteria. Although no single criteria exceeded a 1-year incidence of 20%, a simple composite of HF diagnosis (International Classification of Diseases, Tenth Revision-I50) or use of loop diuretic produced a 1-year incidence 26.1% that was used as the benchmark outcome for evaluation of risk factors. Age ≥65 years, Black race, low-density lipoprotein ≥100 mg/100 ml, elevated hemoglobin A1c (7% to 9% and >9%), and body mass index≥35 kg/m2 were also associated with increased risk of HF. In conclusion, patients with anterior wall STEMI continue to be at high risk for new-onset HF. In the absence of structured, prospective, systematically adjudicated diagnostic criteria, composite definitions are more likely to yield accurate estimates of HF incidence.

Learning healthcare systems in cardiology: A qualitative interview study on ethical dilemmas of a learning healthcare system

Background

Implementation of an LHS in cardiology departments presents itself with ethical challenges, including ethical review and informed consent. In this qualitative study, we investigated stakeholders' attitudes toward ethical issues regarding the implementation of an LHS in the cardiology department.

Methods

We conducted a qualitative study using 35 semi-structured interviews and 5 focus group interviews with 34 individuals. We interviewed cardiologists, research nurses, cardiovascular patients, ethicists, health lawyers, epidemiologists/statisticians and insurance spokespersons.

Results

Respondents identified different ethical obstacles for the implementation of an LHS within the cardiology department. These obstacles were mainly on ethical oversight in LHSs; in particular, informed con sent and data ownership were discussed. In addition, respondents reported on the role of patients in LHS. Respondents described the LHS as a possibility for patients to engage in both research and care. While the LHS can promote patient engagement, patients might also be reduced to their data and are therefore at risk, according to respondents.

Conclusions

Views on the ethical dilemmas of a LHSs within cardiology are diverse. Similar to the literary debate on oversight, there are different views on how ethical oversight should be regulated. This study adds to the literary debate on oversight by highlighting that patients wish to be informed about the learning activities within the LHS they participate in, and that they wish to actively contribute by sharing their data and identifying learning goals, provided that informed consent is obtained.

The use and role of digital technology in learning health systems: A scoping review

OBJECTIVE

The review aimed to identify which digital technologies are proposed or used within learning health systems (LHS) and to analyze the extent to which they support learning processes in LHS.

MATERIALS AND METHODS

Multiple databases and grey literature were searched with terms related to LHS. Manual searches and backward searches of reference lists were also undertaken. The review considered publications from 2007 to 2022. Records focusing on LHS, referring to one or more digital technologies, and describing how at least one digital technology could be used in LHS were included.

RESULTS

2046 records were screened for inclusion and 154 records were included in the analysis. Twenty categories of digital technology were identified. The two most common ones across records were data recording and processing and electronic health records. Digital technology was primarily leveraged to support data access and aggregation and data analysis, two of the seven recognized learning processes within LHS learning cycles.

DISCUSSION

The results of the review show that a wide array of digital technologies is being leveraged to support learning cycles within LHS. Nevertheless, an over-reliance on a narrow set of technologies supporting knowledge discovery, a lack of direct evaluation of digital technologies and ambiguity in technology descriptions are hindering the realization of the LHS vision.

CONCLUSION

Future LHS research and initiatives should aim to integrate digital technology to support practice change and impact evaluation. The use of recognized evaluation methods for health information technology and more detailed descriptions of proposed technologies are also recommended.

Mobilizing registry data for quality improvement: A convergent mixed-methods analysis and application to spinal cord injury

Introduction

The rising prevalence of complex chronic conditions and growing intricacies of healthcare systems emphasizes the need for interdisciplinary partnerships to advance coordination and quality of rehabilitation care. Registry databases are increasingly used for clinical monitoring and quality improvement (QI) of health system change. Currently, it is unclear how interdisciplinary partnerships can best mobilize registry data to support QI across care settings for complex chronic conditions.

Purpose

We employed spinal cord injury (SCI) as a case study of a highly disruptive and debilitating complex chronic condition, with existing registry data that is underutilized for QI. We aimed to compare and converge evidence from previous reports and multi-disciplinary experts in order to outline the major elements of a strategy to effectively mobilize registry data for QI of care for complex chronic conditions.

Methods

This study used a convergent parallel-database variant mixed design, whereby findings from a systematic review and a qualitative exploration were analyzed independently and then simultaneously. The scoping review used a three-stage process to review 282 records, which resulted in 28 articles reviewed for analysis. Concurrent interviews were conducted with multidisciplinary-stakeholders, including leadership from condition-specific national registries, members of national SCI communities, leadership from SCI community organizations, and a person with lived experience of SCI. Descriptive analysis was used for the scoping review and qualitative description for stakeholder interviews.

Results

There were 28 articles included in the scoping review and 11 multidisciplinary-stakeholders in the semi-structured interviews. The integration of the results allowed the identification of three key learnings to enhance the successful design and use of registry data to inform the planning and development of a QI initiative: enhance utility and reliability of registry data; form a steering committee lead by clinical champions; and design effective, feasible, and sustainable QI initiatives.

Conclusion

This study highlights the importance of interdisciplinary partnerships to support QI of care for persons with complex conditions. It provides practical strategies to determine mutual priorities that promote implementation and sustained use of registry data to inform QI. Learnings from this work could enhance interdisciplinary collaboration to support QI of care for rehabilitation for persons with complex chronic conditions.

Stroke Learning Health Systems: A Topical Narrative Review With Case Examples

To our knowledge, the adoption of Learning Health System (LHS) concepts or approaches for improving stroke care, patient outcomes, and value have not previously been summarized. This topical review provides a summary of the published evidence about LHSs applied to stroke, and case examples applied to different aspects of stroke care from high and low-to-middle income countries. Our attempt to systematically identify the relevant literature and obtain real-world examples demonstrated the dissemination gaps, the lack of learning and action for many of the related LHS concepts across the continuum of care but also elucidated the opportunity for continued dialogue on how to study and scale LHS advances. In the field of stroke, we found only a few published examples of LHSs and health systems globally implementing some selected LHS concepts, but the term is not common. A major barrier to identifying relevant LHS examples in stroke may be the lack of an agreed taxonomy or terminology for classification. We acknowledge that health service delivery settings that leverage many of the LHS concepts do so operationally and the lessons learned are not shared in peer-reviewed literature. It is likely that this topical review will further stimulate the stroke community to disseminate related activities and use keywords such as learning health system so that the evidence base can be more readily identified.

