Secondary Use Of Primary Health Data In A Clinical Trials Digital Revolution

By Sergey Jakimov, managing partner, LongeVC

Clinical trials stand at the cusp of a transformative evolution in the coming years. At the heart of this shift lies the secondary use of primary health data. This piece discusses how this previously collected invaluable data, when synergized with cutting-edge digital solutions, is set to revolutionize clinical research. We'll demystify what secondary usage entails, chart out the prerequisites for data to be effectively integrated into clinical trials, and unveil the myriad ways it can be employed in these trials. Furthermore, we'll shed light on the burgeoning applications of AI and other technological advancements, emphasizing the potential of primary medical data in clinical settings. Through such integration, we boost research efficacy and bolster its real-world relevance.

Understanding The Secondary Use Of Primary Health Data

Primary health data encompasses patient information derived from historical (retrospective) and ongoing (prospective) medical encounters. When this data, primarily collected for firsthand medical purposes, is applied beyond its original intent, we refer to it as the “secondary use” of primary health data.

While primary health data is gathered with specific research or goals (most often, patient treatment) in mind, its secondary use is multifaceted. It provides a unique perspective, enabling researchers to identify patterns, trends, and insights into patient behaviors and disease pathways. Moreover, it can potentially be used to identify the best-fitting candidates for clinical trials, seamlessly perform market access studies, and as a basis for prospective data generation.

This depth of information makes primary health data indispensable for AI tools, presenting opportunities for predictive modeling and innovative patient care strategies.

The secondary use of primary health data has proven essential in market access studies. Pharmaceutical companies utilize this data to navigate market demand, quantify the disease burden and the quality of life of certain groups of patients (e.g., those with chronic conditions), understand disease prevalence, and anticipate potential hurdles. Here lies its significance in clinical trials, where it's crucial to identify and recruit the right patient profiles.

Leveraging Primary Health Data For Secondary Use

To optimize the secondary benefits of primary health data, particular prerequisites must be met:

1. Compliance: Beyond patient consent, it's imperative to grasp regional data management norms and guarantee that data isn't transferred to third parties. Any digital solution should adhere to GDPR and HIPAA guidelines, ensuring data anonymization safeguards patient privacy. A separate data protection ISO certification is also desirable. In fact, if one wants to fully emphasize the privacy-preserving nature of any digital secondary health utilization solution, the rule of thumb is to look into the models that offer compliant data usage on-premises and limit (or eliminate) the possibility of clinical data items being transferred outside of healthcare institutions. As a result, centralized data clouds, especially those introduced by private companies and not as part of governmental incentives, are slowly becoming redundant.
2. Data Relevancy: Continuously updating data is essential. While certain insights can be generated from retrospective items only, most of its value is in its real-time relevance, ensuring insights remain up to date and actionable. For example, if data is used to identify candidates for clinical trials, one must ensure that mentioned patients are still accessible by or at the clinical site.
3. Interoperability: Different hospitals have varied EMRs. Standardizing this data ensures a cohesive, searchable structure, leaving no critical information behind. Interoperability and overall data uniformity together, unfortunately, remain one of the biggest unsolved issues in the field. Given the hundreds of different EMRs in use and healthcare institutions varying dramatically in their level of digitalization, one might wonder if implementing a single standard is even possible. It, thus, comes down to data normalization and cleanup engines of secondary health data usage enabling solutions that bridge the gap between conflicting formats and uneven digitalization levels.
4. Technological Infrastructure: Storing the data is just the beginning. It is pivotal to present it uniformly, efficiently tracking potential patients and managing their consent.

Primary Health Data's Symbiosis With AI

The union of primary health data with emergent technological advancements offers unprecedented promise in clinical research. Techniques such as federated learning, where AI training occurs on-site within the clinical facilities instead of being sent elsewhere, facilitate this data's real-time and accurate analysis.

Initiating The Clinical Trial With Digital Tools

Identifying potential candidates is merely the initial phase. It's imperative to ensure seamless patient engagement and their subsequent inclusion in the study. Here, the advent of digital tools plays an indispensable role. They allow for the digital acquisition of informed consent, which is easier to manage for the patients, too, including if they want to revoke their consent. Additionally, by digitally harnessing real-world data, candidates can be prescreened based on the trial's inclusion-exclusion criteria. This digital shift accelerates the selection of eligible participants and presents a cost-efficient alternative to traditional methods. To contextualize, standard protocols require candidates to sign paper consent forms in person and address basic exclusion questions on-site. Drawing from data presented in recent studies, including one from PubMed Central, which estimates the considerable financial burdens of pivotal clinical trials, the traditional method might cost up to $500 per patient per visit.¹ In contrast, a digital-focused approach expedites the process and potentially saves substantial resources.

Other Promising Avenues Enabled By Secondary Health Data

Venturing beyond clinical trials, the secondary use of primary health data opens doors to several transformative applications. Federated learning offers simultaneous on-premises AI training across multiple institutions. Furthermore, initiatives like digital prescreening for diseases underscore the potential of compliant data processing in preventive care. A shift in mindset is also evident among study sponsors who, instead of conventional clinical trials, lean toward market access studies. Such endeavors aim to unravel the complexities of disease burdens on specific patient groups or gather quality of life (QoL) metrics, which can be pivotal for advocating treatment reimbursements in particular markets. Lastly, the essence of pharmacovigilance studies, particularly during Phase 4, emphasizes the relentless tracking of potential side effects, ensuring treatments remain effective and safe in real-world applications.

Utilizing primary health data for secondary purposes in clinical trials opens up a plethora of advantages. It paves the way for more swift and precise clinical trials when harnessed correctly.

References:

1. Moore TJ, Heyward J, Anderson G, Alexander GC. Variation in the estimated costs of pivotal clinical benefit trials supporting the U.S. approval of new therapeutic agents, 2015-2017: a cross-sectional study. BMJ Open. 2020 Jun 11;10(6):e038863. doi: 10.1136/bmjopen-2020-038863. PMID: 32532786; PMCID: PMC7295430.

About The Author:

Sergey Jakimov is a founding partner of LongeVC, a venture capital company supporting early-stage biotech and longevity-focused founders and startups. Sergey also founded a company that offers a toolkit for data discovery and patient engagement, along with other deep-tech ventures.