Technology

Protocol Feasibility in the EMR age

Technology

13:00, August 30 2017

image

Sameer Tandon, Head, Strategic Alliances & Customer Transaction, Novartis, explores protocol feasibility in the era of electronic medical records

Classic protocol feasibility is conducted in several steps that can often take weeks or months before we can truly understand evidence of our capability to recruit patients for a clinical trial. These steps often include feasibility questionnaires, phone screens with potential principal investigators (PIs), pre-study site visits (PSSVs), and patient chart reviews. The initial identification of patients to support enrollment can often be depicted in a “funnel” as shown below:

Figure 1

This methodology has evolved somewhat over the years with expedited approaches for exchange of information, such as electronic confidential disclosure agreements (CDAs) and web-based feasibility questionnaires. However, the feedback on potential patients available is often an educated guess. The industry is ready to start leveraging a more precise measure of protocol predictability by utilizing electronic medical records (EMRs).

What is an EMR and how can it help facilitate Clinical Research?

The Electronic Health Record (EHR) or EMR is a longitudinal electronic record of patient health information generated by one or more encounters in any care delivery setting. Included in this information are patient demographics, progress notes, problems, medications, vital signs, past medical history, immunizations, laboratory data, and radiology reports.

The EMR automates and streamlines the clinician's workflow. The EHR has the power to generate a complete record of a clinical patient encounter, as well as supporting other care-related activities directly or indirectly via interface—including evidence-based decision support, quality management, and outcomes reporting.”

Health Information Management Systems Society

Today we can rely on electronic health records (EHRs) to help us facilitate clinical research and improve our methodology to predict patient enrollment. Over the past few years, several service providers have enabled end-user applications for sponsors that can generate queries using de-identified data for a deeper understanding of patient characteristics:

Figure 2

How does it work?

Ideally, the end-user application is designed to facilitate clinical research and collaboration between pharmaceutical companies and healthcare providers. The core service helps identify potential overly-restrictive inclusion and exclusion criteria (IC/EC) being considered for a new protocol at the synopsis development phase by leveraging electronic health record data. One major objective is to encourage interaction at this level between sponsor and institutions which can often lead to a better understanding of our targeted patients or possible collaboration in the trial itself.

This ability to query millions of patients at multiple hospitals allows for a powerful methodology to obtain real-world insights on patient characteristics. The first step is to take your protocol and review your inclusion/exclusion criteria and begin generating a query. By entering what we consider our “must haves” versus our “cannot haves” we start building a robust analysis on how our protocol is matched with live patients in the hospital network.

One of the benefits of course is that we can proactively make adjustments in the query to visualize the impact of our changes with just a few key strokes. These adjustments could have a significant impact by increasing the number of eligible patients and avoiding costly protocol amendments. This approach should also allow us to think about reducing patient burden. As we develop a query and look at the types of visits, procedures, and labs that will be required, it can provide powerful insight on how we may want to reduce or eliminate some of these tests to encourage patient participation. A few service providers have the capability to predict annual patient turnover based on the matched characteristics, so this is an important step for any query generation.

Traditional vs. EMR Approach

The traditional approach to protocol feasibility is to first start off with engaging PIs and sites. This usually starts with a phone call or email based on a site list where we send a feasibility questionnaire along with other documents, such as the protocol and investigator brochure for review and feedback. The feasibility questionnaire response is a key driver in our proficiency to assess if a site has patients that match our protocol criteria. This data is provided as result of PI/study coordinator input based on chart reviews, current patients in their database/EMR, or referral networks from other participating research centers. However, this is not a very precise measurement in our ability to predict how we can enroll patients for an upcoming trial as depicted in Figure 1. Industry has unfortunately learned that our sites are often a bit optimistic in their predictions until they start screening patients. We do not really know until we start screening if a site will identify patients to match the challenging criteria which are typical of a clinical trial protocol.

Figure 3

The major difference here is quickly “diving” into the IC/EC and testing our protocol hypothesis immediately before investing in expensive study start-up procedures. Consider the time saving and cost reduction if we can quickly predict eligible patients at prospective sites before we initiate all of the robust site activation procedures? The “traditional approach” relies heavily on sites to immediately enroll patients which may or may not exist!

Does this really add value?

In a short answer “yes” this can add significant value to your protocol feasibility steps and should be a supplement to the standard procedures you follow when initiating a trial concept. The potential to “test” your hypothesis with live data in a hospital network will provide your study teams the real-world analysis we need when planning our programs. The proficiency to make adjustments to your protocol before you finalize can significantly reduce the potential for protocol amendments. A number of service providersenable you to reach out to treating physicians to allow for that initial dialogue to better understand the query that has been generated along with the types of patients that may fit the criteria proposed. These early conversations before protocol finalization will add value and help reduce time to enroll and in theory decrease unnecessary start-up costs for sites that may not contribute to the trial. It is essential and perhaps should be considered an obligation for every sponsor to implement an EMR analysis for all protocols to ensure we are designing trials that meet the needs of our patients.

Figure 4

 

Sameer Tandon, MBA

Head, Strategic Alliances & Customer Transactions

Novartis Pharmaceuticals Corporation

Email: Sameer.tandon@novartis.com

Comments

Post a comment

Comments may be moderated for spam, obscenities or defamation.