The idea of mining electronic health record (EHR) data to drive clinical trial recruitment has gained momentum in recent years. The significant uptick in EHR adoption rates –estimated to approach 90 percent of all physicians in 2019, thanks to the HITECH Act – has only fueled this trend. Here is a brief look at the current use of EHRs for clinical trial recruitment, including examples of hospitals that are doing it right, and concerns about potential threats to patient privacy.
How Does It Work?
Adequate clinical trial enrollment is essential for a study’s success. Without it, research can be stunted by costly delays, or worse, cancelled altogether. EHRs contain structured information about patients, which allows for quicker and more efficient retrieval of patient information than traditional methods of record keeping. Researchers use EHRs to sort through specific criteria designated by the investigator and approved by the institutional review board (IRB) to quickly identify patients that qualify for clinical trial participation. Data searches can be used to select terms relevant to the condition being studied, such as disease, sex, age, and medication. Once a qualified candidate for a clinical trial is identified, the individual’s physician receives an alert via the EHR. The healthcare providerthen decides whether or not to suggest participation in the clinical trial. Pop-up alerts provide physicians with relevant links to learn more about the trial, which providerscan share with patients. If the patient is interested, the physician can use the EHR to email a professional trained to discuss the research. On the other hand, if the patient decides he or she does not want to participate, the physician can note in the EHR so other healthcare professionals will not ask the patient to join other trials.
Who Is Doing It Well?
The Mayo Clinic has been a pioneer in leveraging EHR data to discover potential clinical trial candidates. Using natural language processing (NLP) and an algorithm engineered to search through EHRs, including free-text clinician notes, the Mayo Clinic successfully located patients with heart disease, cardiomyopathy, and congestive heart failure for a 2007 study.
Another early adopter is Holston Medical Group (HMG). HMG is a 102-provider multispecialty physician group in Kingsport, Tennessee. Since 1996, HMG has used an EHR to enroll thousands of patients in nearly 275 clinical trials, earning a reputation for exceptional speed and efficiency. The organization was named as the nation's top enrollment site for the 10,000-patient ACCORD diabetes study in 2005 by the National Institutes of Health (NIH).
Similarly, Penn Medicine has made great strides in mining its EHR data to help find subjects for clinical trials.Using a new clinical trial application called Penn Research Trial Advisory, Penn Medicine was able to recruit 200 patients from the ob-gyn department for infertility clinical trials in 2010. This represents an 87% increase in the number of physician-referred patients for clinical trials compared to the 4-month period before the new EHR recruitment procedure.
Ownership, consent, and access to health data are key issues in using EHR data for clinical trial recruitment. The HIPAA Privacy Rule requires protected health information (PHI) to be either de-identified or authorized by the patient for release.Patients may perceive the identification of study candidates through EHRs as a breach in physician-patient confidentiality.Perceived infringement on this confidentiality can cause patients to not fully disclose important facts, or worse, avoid medical care entirely. As a result, the patient’s health and treatment may be compromised.
Accordingly, it is often argued that data mining and secondary use of health data requires informed consent. Patient advocates, such as Dr. Deborah C. Peel, founder of Patient Privacy Rights, often believe patients have a right to know who is viewing their medical information, why it’s being accessed, and how it is being used. Moreover, they believe that patients should have a say in the decisions about the access, content, and ownership of EHR data. Since autonomy is threatened when an individual’s personal health information is shared without that person’s knowledge, it would appear to be a violation of patient autonomy for this data to be used beyond this purpose and aggregated to generate new information about patients without their knowledge or permission.
Despite these ethical concerns, it is possible to use EHR data for clinical trial recruitment while maintaining respect for patient privacy. A study’s clinical trial recruitment plan must be approved by an IRB before recruitment takes place. Researchers file an Application for IRB Waiver of HIPAA Privacy to be authorized to use PHI for prescreening if they intend to use EHRs to identify potential research participants. IRBs reserve the right to reject such requests if they determine the practice would violate the federal health information privacy regulations under HIPAA.
Furthermore, health care professionals have played a vital role in clinical recruitment long before EHRs were established. Non-electronic clinical trial recruitment strategies are commonplace in physicians’ offices, as research demonstrates individuals are more likely to participate in a study if their doctor suggests it. The longstanding history of clinicians culling research subjects from their practices begs the question: Why does it seem more ethically unsettling when this process is done electronically, rather than manually or personally?
While critics may look at EHR data as exceptional because of its digital nature, its use may not be so different than traditional methods after all. Moreover, the acceleration of the recruitment process can help researchers advance new treatments to the market quicker and at a lower cost, which can contribute to better health outcomes at the societal level.Collaboration among ethicists, researchers, clinicians, informatics specialists, and policy makers is necessary to establish guidelines that ensure the safety and privacy of patients while optimizing EHR data to promote clinical research.
Do you use EHRs to support clinical trial recruitment in your profession? Are you a patient who joined a clinical trial after you were identified through an EHR? Tell us about your experience in the comments below!
Donna Hanrahan is HITLAB's Community Manager and Research Associate.