Electronic health record (EHR) data is collected during the ordinary course of a patient’s interactions with the healthcare system. It encompasses a wide range of information collected during patient care and medical research, including:
- Patient demographics
- Diagnoses
- Medications
- Procedures
- Treatment history
- Laboratory tests
- Diagnostic images
EHR data typically go beyond standard clinical data collected by a healthcare provider and include a broader view of patient care. EHRs are built to share information with other healthcare providers, such as laboratories and specialists, so they contain information from all the clinicians involved.
Trends in EHR Data
With technological advancements and guidance from the FDA, common themes surrounding EHR data are emerging.
- Acceptance for Evidence Generation: There's a growing acceptance of using EHR-derived data to support evidence generation for regulatory submissions. This acceptance is facilitated by FDA guidance, which outlines the assessment of EHR and medical claims data for regulatory decision-making regarding drugs and biological products.
- Expanded Collaboration: EHR software companies, health systems, and specialty provider networks are increasingly collaborating to support research initiatives. This expands the range of data partners beyond traditional EHR data aggregators.
- Diverse Data Usage: EHR data usage is being driven by various factors, including the rise of biomarker-specific therapies and the increasing use of biometric changes, Patient Reported Outcomes (PROs), and Clinician Reported Outcomes (ClinROs) to understand disease severity and treatment effectiveness across multiple medical specialties such as oncology, autoimmune diseases, dermatology, ophthalmology, orthopedics, respiratory, and cardio.
- Emerging Use Cases: The emerging use cases for EHR data focus on improving clinical trial efficiency through faster patient identification, synthetic control arms, pharmacovigilance, safety monitoring, and the application of Artificial Intelligence (AI) to create predictive models for disease outcomes.
- Advancements in NLP and ML: Advancements in Natural Language Processing (NLP) and Machine Learning (ML) enable the mining and unlocking of unstructured data within EHRs. This unstructured data, which constitutes approximately 80% of EHR data, helps contextualize the reasons behind a patient's diagnosis and treatment. However, scaling this process remains challenging.
- Specialty EHR Data Providers: There is a rise in specialty EHR data providers focusing on specific disease areas. These partners offer more specialized data elements relevant to those particular diseases, enhancing the depth and specificity of available data for research and analysis.
Using EHR Data
Combining EHR data with real-world data provides a wealth of information. It offers researchers an opportunity to uncover new findings and improve patient care. Key applications of clinical data include:
- Monitoring long-term efficacy and safety of clinical interventions
- Comparative effectiveness research
- Supporting regulatory approval through real-world evidence (RWE)
- Recruiting patients for clinical trials and observational registries
Emerging uses for EHR data center around improving clinical trial efficiency through faster patient identification or synthetic control arms, pharmacovigilance and safety monitoring, and applying AI to create models to predict disease or outcomes. EHR software companies, health systems, and specialty provider networks are becoming active collaborators in supporting research initiatives, expanding choice of data partner beyond traditional EHR data aggregators.
Monitoring long-term efficacy and safety of clinical interventions
Researchers use clinical data to evaluate the effectiveness and safety of new treatments, medications, or medical devices. By analyzing patient outcomes, adverse events, and other relevant variables, researchers can assess the impact of interventions and make evidence-based recommendations.
Case Study: Extending the effectiveness and safety analysis of a new focal seizure treatment
Situation: In a Phase IV study, a biopharmaceutical company assessed a newly approved anti-seizure medication as adjunctive therapy for focal seizures. The results showed reduced seizure frequency and high patient retention, indicating potential benefits for managing focal seizures.
Need: The client aimed to gain insights, focusing on several key aspects:
- Understanding the patient’s seizure frequency prior to the trial and the overall cost of their care.
- Assessing the impact of the therapy on healthcare utilization both during and after the trial.
- Identifying factors that contribute to patient discontinuation and adverse events.
- Evaluating the long-term effectiveness and safety of the therapy after 6 months and 1 year of exposure.
- Minimizing the expense and challenge of asking already overburdened clinical sites for prospective data collection
Solution: The biopharmaceutical company de-identified and linked EHR data with claims data using Datavant’s data connectivity technology. This empowered the company to:
- Establish a baseline of healthcare utilization and assess the impact of the new therapy on overall cost of care.
- Evaluate long-term treatment efficacy and safety of the treatment.
Comparative effectiveness research
Clinical data facilitates comparative effectiveness research, where clinical outcomes of different interventions or treatment strategies are compared to determine which provides the best results for specific patient populations. This research aids in making informed decisions about treatment options and resource allocation.