Near Real-time Natural Language Processing for the Extraction of Abdominal Aortic Aneurysm Diagnoses From Radiology Reports: Algorithm Development and Validation Study

BACKGROUND

Management of abdominal aortic aneurysms (AAAs) requires serial imaging surveillance to evaluate the aneurysm dimension. Natural language processing (NLP) has been previously developed to retrospectively identify patients with AAA from electronic health records (EHRs). However, there are no reported studies that use NLP to identify patients with AAA in near real-time from radiology reports.

OBJECTIVE

This study aims to develop and validate a rule-based NLP algorithm for near real-time automatic extraction of AAA diagnosis from radiology reports for case identification.

METHODS

The AAA-NLP algorithm was developed and deployed to an EHR big data infrastructure for near real-time processing of radiology reports from May 1, 2019, to September 2020. NLP extracted named entities for AAA case identification and classified subjects as cases and controls. The reference standard to assess algorithm performance was a manual review of processed radiology reports by trained physicians following standardized criteria. Reviewers were blinded to the diagnosis of each subject. The AAA-NLP algorithm was refined in 3 successive iterations. For each iteration, the AAA-NLP algorithm was modified based on performance compared to the reference standard.

RESULTS

A total of 360 reports were reviewed, of which 120 radiology reports were randomly selected for each iteration. At each iteration, the AAA-NLP algorithm performance improved. The algorithm identified AAA cases in near real-time with high positive predictive value (0.98), sensitivity (0.95), specificity (0.98), F1 score (0.97), and accuracy (0.97).

CONCLUSIONS

Implementation of NLP for accurate identification of AAA cases from radiology reports with high performance in near real time is feasible. This NLP technique will support automated input for patient care and clinical decision support tools for the management of patients with AAA.  .

Optimizing cardiovascular risk assessment and registration in a developing cardiovascular learning health care system: Women benefit most

Since 2015 we organized a uniform, structured collection of a fixed set of cardiovascular risk factors according the (inter)national guidelines on cardiovascular risk management. We evaluated the current state of a developing cardiovascular towards learning healthcare system-the Utrecht Cardiovascular Cohort Cardiovascular Risk Management (UCC-CVRM)-and its potential effect on guideline adherence in cardiovascular risk management. We conducted a before-after study comparing data from patients included in UCC-CVRM (2015-2018) and patients treated in our center before UCC-CVRM (2013-2015) who would have been eligible for UCC-CVRM using the Utrecht Patient Oriented Database (UPOD). Proportions of cardiovascular risk factor measurement before and after UCC-CVRM initiation were compared, as were proportions of patients that required (change of) blood pressure, lipid, or blood glucose lowering treatment. We estimated the likelihood to miss patients with hypertension, dyslipidemia, and elevated HbA1c before UCC-CVRM for the whole cohort and stratified for sex. In the present study, patients included up to October 2018 (n = 1904) were matched with 7195 UPOD patients with similar age, sex, department of referral and diagnose description. Completeness of risk factor measurement increased, ranging from 0% -77% before to 82%-94% after UCC-CVRM initiation. Before UCC-CVRM, we found more unmeasured risk factors in women compared to men. This sex-gap resolved in UCC-CVRM. The likelihood to miss hypertension, dyslipidemia, and elevated HbA1c was reduced by 67%, 75% and 90%, respectively, after UCC-CVRM initiation. A finding more pronounced in women compared to men. In conclusion, a systematic registration of the cardiovascular risk profile substantially improves guideline adherent assessment and decreases the risk of missing patients with elevated levels with an indication for treatment. The sex-gap disappeared after UCC-CVRM initiation. Thus, an LHS approach contributes to a more inclusive insight into quality of care and prevention of cardiovascular disease (progression).

Impact of design characteristics among studies comparing coronary computed tomography angiography to noninvasive functional testing in chronic coronary syndromes

BACKGROUND

Coronary computed tomography angiography (CCTA) is widely adopted to detect obstructive coronary artery disease (CAD) in patients with chronic coronary syndromes (CCS). However, it is unknown to which extent study-specific characteristics yield different conclusions.

METHODS

We summarized non-randomized and randomized studies comparing CCTA and noninvasive functional testing for CCS with information on the outcome of myocardial infarction (MI). We evaluated the differential effect according to study characteristics using random-effect meta-analysis with Hartung-Knapp-Sidik-Jonkman adjustments.

RESULTS

Fifteen studies (8 non-randomized, 7 randomized) were included. CCTA was associated with decrease in relative (odds ratio (OR) 0.54, 95%CI 0.47 to 0.62, P < .001) and absolute MI risk (risk difference (RD) -0.4%, 95%CI -0.6 to -0.1, P = .005). The results remained consistent among the non-randomized (RD -0.4%, 95%CI -0.7 to -0.1, P=.029), but not among the randomized trials where there was no difference in the observed risk (RD 0.2%, 95%CI -0.6 to 0.1, P = .158). CCTA was not associated with MI reduction in studies with clinical outcome definition (OR 0.77, 95%CI 0.41 to 1.44, P = .212), research driven follow-up (OR 0.54, 95%CI 0.24 to 1.21, P = .090), central event assessment (OR 0.63, 95%CI 0.21 to 1.86, P = .207), outcome adjudication (OR 0.74, 95%CI 0.24 to 2.23, P = .178), or at low-risk of bias (OR 0.74, 95%CI 0.24 to 2.23, P = .178).

CONCLUSIONS

Among studies of any design, CCTA was associated with lower risk of MI in CCS compared to noninvasive functional testing. This benefit was diminished among studies with clinical outcome definition, central outcome assessment/adjudication or at low-risk of bias.

Applications of artificial intelligence for patients with peripheral artery disease

OBJECTIVE

Applications of artificial intelligence (AI) have been reported in several cardiovascular diseases but its interest in patients with peripheral artery disease (PAD) has been so far less reported. The aim of this review was to summarize current knowledge on applications of AI in patients with PAD, to discuss current limits, and highlight perspectives in the field.

METHODS

We performed a narrative review based on studies reporting applications of AI in patients with PAD. The MEDLINE database was independently searched by two authors using a combination of keywords to identify studies published between January 1995 and December 2021. Three main fields of AI were investigated including natural language processing (NLP), computer vision and machine learning (ML).

