Chart Review and EMR Extractions in RWE Research

Two primary methods used to gather data for Real World Evidence (RWE) research are chart reviews and Electronic Medical Record (EMR) extractions.


Chart Review involves manually extracting data from patient medical records, including both paper-based charts and digital documents. This method provides a rich source of detailed clinical information.


EMR Extractions involve the automated retrieval of structured data from electronic health records. This method leverages existing digital infrastructure to gather large amounts of data efficiently.


Chart review is a valuable method when detailed, context-rich clinical data are required. It is particularly useful for studying rare conditions, complex cases, longitudinal histories, and qualitative aspects of patient care. While it is labor-intensive and has limited scalability, the depth and accuracy of information it provides can be key for certain types of research. In contrast, EMR extractions offer efficiency, scalability, and comprehensive data coverage. However, they require careful attention to data quality, interoperability, and privacy.


To determine the most suitable method to use for a study, various factors need to be considered. This article explores applicability, advantages and disadvantages of the two methods.


When to Use Chart Review

  1. Detailed Clinical Information: When the research requires in-depth clinical details that are not captured in structured fields of EMR. For example, studies needing precise clinical notes, detailed descriptions of patient symptoms, and physician's qualitative assessments.
  2. Rare Conditions or Complex Cases: When studying rare diseases or complex medical conditions where detailed case histories and individualized patient information are key, in research on rare genetic disorders where comprehensive clinical documentation is necessary for understanding disease progression and treatment outcomes.
  3. Longitudinal Patient Histories: When investigating long-term outcomes or longitudinal patient histories that involve multiple episodes of care and nuanced clinical narratives, for studies tracking the progression of chronic diseases over several years, requiring detailed documentation of changes in symptoms and treatments.
  4. Contextual and Qualitative Data: When qualitative data and contextual information are essential to the research questions and when exploring patient adherence to treatment regimens, where understanding the reasons behind non-adherence from clinical notes is important.
  5. Pilot or Exploratory Studies: For preliminary or pilot studies where detailed insights are needed before scaling up to larger studies, in initial exploration of new therapeutic interventions where comprehensive patient feedback and detailed clinical observations are necessary.
  6. Complex Outcomes or Adverse Events: When studying complex outcomes or adverse events that require thorough investigation of patient records to understand the context and contributing factors. For example, when researching on adverse drug reactions where detailed clinical documentation and physician assessments are critical to identify causality and patterns.



Chart Review advantages:

1. Rich, detailed Contextual Data: Chart reviews capture nuanced clinical details and contextual information that may not be structured or recorded in EMRs. Provides access to unstructured data that offer deeper insights into patient care and clinical decision-making.

2. Flexibility: Researchers can tailor data collection to specific research questions and study objectives.

3. Validation and Quality Control: Direct review of original records allows for high accuracy and the potential for correcting discrepancies.


Chart Review disadvantages:

1. Labor-Intensive: The process is time-consuming and resource-intensive, requiring significant human effort for data extraction and interpretation.

2. Subjectivity: Different reviewers may interpret data differently, leading to potential bias and variability in data quality.

3. Scalability: Due to its manual nature, chart review is less scalable, making it impractical for large-scale studies.

4. Data privacy and consent: It might be needed to seek consent from the patients to have access to their charts.



  • Personnel Costs: High due to the need for trained clinical reviewers and data abstractors.
  • Time Costs: Extensive time required for manual data extraction and validation.
  • Training Costs: Costs associated with training personnel to ensure consistent and accurate data extraction.
  • Operational Costs: Costs for accessing and handling medical records, especially if paper records need to be digitized.



When to Use EMR Extractions

  1. Large-Scale Population Studies: When studying health outcomes across extensive patient populations to identify trends and patterns.
  2. Longitudinal Studies: For tracking patient outcomes over time, leveraging the continuous and comprehensive nature of EMR data.
  3. Comparative Effectiveness Research: To compare the real-world effectiveness of different treatments or interventions using broad and diverse datasets.
  4. Safety Surveillance: For monitoring adverse events and safety profiles of medications or medical devices in real-world settings.
  5. Health Services Research: To analyze healthcare utilization, costs, and quality of care across different healthcare systems.
  6. Precision Medicine: Utilizing detailed patient data to tailor treatments based on individual characteristics and genetic information.


