EHR vs. eSource: Why so many research sites don’t use EHR as the source data collection tool
EHR’s aren’t built to support clinical research the same way as eSource
This may be surprising to many, but it’s important to understand that an EHR is not designed to support the needs of a clinical trial. The EHR’s primary purpose is to give healthcare providers a flexible tool to document patient encounters and generate appropriate billing for insurance companies. EHR systems utilize standardized templates optimized for delivery of care, not research.
By contrast, clinical trials have very specific and lengthy data requirements that vary with each protocol. To comply with study protocols, research staff have to schedule visits in a prescribed order, with limited variance against target dates; collect detailed and structured data and follow rigidly prescribed workflows that dictate everything from how the vitals are to be taken (sitting vs. supine, right arm vs. left arm, etc.) to how the IP is administered.
It is extremely difficult for most research personnel to configure the EHR visit schedules and data templates to meet the complex needs of a particular study. No health care institution could justify expending the IT resources required to reconfigure the EHR to accommodate 5-10 study participants when they are managing hundreds of thousands of patients on the care side.
Care data and research data are different
Furthermore, the types of data used in the conduct of clinical care versus that of clinical research are of limited value to each other. On the care side, multiple providers need to access it, and each provider makes incremental contributions to the patient’s medical history. Over time, the data becomes large, messy, and outdated. Medical conditions, for example, often contain conditions that have self-resolved; hypotheses that did not bear out; or different diagnoses for the same underlying condition.
Research requires precision. Study sponsors rely on the Principal Investigator to document the actual medical conditions of the patients, not past conditions or provisional hypotheses, and the actual medications the patient is taking, not just those that were prescribed but not taken. In addition, research protocols require documentation of study-specific data, such as pill counts or study-specific compliance, that are of value to the study sponsor but not to other care providers.
The EHR is used for standard of care procedures but not research procedures
EHR is entrenched in health care, so when a protocol’s procedure overlaps with standard of care, many sites will utilize the EHR to perform that procedure. In oncology studies, standard of care procedures can be the majority of procedures, meaning that these studies do rely on the EHR for much of their source data collection – although the research-specific procedures are often done outside of the EHR, on paper charts. Outside of oncology and surgery, however, most interventional studies have limited overlap with standard of care. In fact, many studies utilize dedicated research-only sites, or sites aligned with health care practices but which actively recruit volunteers outside the patients they serve. For this reason, the overwhelming majority of source data is collected outside an EHR.
Today’s leading sites are transitioning from paper to eSource
Historically, research sites have built detailed source templates in Word. These templates contain detailed instructions on how to perform each procedure, and a series of questions and answers the site team must complete. As these templates are completed, the coordinator will store the completed paper copies in patient-specific binders, to be reviewed by the CRA during on-site visits. If medical records are referenced (e.g., in medical history or concomitant medications), they are printed and annotated, then inserted into the binder. Experienced CRA’s are well familiar with images like these from a research site:
This is where eSource comes in. eSource gives sites a tool they can use to configure study-specific visit schedules and procedures. It eliminates the use of paper, builds in quality checks at the point of data capture, and enables true remote monitoring – where all site source data can be reviewed by the monitoring team, with no requirement to book travel or visit the site’s facility.
Over 2000 sites in 27 countries use CRIO eSource for their clinical trial data capture needs. Many, if not most, still use the EHR – but mainly as a source of potential patients, and as an input to the clinical trial data (medical history, medications and adverse events). They will notate in the EHR the fact of the patient’s participation in the trial; the nature of the trial; and critical milestones, such as randomization or IP administration. But none of CRIO’s clients use the EHR as their primary source data collection tool – in fact, several of CRIO’s clients tried and failed with EHR before adopting CRIO eSource.
Central eSource advances clinical research
Now, with Central eSource, sponsors can benefit from CRIO eSource across their entire study. Sponsors are able to specify a standard electronic source template, pre-mapped to the eCRF, and then send the central eSource template to the sites on the study. With CRIO’s support, sites can configure the eSource template locally to reflect their SOPs, then leverage CRIO’s built-in EDC integration to populate the EDC at the click of a button.
With better quality touch-points embedded throughout their processes, CRIO’s eSource-enabled sites are able to reduce protocol deviations by up to 50% and improve data quality by 70%. This enhances PI oversight, promotes patient safety, and frees up the sites to focus on patient recruitment and retention and not clerical work.