RESULTS

NLP and ML brought new tools to improve the screening, the diagnosis and classification of the severity of PAD. ML was also used to develop predictive models to better assess the prognosis of patients and develop real-time prediction models to support clinical decision-making. Studies related to computer vision mainly aimed at creating automatic detection and characterization of arterial lesions based on Doppler ultrasound examination or computed tomography angiography. Such tools could help to improve screening programs, enhance diagnosis, facilitate presurgical planning, and improve clinical workflow.

CONCLUSIONS

AI offers various applications to support and likely improve the management of patients with PAD. Further research efforts are needed to validate such applications and investigate their accuracy and safety in large multinational cohorts before their implementation in daily clinical practice.

Stepwise implementation of a cardiovascular risk management care program in primary care

Background

Primary care plays a pivotal role in sustainable cardiovascular risk management (CVRM) but little is known about the organizational process of implementing the guidelines. The aim of the study was to describe the approach taken by a primary care group to implement the CVRM guideline.

Methods

Stepwise introduction and implementation of a programmatic CVRM care program was organized and facilitated by the care group between April 2010 and January 2013 in 137 affiliated general practices with 188 general practitioners (GPs), in the vicinity of Eindhoven, Netherlands. Care group support comprised sufficient staff, support with data extraction based on ICPC and ATC codes and with identification of eligible patients by scrutinizing patient health records and adequate coding of disease conditions.

Results

Patient selection based on availability of structured information on ICPC codes and risk factor levels from the electronic health records, led to 38,675 eligible patients in 2013. December 2019, the CVRM program was still running in 151 practices with 51,416 patients receiving programmatic CVRM care. Linking problems between 8 different electronic health record systems and the multidisciplinary information system for integrated care delayed adequate data collection until the beginning of 2013.

Conclusion

Commitment of affiliated GPs, a structured approach with adequate coding of diagnoses and risk factors, central data registration and additional funding for sufficient staff support are important conditions for the introduction and implementation of successful and sustainable programmatic CVRM care. This approach constitutes the basis for long-term follow up and annual evaluation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12875-021-01602-w.

Patient and Clinician Perceptions of Precision Cardiology Care: Findings From the HeartCare Study

BACKGROUND

Routine genome-wide screening for cardiovascular disease risk may inform clinical decision-making. However, little is known about whether clinicians and patients would find such testing useful or acceptable within the context of a genomics-enabled learning health system.

METHODS

We conducted surveys with patients and their clinicians who were participating in the HeartCare Study, a precision cardiology care project that returned results from a next-generation sequencing panel of 158 genes associated with cardiovascular disease risk. Six weeks after return of results, we assessed patients' and clinicians' perceived utility and disutility of HeartCare, the effect of the test on clinical recommendations, and patients' attitudes toward integration of research and clinical care.

RESULTS

Among 666 HeartCare patients with a result returned during the survey study period, 42.0% completed a full or partial survey. Patient-participants who completed a full survey (n=224) generally had positive perceptions of HeartCare independent of whether they received a positive or negative result. Most patient-participants considered genetic testing for cardiovascular disease risk to have more benefit than risk (88.3%) and agreed that it provided information that they wanted to know (81.2%), while most disagreed that the test caused them to feel confused (77.7%) or overwhelmed (78.0%). For 122 of their patients with positive results, clinicians (n=13) reported making changes in clinical care for 66.4% of patients, recommending changes in health behaviors for 36.9% of patients, and recommending to 33.6% of patients that their family members have clinical testing.

CONCLUSIONS

Both patients and clinicians thought the HeartCare panel screen for cardiovascular disease risk provided information that was useful in terms of personal or health benefits to the patient and that informed clinical care without causing patients to be confused or overwhelmed. Further research is needed to assess perceptions of genome-wide screening among the US cardiology clinic population.

Natural language processing for identification of hypertrophic cardiomyopathy patients from cardiac magnetic resonance reports

Background

Cardiac magnetic resonance (CMR) imaging is important for diagnosis and risk stratification of hypertrophic cardiomyopathy (HCM) patients. However, collection of information from large numbers of CMR reports by manual review is time-consuming, error-prone and costly. Natural language processing (NLP) is an artificial intelligence method for automated extraction of information from narrative text including text in CMR reports in electronic health records (EHR). Our objective was to assess whether NLP can accurately extract diagnosis of HCM from CMR reports.

Methods

An NLP system with two tiers was developed for information extraction from narrative text in CMR reports; the first tier extracted information regarding HCM diagnosis while the second extracted categorical and numeric concepts for HCM classification. We randomly allocated 200 HCM patients with CMR reports from 2004 to 2018 into training (100 patients with 185 CMR reports) and testing sets (100 patients with 206 reports).

Results

NLP algorithms demonstrated very high performance compared to manual annotation. The algorithm to extract HCM diagnosis had accuracy of 0.99. The accuracy for categorical concepts included HCM morphologic subtype 0.99, systolic anterior motion of the mitral valve 0.96, mitral regurgitation 0.93, left ventricular (LV) obstruction 0.94, location of obstruction 0.92, apical pouch 0.98, LV delayed enhancement 0.93, left atrial enlargement 0.99 and right atrial enlargement 0.98. Accuracy for numeric concepts included maximal LV wall thickness 0.96, LV mass 0.99, LV mass index 0.98, LV ejection fraction 0.98 and right ventricular ejection fraction 0.99.

Conclusions

NLP identified and classified HCM from CMR narrative text reports with very high performance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-02017-y.

Translating periodontal data to knowledge in a learning health system

BACKGROUND

A learning health system (LHS) is a health system in which patients and clinicians work together to choose care on the basis of best evidence and to drive discovery as a natural outgrowth of every clinical encounter to ensure the right care at the right time. An LHS for dentistry is now feasible, as an increased number of oral health care encounters are captured in electronic health records (EHRs).

METHODS

The authors used EHRs data to track periodontal health outcomes at 3 large dental institutions. The 2 outcomes of interest were a new periodontitis case (for patients who had not received a diagnosis of periodontitis previously) and tooth loss due to progression of periodontal disease.

RESULTS

The authors assessed a total of 494,272 examinations (new periodontitis outcome: n = 168,442; new tooth loss outcome: n = 325,830), representing a total of 194,984 patients. Dynamic dashboards displaying performance on both measures over time allow users to compare demographic and risk factors for patients. The incidence of new periodontitis and tooth loss was 4.3% and 1.2%, respectively.

