The Electronic Health Record (EHR) includes a whole range of data in comprehensive and summary form. It's a digital version of a patient's paper chart. EHR Systems combined as a single, large database, become a powerful tool to conduct comparative effectiveness research. The many modules available in the EHR make refining your research question critical to producing reliable and accurate results.
Things to consider when crafting your EHR query:
- Inclusion criteria. What conditions are you looking for?
- Use of ICD10 (International Classification of Diseases) are widely used in the medical record and can be easily queried. A link to a compendium of these codes can be found at right.
- Is gender a consideration?
- does it matter where in the medical record a condition is located? i.e. Problem list, discharge diagnosis or both?
- Is the location of a visit important? Do you include rural practices or limit to only certain clinics/care centers.
- Exclusion criteria. What don't you want included?
- Healthy other than the condition of interest?
- Are certain medication regimes acceptable? i.e. Statins, NSAIDs, etc.
- Cancer at any point in the history?
- Discrete data. Can the data you are interested in be located as a discrete field in the record?
- Calculated fields like pulmonary function tests and ejection fraction are not discrete fields
- Text files like pathology or radiology results are also not discrete fields.
- Some discrete fields like encounter and procedural data are only available since the implementation of EPIC in 2012.
- Are you asking for Protected Health Information?