EMR Extractions advantages:

1. Efficiency and Scalability: Automated processes enable the rapid extraction of large datasets, making it suitable for large-scale studies involving extensive patient populations across multiple sites.

2. Consistency and Standardization: EMR data is typically structured and standardized, facilitating easier analysis and reducing variability compared to manual data collection.

3. Timeliness: Real-time or near-real-time data availability enhances the relevance and immediacy of research findings.

4. Broad Scope: Comprehensive datasets encompassing a wide range of clinical parameters including diverse patient populations, improving the generalizability of findings.


EMR Extractions disadvantages:

1. Data Quality: EMR data can be incomplete or inaccurate due to inconsistent documentation practices. Missing data and entry errors can impact study outcomes. The use of statistical methods to handle missing data, such as imputation techniques might be required.

2. Data Heterogeneity: Variability in data entry practices and EMR system configurations can lead to inconsistencies.Data harmonization techniques and standard operating procedures might need to be implemented.

2. Limited Detail: Structured data may lack the nuance and context found in chart reviews. Important clinical notes and contextual information may be overlooked.

3. Technical Challenges: Extracting data from different EMR systems requires advanced technical expertise and can be complicated by interoperability issues.

4. Bias and Confounding: Potential for systematic biases related to data recording practices and patient selection. Statistical methods to adjust for confounding factors and biases might be needed.



  • Initial Setup Costs: High initial investment in IT infrastructure and software for data extraction.
  • Maintenance Costs: Ongoing costs for maintaining and updating EMR systems and extraction tools.
  • Personnel Costs: Lower than chart reviews but requires skilled IT professionals and data scientists.
  • Interoperability Costs: Costs associated with ensuring data compatibility and integration across different EMR systems.
  • Data Licensing Costs: Potential costs for accessing EMR data, especially from multiple healthcare providers or systems.



Comparing Chart Review and EMR Extractions


Nature of Data:

  • Chart Review: Provides detailed, unstructured data from individual patient records.
  • EMR Extractions: Offers standardized, structured data from electronic records.


Data Granularity:

  • Chart Review: High granularity with rich, context-specific information.
  • EMR Extractions: Less granular but more consistent and easier to analyze.



  • Chart Review: Limited by the manual, labor-intensive nature of the process.
  • EMR Extractions: Highly scalable, suitable for large cohorts across multiple sites.


Potential for Bias:

  • Chart Review: Higher risk of subjective interpretation and variability among reviewers.
  • EMR Extractions: Lower risk of subjectivity but susceptible to systematic biases inherent in data recording practices.


Resource Requirements:

  • Chart Review: Requires significant human resources and time.
  • EMR Extractions: Requires technical infrastructure and expertise but less manual effort.


Cost Implications:

  • Chart Review: Higher personnel and time costs, potentially higher operational costs.
  • EMR Extractions: Higher initial setup and maintenance costs, potentially lower ongoing personnel costs.



Applications in RWE Research

Chart Review:

  • Ideal for studies requiring detailed clinical insights and contextual information.
  • Generally, chart reviews involve smaller sample sizes, often ranging from dozens to a few hundred patients, depending on the study’s scope and available resources.
  • Useful for validating findings from EMR data or other sources.

EMR Extractions:

  • Suitable for large-scale population studies (often ranging from thousands to hundreds of thousands of patients, depending on the study design and data availability), longitudinal research, and comparative effectiveness studies.
  • EMR extractions typically involve large sample sizes, often.
  • Efficient for safety surveillance, health services research, and precision medicine applications.



Both chart review and EMR extractions play vital roles in RWE research. The choice between these methods depends on the specific research objectives, available resources, cost considerations, and the nature of the data needed. By understanding the strengths, limitations, and costs of each approach, researchers can effectively leverage both methods to enhance the quality and impact of RWE research.