CONCLUSIONS

Periodontal disease, diagnosis, prevention, and treatment are particularly well suited for an LHS model. The results showed the feasibility of automated extraction and interpretation of critical data elements from the EHRs. The 2 outcome measures are being implemented as part of a dental LHS. The authors are using this knowledge to target the main drivers of poorer periodontal outcomes in a specific patient population, and they continue to use clinical health data for the purpose of learning and improvement.

PRACTICAL IMPLICATIONS

Dental institutions of any size can conduct contemporaneous self-evaluation and immediately implement targeted strategies to improve oral health outcomes.

Augmented Intelligence to Identify Patients With Advanced Heart Failure in an Integrated Health System

BACKGROUND

Timely referral for specialist evaluation in patients with advanced heart failure (HF) is a Class 1 recommendation. However, the transition from stage C HF to advanced or stage D HF often goes undetected in routine care, resulting in delayed referral and higher mortality rates.

OBJECTIVES

The authors sought to develop an augmented intelligence-enabled workflow using machine learning to identify patients with stage D HF and streamline referral.

METHODS

We extracted data on HF patients with encounters from January 1, 2007, to November 30, 2020, from a HF registry within a regional, integrated health system. We created an ensemble machine learning model to predict stage C or stage D HF and integrated the results within the electronic health record.

RESULTS

In a retrospective data set of 14,846 patients, the model had a good positive predictive value (60%) and low sensitivity (25%) for identifying stage D HF in a 100-person, physician-reviewed, holdout test set. During prospective implementation of the workflow from April 1, 2021, to February 15, 2022, 416 patients were reviewed by a clinical coordinator, with agreement between the model and the coordinator in 50.3% of stage D predictions. Twenty-four patients have been scheduled for evaluation in a HF clinic, 4 patients started an evaluation for advanced therapies, and 1 patient received a left ventricular assist device.

CONCLUSIONS

An augmented intelligence-enabled workflow was integrated into clinical operations to identify patients with advanced HF. Endeavors such as this require a multidisciplinary team with experience in design thinking, informatics, quality improvement, operations, and health information technology, as well as dedicated resources to monitor and improve performance over time.

Connected Health Innovation Research Program (C.H.I.R.P.): A bridge for digital health and wellness in cardiology and oncology

Study objective

Cancer and heart disease are leading causes of mortality, and cardio-oncology is emerging as a new field addressing the cardiovascular toxicities related to cancer and cancer therapy. Interdisciplinary research platforms that incorporate digital health to optimize cardiovascular health and wellness in cancer survivors are therefore needed as we advance in the digital era. Our goal was to develop the Connected Health Innovation Research Program (C.H.I.R.P.) to serve as a foundation for future integration and assessments of adoption and clinical efficacy of digital health tools for cardiovascular health and wellness in the general population and in oncology patients.

Design/setting/participants

Partner companies were identified through the American Medical Association innovation platform, as well as LinkedIn and direct contact by our team. Company leaders met with our team to discuss features of their technology or software. Non-disclosure agreements were signed and data were discussed and obtained for descriptive or statistical analysis.

Results

A suite of companies with technologies focused on wellness, biometrics tracking, audio companions, oxygen saturation, weight trends, sleep patterns, heart rate variability, electrocardiogram patterns, blood pressure patterns, real-time metabolism tracking, instructional video modules, or integration of these technologies into electronic health records was collated. We formed an interdisciplinary research team and established an academia-industry collaborative foundation for connecting patients with wellness digital health technologies.

Conclusions

A suite of software and device technologies accessible to the cardiology and oncology population has been established and will facilitate retrospective, prospective, and case research studies assessing adoption and clinical efficacy of digital health tools in cardiology/oncology.

Conceptualizing learning health systems: A mapping review

Introduction

Health systems worldwide face the challenge of increasing population health with high-quality care and reducing health care expenditure growth. In pursuit for a solution, regional cross-sectoral partnerships aim to reorganize and integrate services across public health, health care and social care. Although the complexity of regional partnerships demands an incremental strategy, it is yet not known how learning works within these partnerships. To understand learning in regional cross-sectoral partnerships for health, this study aims to map the concept Learning Health System (LHS).

Methods

This mapping review used a qualitative text analysis approach. A literature search was conducted in Embase and was limited to English-language papers published in the period 2015-2020. Title-abstract screening was performed using established exclusion criteria. During full-text screening, we combined deductive and inductive coding. The concept LHS was disentangled into aims, design elements, and process of learning. Data extraction and analysis were performed in MAX QDA 2020.

Results

In total, 155 articles were included. All articles used the LHS definition of the Institute of Medicine. The interpretation of the concept LHS varied widely. The description of LHS contained 25 highly connected aims. In addition, we identified nine design elements. Most elements were described similarly, only the interpretation of stakeholders, data infrastructure and data varied. Furthermore, we identified three types of learning: learning as 1) interaction between clinical practice and research; 2) a circular process of converting routine care data to knowledge, knowledge to performance; and performance to data; and 3) recurrent interaction between stakeholders to identify opportunities for change, to reveal underlying values, and to evaluate processes. Typology 3 was underrepresented, and the three types of learning rarely occurred simultaneously.

Conclusion

To understand learning within regional cross-sectoral partnerships for health, we suggest to specify LHS-aim(s), operationalize design elements, and choose deliberately appropriate learning type(s).

The pursuit of health equity in digital transformation, health informatics, and the cardiovascular learning healthcare system

African Americans have a higher rate of cardiovascular morbidity and mortality and a lower rate of specialty consultation and treatment than Caucasians. These disparities also exist in the care and treatment of chemotherapy-related cardiovascular complications. African Americans suffer from cardiotoxicity at a higher rate than Caucasians and are underrepresented in clinical trials aimed at preventing cardiovascular injury associated with cancer therapies. To eliminate racial and ethnic disparities in the prevention of cardiotoxicity, an interdisciplinary and innovative approach will be required. Diverse forms of digital transformation leveraging health informatics have the potential to contribute to health equity if they are implemented carefully and thoughtfully in collaboration with minority communities. A learning healthcare system can serve as a model for developing, deploying, and disseminating interventions to minimize health inequities and maximize beneficial impact.

