Transformer-based Multi-target Regression on Electronic Health Records for Primordial Prevention of Cardiovascular Disease

Author(s):  
Raphael Poulain ◽  
Mehak Gupta ◽  
Randi Foraker ◽  
Rahmatollah Beheshti
Circulation ◽  
2018 ◽  
Vol 137 (suppl_1) ◽  
Author(s):  
Brittany M Bogle ◽  
Wayne D Rosamond ◽  
Aaron R Folsom ◽  
Paul Sorlie ◽  
Elsayed Z Soliman ◽  
...  

Background: Accurate community surveillance of cardiovascular disease requires hospital record abstraction, which is typically a manual process. The costly and time-intensive nature of manual abstraction precludes its use on a regional or national scale in the US. Whether an efficient system can accurately reproduce traditional community surveillance methods by processing electronic health records (EHRs) has not been established. Objective: We sought to develop and test an EHR-based system to reproduce abstraction and classification procedures for acute myocardial infarction (MI) as defined by the Atherosclerosis Risk in Communities (ARIC) Study. Methods: Records from hospitalizations in 2014 within ARIC community surveillance areas were sampled using a broad set of ICD discharge codes likely to harbor MI. These records were manually abstracted by ARIC study personnel and used to classify MI according to ARIC protocols. We requested EHRs in a unified data structure for the same hospitalizations at 6 hospitals and built programs to convert free text and structured data into the ARIC criteria elements necessary for MI classification. Per ARIC protocol, MI was classified based on cardiac biomarkers, cardiac pain, and Minnesota-coded electrocardiogram abnormalities. We compared MI classified from manually abstracted data to (1) EHR-based classification and (2) final ICD-9 coded discharge diagnoses (410-414). Results: These preliminary results are based on hospitalizations from 1 hospital. Of 684 hospitalizations, 355 qualified for full manual abstraction; 83 (23%) of these were classified as definite MI and 78 (22%) as probable MI. Our EHR-based abstraction is sensitive (>75%) and highly specific (>83%) in classifying ARIC-defined definite MI and definite or probable MI (Table). Conclusions: Our results support the potential of a process to extract comprehensive sets of data elements from EHR from different hospitals, with completeness and accuracy sufficient for a standardized definition of hospitalized MI.


2018 ◽  
Vol 7 (3) ◽  
pp. e000071
Author(s):  
Smita Bakhai ◽  
Aishwarya Bhardwaj ◽  
Parteet Sandhu ◽  
Jessica L. Reynolds

The 2013 American College of Cardiology/American Heart Association (ACC/AHA) guidelines focus on atherosclerotic cardiovascular disease (ASCVD) risk reduction, using a Pooled Cohort Equation to calculate a patient’s 10-year risk score, which is used to guide initiation of statin therapy. We identified a gap of evidence-based treatment for hyperlipidaemia in the Internal Medicine Clinic. Therefore, the aim of this study was to increase calculation of ASCVD risk scores in patients between the ages of 40 and 75 years from a baseline rate of less than 1% to 10%, within 12 months, for primary prevention of ASCVD. Root cause analysis was performed to identify materials/methods, provider and patient-related barriers. Plan-Do-Study-Act cycles included: (1) creation of customised workflow in electronic health records for documentation of calculated ASCVD risk score; (2) physician education regarding guidelines and electronic health record workflow; (3) refresher training for residents and a chart alert and (4) patient education and physician reminders. The outcome measures were ASCVD risk score completion rate and percentage of new prescriptions for statin therapy. Process measures included lipid profile order and completion rates. Increase in patient wait time, and blood test and medications costs were the balanced measures. We used weekly statistical process control charts for data analysis. The average ASCVD risk completion rate was 14.2%. The mean ASCVD risk completion rate was 4.0%. In eligible patients, the average lipid profile completion rate was 18%. ASCVD risk score completion rate was 33% 1-year postproject period. A team-based approach led to a sustainable increase in ASCVD risk score completion rate. Lack of automation in ASCVD risk score calculation and physician prompts in electronic health records were identified as major barriers. Furthermore, the team identified multiple barriers to lipid blood tests and treatment of increased ASCVD risk based on ACC/AHA guidelines.


2021 ◽  
Author(s):  
Kavita Singh ◽  
Mark D. Huffman ◽  
Nikhil Tandon ◽  
Raji Devarajan ◽  
Dorairaj Prabhakaran ◽  
...  