Case Study: Understanding clinical and economic outcomes in rheumatoid arthritis patients
Situation: A biopharma company wanted to evaluate clinical and economic outcomes associated with lower disease activity states for patients with rheumatoid arthritis.
Need: The company needed to understand whether disease activity in rheumatoid arthritis was associated with adverse events, specifically hospitalization, emergency department visits, mortality, and medical costs.
Solution: By connecting clinical data from a disease registry with claims data, the biopharma company was able to conduct clinical and economic assessments. This resulted in:
- Better understanding of how hospitalization rates correlate with disease activity
- Improved value proposition for payers
Support regulatory approval through RWE
Real world evidence studies utilize clinical data to assess how interventions perform in real-world settings, beyond the controlled environment of clinical trials. This helps bridge the gap between efficacy and effectiveness and provides insights into how treatments work in diverse patient populations.
Case Study: Creating an Oncology External Control Arm
Situation: A top 5 pharma company aimed to conduct a clinical study for an experimental treatment in a solid tumor indication. They needed to establish an external control arm to support the clinical study.
Need: The pharma company needed access to EHR data. This would empower them to compare real-world outcomes with those enrolled in the clinical trial.
Solution: The pharma company accessed data from three electronic medical records through the Datavant ecosystem. This allowed them to analyze the patient population, compare outcomes, and evaluate real-world effectiveness. The external control arm enhanced the standard of care, resulting in:
- Accelerated trial recruitment
- Increased patient access to the experimental drug
- Reduced the time to market.
Recruiting patients for clinical trials and observational registries
Understanding patients through clinical data can help accelerate trial recruitment. By analyzing the characteristics and outcomes of participants in previous trials, researchers can identify potential candidate pools and optimize recruitment strategies for patients that can be found in real-world settings. This improves enrollment rates, reduces recruitment time, and ensures a diverse and representative participant population.
Case Study: Overcoming clinical trial recruitment challenges in NASH
Situation: Recruiting patients for Nonalcoholic fatty liver disease (NAFLD) and Non-alcoholic steatohepatitis (NASH) trials is challenging due to the need for both lab test results and specialized liver imaging, which are often collected separately.
Need: A pharma company needed to link patient data from labs and imaging centers to confirm NAFLD and NASH diagnoses and recruit eligible patients for their trial.
Solution: By connecting blood tests from labs to imaging results, the company was able to confirm a NALFD or NASH diagnosis. The company was able to streamline clinical trial recruitment at sites with a higher concentration of NAFLD and NASH cases.
Privacy and Data Protection Around EHR Data
Privacy and data protection are paramount when working with clinical data to ensure confidentiality and comply with ethical guidelines. Here are some measures that should be implemented:
- De-identification and anonymization: To protect patient privacy, EHR data should undergo de-identification and anonymization processes, removing personally identifiable information while preserving the data’s integrity for research purposes.
- Access controls and restricted data sharing: Access to clinical data should be limited to authorized personnel and organizations. Implementing secure systems with access controls ensures that only approved individuals can access and analyze the data. Data sharing should be conducted through secure channels and agreements that prioritize privacy and compliance.
Health Data Ecosystems Containing EHR Data
Organizations can access a wealth of EHR data through the Datavant ecosystem. Datavant connects disparate data sources, empowering organizations to gain a comprehensive view of patient populations, uncover new insights, and drive better outcomes.
Datavant ecosystem provides access and connectivity to various data sources, including:
- EHR data
- Claims data
- Genomic data
- Mortality data
- SDOH data
- Specialty pharmacy data
Clinical data holds immense potential in transforming healthcare and addressing health disparities. With Datavant, healthcare organizations can securely access EHR data, empowering them to make data-driven decisions and improve patient outcomes.
Additional access to EHR data
Researchers and organizations can also access electronic health records through the following avenues:
- Data-sharing initiatives and repositories: Several initiatives and repositories promote data sharing, allowing researchers to access and analyze clinical data. Examples include the Global Alliance for Genomics and Health, the COVID-19 Research Database, and disease-specific data repositories.
- Collaboration with trial sponsors and investigators: Establishing collaborations with trial sponsors and investigators is another effective way to access clinical data.
Clinical data offers a trove of insights and possibilities for advancing medical knowledge and improving patient outcomes, driving medical innovation and enhancing evidence-based practices.
.svg)