Use of Multi-Modal Data and Machine Learning to Improve Cardiovascular Disease Care

Today's digital health revolution aims to improve the efficiency of healthcare delivery and make care more personalized and timely. Sources of data for digital health tools include multiple modalities such as electronic medical records (EMR), radiology images, and genetic repositories, to name a few. While historically, these data were utilized in silos, new machine learning (ML) and deep learning (DL) technologies enable the integration of these data sources to produce multi-modal insights. Data fusion, which integrates data from multiple modalities using ML and DL techniques, has been of growing interest in its application to medicine. In this paper, we review the state-of-the-art research that focuses on how the latest techniques in data fusion are providing scientific and clinical insights specific to the field of cardiovascular medicine. With these new data fusion capabilities, clinicians and researchers alike will advance the diagnosis and treatment of cardiovascular diseases (CVD) to deliver more timely, accurate, and precise patient care.

Clinical outcomes and patient satisfaction with the use of biological and synthetic meshes in one-stage implant-based breast reconstruction

BACKGROUND

Biological and synthetic meshes were used to cover the damaged muscle and augment the subpectoral pocket in breast reconstruction. However, few studies have directly compared the effects of biological and synthetic meshes. This study analyzed postoperative complications and assessed the patient-reported outcomes with the use of BioDesign® Surgisis and TiLOOP Bra/TiMesh® in one-stage implant-based breast reconstruction.

METHODS

Patients undergoing one-stage implant-based breast reconstruction were enrolled in this study. Post-mastectomy breast reconstructions were facilitated with either Surgisis mesh or TiLOOP mesh. Complications were examined and patient-reported quality-of-life outcomes were evaluated using the BREAST-Q questionnaire (ver 2.0). The multivariate linear regression models were used for data analysis.

RESULTS

Overall, 79 of 116 patients (68%) received breast reconstruction with Surgisis mesh and 37 (32%) with TiLOOP mesh. There was no difference in complication rates between the two groups postoperatively. But patient-reported satisfaction was higher with the use of Surgisis mesh than with TiLOOP mesh (P = 0.05).

CONCLUSIONS

This study reported no difference between the Surgisis group and the TiLOOP group in either complication rates or most patient-reported outcomes postoperatively. Yet the assessment of patient-reported satisfaction showed preference toward Surgisis mesh, a finding with a potential implication for mesh selection.

A Systems-Based Morbidity and Mortality Conference Was Associated With a Transient Reduction in ECG Completion Times

BACKGROUND AND OBJECTIVES

During its monthly morbidity and mortality conference (MMC), the University of Colorado Division of Cardiology reviewed a "near-miss" patient safety event involving the delayed completion of a Stat-priority (ie, statim, meaning high priority) electrocardiogram (ECG). Because critical and interprofessional stakeholders participated in the conference, we hypothesized that the MMC would be associated with reduced ECG completion times.

METHODS

Data were collected for in-hospital ECGs performed at the University of Colorado Hospital between January 1, 2017, and June 30, 2018. An interrupted time series analysis was used to estimate the immediate and ongoing impact of the MMC (held on February 28, 2018) on ECG completion times, stratified by order priority (Stat, Now, or Routine). The percentage of delayed Stat-priority ECGs was analyzed as a secondary outcome.

RESULTS

Before the MMC, ECG completion times were stable for all order priorities ( P > .2), but the proportion of delayed Stat-priority ECGs increased from 5% in January 2017 to 20% in February 2018 ( P < .01). The MMC was associated with an immediate reduction in average daily ECG completion times for Routine (-18.4 minutes, P = .03) and Now (-8 minutes, P = .024) priority ECGs. No reduction was seen for Stat ECGs ( P = .97), though the percentage of delayed Stat ECGs stopped increasing ( P = .63). In the post-MMC period, completion times for Routine-priority ECGs increased and approached pre-MMC levels.

CONCLUSIONS

The MMC was associated with an immediate, but temporary, improvement in ECG completion times. Although the observed clinical benefit of the MMC is novel, these data support the need for more durable reforms to sustain initial improvements.

Establishing an interdisciplinary research team for cardio-oncology artificial intelligence informatics precision and health equity

Study objective

A multi-institutional interdisciplinary team was created to develop a research group focused on leveraging artificial intelligence and informatics for cardio-oncology patients. Cardio-oncology is an emerging medical field dedicated to prevention, screening, and management of adverse cardiovascular effects of cancer/ cancer therapies. Cardiovascular disease is a leading cause of death in cancer survivors. Cardiovascular risk in these patients is higher than in the general population. However, prediction and prevention of adverse cardiovascular events in individuals with a history of cancer/cancer treatment is challenging. Thus, establishing an interdisciplinary team to create cardiovascular risk stratification clinical decision aids for integration into electronic health records for oncology patients was considered crucial.

Design/setting/participants

Core team members from the Medical College of Wisconsin (MCW), University of Wisconsin-Milwaukee (UWM), and Milwaukee School of Engineering (MSOE), and additional members from Cleveland Clinic, Mayo Clinic, and other institutions have joined forces to apply high-performance computing in cardio-oncology.

Results

The team is comprised of clinicians and researchers from relevant complementary and synergistic fields relevant to this work. The team has built an epidemiological cohort of ~5000 cancer survivors that will serve as a database for interdisciplinary multi-institutional artificial intelligence projects.

Conclusion

Lessons learned from establishing this team, as well as initial findings from the epidemiology cohort, are presented. Barriers have been broken down to form a multi-institutional interdisciplinary team for health informatics research in cardio-oncology. A database of cancer survivors has been created collaboratively by the team and provides initial insight into cardiovascular outcomes and comorbidities in this population.

Maternal Cardiovascular Disease After Pre-Eclampsia and Gestational Hypertension: A Narrative Review

Previous literature has highlighted that women who have a pregnancy affected by gestational hypertension or preeclampsia are at higher risk of cardiovascular disease (CVD) in later life. However, CVD is a composite of multiple outcomes, including coronary heart disease, heart failure, and stroke, and the risk of both CVD and hypertensive disorders of pregnancy varies by the population studied. We conducted a narrative review of the risk of cardiovascular outcomes for women with prior gestational hypertension and pre-eclampsia. Previous literature is summarized by country and ethnicity, with a higher risk of CVD and coronary heart disease observed after gestational hypertension and a higher risk of CVD, coronary heart disease and heart failure observed after pre-eclampsia in most of the populations studied. Only one study was identified in a low- or middle-income country, and the majority of studies were conducted in white or mixed ethnicity populations. We discuss potential interventions to mitigate cardiovascular risk for these women in different settings and highlight the need for a greater understanding of the epidemiology of CVD risk after gestational hypertension and pre-eclampsia outside of high-income, white populations.