Abstract Background. Cardiovascular disease (CVD) is pervasive in India, and little is known about the perception of patients and providers about collaborative care in secondary prevention of CVD. To fill this gap, we performed a needs assessment and investigated the barriers and facilitators of the collaborative quality improvement (C-QIP) strategy for secondary prevention of CVD in India.Methods. Between September 2019 – February 2020, we conducted semi-structured in-depth interviews with providers, health administrators, patients and caregivers to understand the challenges and facilitators of the C-QIP strategy consisting of electronic health records-decision support system (EHR-DSS), non-physician health worker and text messages for healthy lifestyle. Also, data were analyzed from the lens of consolidated framework for implementation research (CFIR) to guide effective implementation of the C-QIP strategy. We used an iterative approach for qualitative data analysis based on the framework method. Results. We interviewed 38 physicians, 14 non-physician health workers (nurses, community health workers, pharmacists), 4 health administrators, 16 patients and their caregivers. Challenges perceived from providers’ and health administrators’ perspectives to implement quality in CVD care were related to CFIR actors and inner and outer settings: high patient volume, too few specialists, time-constraints, physician burnout, lack of robust communication system or referral linkage, paucity of electronic health records, lack of patient counsellors, polypharmacy and lack of sustainable financing schemes for outpatient services. In addition, low health literacy, high cost of treatment, misinformation bias, and difficulty in maintaining lifestyle changes were key barriers from patients’ and caregivers’ perspectives. Potential benefits of the C-QIP strategy emerged, such as standardized treatment protocol to minimize variation in care, reduced medication errors, improved physician-patient relationships, and enhanced self-care management. However, concerns were raised about feasibility, adoption, and implementation of EHR-DSS across heterogenous healthcare settings, including related to interoperability, patient confidentiality and data security, appropriateness across diverse patient groups, and care delivery costs.Conclusions. Our findings reveal context-specific, patient-, provider- and health system factors that will influence C-QIP strategy implementation in India. Strategies to optimize chronic care of CVD need to be low-cost, culturally acceptable, targeted, and integrated into existing systems and care pathways to be successful.


BMJ Open ◽  
2020 ◽  
Vol 10 (2) ◽  
pp. e034396 ◽  
Author(s):  
Patrick Rockenschaub ◽  
Vincent Nguyen ◽  
Robert W Aldridge ◽  
Dionisio Acosta ◽  
Juan Miguel García-Gómez ◽  
...  

ObjectivesTo demonstrate how data-driven variability methods can be used to identify changes in disease recording in two English electronic health records databases between 2001 and 2015.DesignRepeated cross-sectional analysis that applied data-driven temporal variability methods to assess month-by-month changes in routinely collected medical data. A measure of difference between months was calculated based on joint distributions of age, gender, socioeconomic status and recorded cardiovascular diseases. Distances between months were used to identify temporal trends in data recording.Setting400 English primary care practices from the Clinical Practice Research Datalink (CPRD GOLD) and 451 hospital providers from the Hospital Episode Statistics (HES).Main outcomesThe proportion of patients (CPRD GOLD) and hospital admissions (HES) with a recorded cardiovascular disease (CPRD GOLD: coronary heart disease, heart failure, peripheral arterial disease, stroke; HES: International Classification of Disease codes I20-I69/G45).ResultsBoth databases showed gradual changes in cardiovascular disease recording between 2001 and 2008. The recorded prevalence of included cardiovascular diseases in CPRD GOLD increased by 47%–62%, which partially reversed after 2008. For hospital records in HES, there was a relative decrease in angina pectoris (−34.4%) and unspecified stroke (−42.3%) over the same time period, with a concomitant increase in chronic coronary heart disease (+14.3%). Multiple abrupt changes in the use of myocardial infarction codes in hospital were found in March/April 2010, 2012 and 2014, possibly linked to updates of clinical coding guidelines.ConclusionsIdentified temporal variability could be related to potentially non-medical causes such as updated coding guidelines. These artificial changes may introduce temporal correlation among diagnoses inferred from routine data, violating the assumptions of frequently used statistical methods. Temporal variability measures provide an objective and robust technique to identify, and subsequently account for, those changes in electronic health records studies without any prior knowledge of the data collection process.


Sign in / Sign up

Export Citation Format

Share Document