The topic of subject enrollment has been evaluated from many different perspectives. We recently explored how pharmacies can impact subject enrollment, and in these articles, much of the literature suggests that referrals from trusted healthcare sources, such as pharmacists, healthcare professionals and physicians tend to have the highest impact on influencing a patient to participate in a clinical trial. In this article, I will describe breakthrough technologies that utilize aggregated Electronic Medical Records (EMR), and how this system leverages trusted referral sources for clinical trial subject enrollment.
Aggregated EMR is a process for querying and mining for specific healthcare information from numerous hospital system EMR databases, and then aggregates the data into a single report. The process is capable of obtaining both structured (i.e., fixed fields filled out in an EMR system, such as gender) and unstructured data (e.g., mention of a certain condition in the physician’s clinical notes). To illustrate, let’s say that a sponsor is looking to recruit patients for an Alzheimer’s clinical trial who are 50 years and older, and also took The Mini Mental State Examination (MMSE). Through aggregated EMR, study teams can query for high quality Alzheimer’s patients by ICD-9 code, filtering patients by age range, and then drilling into clinical notes to uncover patients that have ‘MMSE’ mentioned in the notes. Table 1 and Figure 1 demonstrate results from this query, and how aggregated EMR can map qualified patients for clinical trial subject enrollment.
As mentioned previously, the most effective and impactful way to engage patients is through their healthcare providers and physicians. This aggregated EMR system leverages relationships with hospital systems and physicians to request direct referrals to patients. To elaborate, if a hospital system has 20 satellite sites, and an EMR query captures a patient who was seen a week ago, research coordinators can not only immediately identify the patient through their medical record number, but, also know who their treating physician is. Correspondingly, research coordinators can request the treating physician to qualify and refer that patient to a specific clinical trial.
Some concerns regarding the usage of this technology involve the notion that the query may not capture the appropriate patient profile, and that current EMR systems do not contain sufficient data. The aggregated EMR technology allows study teams to optimize query design in order to evaluate patient populations from numerous view points, thus, creating different patient profiles, but, similar targets, which expands the patient capture radius. Moreover, if an EMR system does not contain sufficient data, traditional screening methods would also prove to be ineffective; the advantages of using an aggregated EMR system enables research teams to efficiently screen and enroll patients.
Protecting patient health information (PHI) is critical, not only from regulatory but also ethical standpoints. Innovative technologies that leverage Natural Language Processing (NLP) technologies, and machine learning algorithms can identify PHI and automatically redact the data from a medical record. Additionally, the medical record does not leave the medical center, as the EMR tool accesses EMR data directly from the source. Correspondingly, clinical researchers can look at a patient’s clinical notes without having to worry about PHI/HIPAA violations, and medical centers would not have to be concerned about putting their patients at risk of exposure.
While aggregated EMR is proven to be a highly effective medium for clinical trial subject enrollment (compared to traditional models), it is always a good idea to hedge budgetary risk by diversifying subject enrollment investments. It is paramount to expose clinical trials to patients in as many channels as the subject enrollment budget can bear, however, since we work with limited budgets, it is important to implement budgetary diversification techniques in order to maximize ROI potential on enrollment campaigns.
Aggregated EMR offers many advantages towards clinical trials including analytical protocol design and optimization (which reduces timeline slippage), evaluating patient populations for label expansion, developing predictive models for key risk and performance indicators regarding medical risk based monitoring and adverse event screening/reporting, health outcomes research, and much more. Although such systems offer significant advantages over traditional subject enrollment methods, study teams should carefully evaluate their clinical trial objectives, and choose an enrollment model and medium that best accesses their targeted patient populations.