Learning healthcare systems and rapid learning in radiation oncology: Where are we and where are we going?

Learning health systems and rapid-learning are well developed at the conceptual level. The promise of rapidly generating and applying evidence where conventional clinical trials would not usually be practical is attractive in principle. The connectivity of modern digital healthcare information systems and the increasing volumes of data accrued through patients' care pathways offer an ideal platform for the concepts. This is particularly true in radiotherapy where modern treatment planning and image guidance offers a precise digital record of the treatment planned and delivered. The vision is of real-world data, accrued by patients during their routine care, being used to drive programmes of continuous clinical improvement as part of standard practice. This vision, however, is not yet a reality in radiotherapy departments. In this article we review the literature to explore why this is not the case, identify barriers to its implementation, and suggest how wider clinical application might be achieved.

Clinical prediction models for hospital falls: a scoping review protocol

INTRODUCTION

Falls remain one of the most prevalent adverse events in hospitals and are associated with substantial negative health impacts and costs. Approaches to assess patients' fall risk have been implemented in hospitals internationally, ranging from brief screening questions to multifactorial risk assessments and complex prediction models, despite a lack of clear evidence of effect in reducing falls in acute hospital environments. The increasing digitisation of hospital systems provides new opportunities to understand and predict falls using routinely recorded data, with potential to integrate fall prediction models into real-time or near-real-time computerised decision support for clinical teams seeking to mitigate fall risk. However, the use of non-traditional approaches to fall risk prediction, including machine learning using integrated electronic medical records, has not yet been reviewed relative to more traditional fall prediction models. This scoping review will summarise methodologies used to develop existing hospital fall prediction models, including reporting quality assessment.

METHODS AND ANALYSIS

This scoping review will follow the Arksey and O'Malley framework and its recent advances, and will be reported using Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews recommendations. Four electronic databases (CINAHL via EBSCOhost, PubMed, IEEE Xplore and Embase) will be initially searched for studies up to 12 November 2020, and searches may be updated prior to final reporting. Additional studies will be identified by reference list review and citation analysis of included studies. No restriction will be placed on the date or language of identified studies. Screening of search results and extraction of data will be performed by two independent reviewers. Reporting quality will be assessed by the adherence to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis.

ETHICS AND DISSEMINATION

Ethical approval is not required for this study. Findings will be disseminated through peer-reviewed publication and scientific conferences.

Clarifying the concept of a learning health system for healthcare delivery organizations: Implications from a qualitative analysis of the scientific literature

The “learning health system” (LHS) concept has been defined in broad terms, which makes it challenging for health system leaders to determine exactly what is required to transform their organization into an LHS. This study provides a conceptual map of the LHS landscape by identifying the activities, principles, tools, and conditions that LHS researchers have associated with the concept. Through a multi‐step screening process, two researchers identified 79 publications from PubMed (published before January 2020) that contained information relevant to the question, “What work is required of a healthcare organization that is operating as an LHS?” Those publications were coded as to whether or not they referenced each of 94 LHS elements in the taxonomy developed by the study team. This taxonomy, named the Learning Health Systems Consolidated Framework (LHS‐CF), organizes the elements into five “bodies of work” (organizational learning, translation of evidence into practice, building knowledge, analyzing clinical data, and engaging stakeholders) and four “enabling conditions” (workforce skilled for LHS work, data systems and informatics technology in place, organization invests resources in LHS work, and supportive organizational culture). We report the frequency that each of the 94 elements was referenced across the 79 publications. The four most referenced elements were: “organization builds knowledge or evidence,” “quality improvement practices are standard practice,” “patients and family members are actively engaged,” and “organizational culture emphasizes and supports learning.” By dissecting the LHS construct into its component elements, the LHS‐CF taxonomy can serve as a useful tool for LHS researchers and practitioners in defining the aspects of LHS they are addressing. By assessing how often each element is referenced in the literature, the study provides guidance to health system leaders as to how their organization needs to evolve in order to become an LHS ‐ while also recognizing that each organization should emphasize elements that are most aligned with their mission and goals.

Towards a Responsible Transition to Learning Healthcare Systems in Precision Medicine: Ethical Points to Consider

Learning healthcare systems have recently emerged as a strategy to continuously use experiences and outcomes of clinical care for research purposes in precision medicine. Although it is known that learning healthcare transitions in general raise important ethical challenges, the ethical ramifications of such transitions in the specific context of precision medicine have not extensively been discussed. Here, we describe three levers that institutions can pull to advance learning healthcare systems in precision medicine: (1) changing testing of individual variability (such as genes); (2) changing prescription of treatments on the basis of (genomic) test results; and/or (3) changing the handling of data that link variability and treatment to clinical outcomes. Subsequently, we evaluate how patients can be affected if one of these levers are pulled: (1) patients are tested for different or more factors than before the transformation, (2) patients receive different treatments than before the transformation and/or (3) patients’ data obtained through clinical care are used, or used more extensively, for research purposes. Based on an analysis of the aforementioned mechanisms and how these potentially affect patients, we analyze why learning healthcare systems in precision medicine need a different ethical approach and discuss crucial points to consider regarding this approach.

User stories as lightweight requirements for agile clinical decision support development

Objective

We sought to demonstrate applicability of user stories, progressively elaborated by testable acceptance criteria, as lightweight requirements for agile development of clinical decision support (CDS).

Materials and Methods

User stories employed the template: As a [type of user], I want [some goal] so that [some reason]. From the “so that” section, CDS benefit measures were derived. Detailed acceptance criteria were elaborated through ensuing conversations. We estimated user story size with “story points,” and depicted multiple user stories with a use case diagram or feature breakdown structure. Large user stories were split to fit into 2-week iterations.

Results

One example user story was: As a rheumatologist, I want to be advised if my patient with rheumatoid arthritis is not on a disease-modifying anti-rheumatic drug (DMARD), so that they receive optimal therapy and can experience symptom improvement. This yielded a process measure (DMARD use), and an outcome measure (Clinical Disease Activity Index). Following implementation, the DMARD nonuse rate decreased from 3.7% to 1.4%. Patients with a high Clinical Disease Activity Index improved from 13.7% to 7%. For a thromboembolism prevention CDS project, diagrams organized multiple user stories.

Discussion

User stories written in the clinician’s voice aid CDS governance and lead naturally to measures of CDS effectiveness. Estimation of relative story size helps plan CDS delivery dates. User stories prove to be practical even on larger projects.

Conclusions

User stories concisely communicate the who, what, and why of a CDS request, and serve as lightweight requirements for agile development to meet the demand for increasingly diverse CDS.

Achieving Optimal Population Cardiovascular Health Requires an Interdisciplinary Team and a Learning Healthcare System: A Scientific Statement From the American Heart Association

Population cardiovascular health, or improving cardiovascular health among patients and the population at large, requires a redoubling of primordial and primary prevention efforts as declines in cardiovascular disease mortality have decelerated over the past decade. Great potential exists for healthcare systems-based approaches to aid in reversing these trends. A learning healthcare system, in which population cardiovascular health metrics are measured, evaluated, intervened on, and re-evaluated, can serve as a model for developing the evidence base for developing, deploying, and disseminating interventions. This scientific statement on optimizing population cardiovascular health summarizes the current evidence for such an approach; reviews contemporary sources for relevant performance and clinical metrics; highlights the role of implementation science strategies; and advocates for an interdisciplinary team approach to enhance the impact of this work.

CREATE: A New Data Resource to Support Cardiac Precision Health

Background

The initiatives of precision medicine and learning health systems require databases with rich and accurately captured data on patient characteristics. We introduce the Clinical Registry, AdminisTrative Data and Electronic Medical Records (CREATE) database, which includes linked data from 4 population databases: Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH; a national clinical registry), Sunrise Clinical Manager (SCM) electronic medical record (city-wide), the Discharge Abstract Database (DAD), and the National Ambulatory Care Reporting System (NACRS). The intent of this work is to introduce a cardiovascular-specific database for pursuing precision health activities using big data analytics.

Methods

We used deterministic data linkage to link SCM electronic medical record data to APPROACH clinical registry data using patient identifier variables. The APPROACH-SCM data set was subsequently linked to DAD and NACRS to obtain inpatient and outpatient cohort data. We further validated the quality of the linkage, where applicable, in these databases by comparing against the Alberta Health Insurance Care Plan registry database.

Results

We achieved 99.96% linkage across these 4 databases. Currently, there are 30,984 patients with 35,753 catheterizations in the CREATE database. The inpatient cohort contained 65.75% (20,373/30,984) of the patient sample, whereas the outpatient cohort contained 29.78% (9226/30,984). The infrastructure and the process to update and expand the database has been established.

Conclusions

CREATE is intended to serve as a database for supporting big data analytics activities surrounding cardiac precision health. The CREATE database will be managed by the Centre for Health Informatics at the University of Calgary, and housed in a secure high-performance computing environment.

Standardized definitions for evaluation of heart failure therapies: scientific expert panel from the Heart Failure Collaboratory and Academic Research Consortium

The Heart Failure Academic Research Consortium is a partnership between the Heart Failure Collaboratory (HFC) and Academic Research Consortium (ARC), comprised of leading heart failure (HF) academic research investigators, patients, United States (US) Food and Drug Administration representatives, and industry members from the US and Europe. A series of meetings were convened to establish definitions and key concepts for the evaluation of HF therapies including optimal medical and device background therapy, clinical trial design elements and statistical concepts, and study endpoints. This manuscript summarizes the expert panel discussions as consensus recommendations focused on populations and endpoint definitions; it is not exhaustive or restrictive, but designed to stimulate HF clinical trial innovation.

Phenotyping severity of patient-centered outcomes using clinical notes: A prostate cancer use case

Introduction

A learning health system (LHS) must improve care in ways that are meaningful to patients, integrating patient‐centered outcomes (PCOs) into core infrastructure. PCOs are common following cancer treatment, such as urinary incontinence (UI) following prostatectomy. However, PCOs are not systematically recorded because they can only be described by the patient, are subjective and captured as unstructured text in the electronic health record (EHR). Therefore, PCOs pose significant challenges for phenotyping patients. Here, we present a natural language processing (NLP) approach for phenotyping patients with UI to classify their disease into severity subtypes, which can increase opportunities to provide precision‐based therapy and promote a value‐based delivery system.

Methods

Patients undergoing prostate cancer treatment from 2008 to 2018 were identified at an academic medical center. Using a hybrid NLP pipeline that combines rule‐based and deep learning methodologies, we classified positive UI cases as mild, moderate, and severe by mining clinical notes.

Results

The rule‐based model accurately classified UI into disease severity categories (accuracy: 0.86), which outperformed the deep learning model (accuracy: 0.73). In the deep learning model, the recall rates for mild and moderate group were higher than the precision rate (0.78 and 0.79, respectively). A hybrid model that combined both methods did not improve the accuracy of the rule‐based model but did outperform the deep learning model (accuracy: 0.75).

Conclusion

Phenotyping patients based on indication and severity of PCOs is essential to advance a patient centered LHS. EHRs contain valuable information on PCOs and by using NLP methods, it is feasible to accurately and efficiently phenotype PCO severity. Phenotyping must extend beyond the identification of disease to provide classification of disease severity that can be used to guide treatment and inform shared decision‐making. Our methods demonstrate a path to a patient centered LHS that could advance precision medicine.

Health Outcomes from Home Hospitalization: Multisource Predictive Modeling

BACKGROUND

Home hospitalization is widely accepted as a cost-effective alternative to conventional hospitalization for selected patients. A recent analysis of the home hospitalization and early discharge (HH/ED) program at Hospital Clínic de Barcelona over a 10-year period demonstrated high levels of acceptance by patients and professionals, as well as health value-based generation at the provider and health-system levels. However, health risk assessment was identified as an unmet need with the potential to enhance clinical decision making.

OBJECTIVE

The objective of this study is to generate and assess predictive models of mortality and in-hospital admission at entry and at HH/ED discharge.

METHODS

Predictive modeling of mortality and in-hospital admission was done in 2 different scenarios: at entry into the HH/ED program and at discharge, from January 2009 to December 2015. Multisource predictive variables, including standard clinical data, patients' functional features, and population health risk assessment, were considered.

RESULTS

We studied 1925 HH/ED patients by applying a random forest classifier, as it showed the best performance. Average results of the area under the receiver operating characteristic curve (AUROC; sensitivity/specificity) for the prediction of mortality were 0.88 (0.81/0.76) and 0.89 (0.81/0.81) at entry and at home hospitalization discharge, respectively; the AUROC (sensitivity/specificity) values for in-hospital admission were 0.71 (0.67/0.64) and 0.70 (0.71/0.61) at entry and at home hospitalization discharge, respectively.

CONCLUSIONS

The results showed potential for feeding clinical decision support systems aimed at supporting health professionals for inclusion of candidates into the HH/ED program, and have the capacity to guide transitions toward community-based care at HH discharge.

Implementation of a Cohort Retrieval System for Clinical Data Repositories Using the Observational Medical Outcomes Partnership Common Data Model: Proof-of-Concept System Validation

BACKGROUND

Widespread adoption of electronic health records has enabled the secondary use of electronic health record data for clinical research and health care delivery. Natural language processing techniques have shown promise in their capability to extract the information embedded in unstructured clinical data, and information retrieval techniques provide flexible and scalable solutions that can augment natural language processing systems for retrieving and ranking relevant records.

OBJECTIVE

In this paper, we present the implementation of a cohort retrieval system that can execute textual cohort selection queries on both structured data and unstructured text-Cohort Retrieval Enhanced by Analysis of Text from Electronic Health Records (CREATE).

METHODS

CREATE is a proof-of-concept system that leverages a combination of structured queries and information retrieval techniques on natural language processing results to improve cohort retrieval performance using the Observational Medical Outcomes Partnership Common Data Model to enhance model portability. The natural language processing component was used to extract common data model concepts from textual queries. We designed a hierarchical index to support the common data model concept search utilizing information retrieval techniques and frameworks.

RESULTS

Our case study on 5 cohort identification queries, evaluated using the precision at 5 information retrieval metric at both the patient-level and document-level, demonstrates that CREATE achieves a mean precision at 5 of 0.90, which outperforms systems using only structured data or only unstructured text with mean precision at 5 values of 0.54 and 0.74, respectively.

CONCLUSIONS

The implementation and evaluation of Mayo Clinic Biobank data demonstrated that CREATE outperforms cohort retrieval systems that only use one of either structured data or unstructured text in complex textual cohort queries.

Venous Thromboembolism Research Priorities: A Scientific Statement From the American Heart Association and the International Society on Thrombosis and Haemostasis

Venous thromboembolism is a major cause of morbidity and mortality. The impact of the US Surgeon General's The Surgeon General's Call to Action to Prevent Deep Vein Thrombosis and Pulmonary Embolism in 2008 has been lower than expected given the public health impact of this disease. This scientific statement highlights future research priorities in venous thromboembolism, developed by experts and a crowdsourcing survey across 16 scientific organizations. At the fundamental research level (T0), researchers need to identify pathobiological causative mechanisms for the 50% of patients with unprovoked venous thromboembolism and to better understand mechanisms that differentiate hemostasis from thrombosis. At the human level (T1), new methods for diagnosing, treating, and preventing venous thromboembolism will allow tailoring of diagnostic and therapeutic approaches to individuals. At the patient level (T2), research efforts are required to understand how foundational evidence impacts care of patients (eg, biomarkers). New treatments, such as catheter-based therapies, require further testing to identify which patients are most likely to experience benefit. At the practice level (T3), translating evidence into practice remains challenging. Areas of overuse and underuse will require evidence-based tools to improve care delivery. At the community and population level (T4), public awareness campaigns need thorough impact assessment. Large population-based cohort studies can elucidate the biological and environmental underpinnings of venous thromboembolism and its complications. To achieve these goals, funding agencies and training programs must support a new generation of scientists and clinicians who work in multidisciplinary teams to solve the pressing public health problem of venous thromboembolism.

Venous thromboembolism research priorities: A scientific statement from the American Heart Association and the International Society on Thrombosis and Haemostasis

Venous thromboembolism (VTE) is a major cause of morbidity and mortality. The impact of the Surgeon General's Call to Action in 2008 has been lower than expected given the public health impact of this disease. This scientific statement highlights future research priorities in VTE, developed by experts and a crowdsourcing survey across 16 scientific organizations. At the fundamental research level (T0), researchers need to identify pathobiologic causative mechanisms for the 50% of patients with unprovoked VTE and better understand mechanisms that differentiate hemostasis from thrombosis. At the human level (T1), new methods for diagnosing, treating, and preventing VTE will allow tailoring of diagnostic and therapeutic approaches to individuals. At the patient level (T2), research efforts are required to understand how foundational evidence impacts care of patients (eg, biomarkers). New treatments, such as catheter-based therapies, require further testing to identify which patients are most likely to experience benefit. At the practice level (T3), translating evidence into practice remains challenging. Areas of overuse and underuse will require evidence-based tools to improve care delivery. At the community and population level (T4), public awareness campaigns need thorough impact assessment. Large population-based cohort studies can elucidate the biologic and environmental underpinings of VTE and its complications. To achieve these goals, funding agencies and training programs must support a new generation of scientists and clinicians who work in multidisciplinary teams to solve the pressing public health problem of VTE.

Learning health care systems: Highly needed but challenging

BACKGROUND

Learning health care systems (LHSs) have the potential to transform health care. However, this transformation process faces significant challenges.

MATERIALS AND METHODS

Based on proposals and early examples of LHSs in the literature and conceptual analysis of the LHS mission, we provide four models with distinct organizational and ethical implications that may facilitate the transformation.

RESULTS

An LHS could be developed in the following ways: by taking away practical impediments that prevent patients and professionals from engaging in scientific research (model 1: optimization LHS); by routinely analyzing observational data from electronic health records and other sources (model 2: comprehensive data LHS); by making clinical decisions based on the outcomes of the aforementioned data analyses and directly evaluating the outcomes in order to continuously improve decision-making (model 3: real-time LHS); or by embedding clinical trials into routine care delivery (model 4: full LHS).

CONCLUSIONS

Each model has different ethical implications for consent and oversight. Also, the four-model approach shows that reorganizing a health care center into an LHS is not an all-or-nothing decision. Rather, it is a choice from a menu of possibilities. Instead of discussing the advantages and disadvantages of the LHS menu in its entirety, the medical community should focus on the designs and ethical aspects of each of the separate options.