.png){kind=small}
In the high-stakes realm of clinical research, the accuracy and integrity of patient data are paramount. Traditionally, quality control (QC) in clinical trials has been a painstaking process, involving manual checks and balances designed to ensure data integrity and adherence to protocols. The manual nature of reviewing this data is not only a time-consuming, burdensome task for site staff, but also introduces risks of oversight failures due to the sheer volume of data and the fatigue factors inherent in repetitive, detailed-oriented tasks.
The Shortcomings of Traditional QC Methods
Time-Consuming Processes: Traditional QC methods require extensive human labor to review and verify each data point collected during the trial. This process is slow and often becomes a bottleneck in the study timeline.
Susceptibility to Human Error: Manual data handling increases the risk of errors due to misinterpretation, oversight, or simple data entry mistakes. These errors can compromise data integrity and affect trial outcomes.
Cost Implications: Every hour spent on manual quality control is a cost added to the trial. More importantly, errors or delays can lead to significant additional costs related to trial extensions, regulatory fines, or failed audits.
Scalability Issues: As clinical trials grow in size and complexity, scaling traditional QC methods becomes impractical. More data and more complex protocols mean exponentially greater manual effort is required
E-source technology has undoubtedly been beneficial - automatically flagging certain issues at the point of entry. However, e-source solutions often handle basic scenarios well (i.e. time windows or missing information) but lack the sophistication required to catch more complex scenarios. For example, if a certain lab result is an AESI, if a new conmed is a prohibited medication, if a certain measurement is a withdrawal criteria... these require manual analysis and assessment. Plus, many sites hesitate to fully integrate e-source systems due to various challenges.
Queries and Deviations
When issues in data collection do fall between the cracks and are entered into the EDC, they lead to queries and deviations. The presence of queries requires research staff to revisit the original data sources, verify accuracy, and make necessary corrections, which can be time-consuming and delay data analysis and trial progression. The negative implications of queries and deviations include:
Increased Costs and Delays: Resolving queries and addressing deviations often requires additional resources and time, which can extend study timelines and increase costs.
Regulatory Implications: High rates of deviations may lead to scrutiny from regulatory bodies, potentially resulting in audits or fines if the deviations affect trial integrity.
Impact on Data Quality: Significant deviations may compromise the reliability of trial results, affecting the overall conclusions that can be drawn.
Reputational Damage: Frequent queries and deviations can damage the reputation of the clinical trial sites, affecting their ability to participate in future studies.
Enter Sentinel:
A second set of eyes automatically reviewing your patient visit documentation.
Catch Problems Before They're Problems...Quri's groundbreaking Automated Quality Control solution, Sentinel, is tailored specifically for clinical research sites to catch problems before they turn into queries and deviations. Sentinel operates as a rule-based system enhanced by Artificial Intelligence, specifically leveraging Large Language Models (LLMs). This enables PhD-level complex reasoning when assessing incoming data, understanding the context and background knowledge required to connect the dots.
Rule Set-Up
Sentinel begins by ingesting the clinical trial protocol and any associated manuals. This crucial first step allows the system to understand the specific requirements and standards set forth for the trial, such as eligibility criteria, prohibited medications, adverse events of special interest, and other specific procedural and visit requirements. Based on this, Sentinel sets up rules for each visit. These rules instruct the the AI on what to look for and what to generate alerts for when reviewing visit documentation.
Assess Pre-Screening Eligibility
Upload the patient's medical records, pre-screen/intake forms, medical history notes, whatever... Sentinel will auto-assess if the patient matches the trial's initial criteria.
Patient Visit Documentation Upload
When a patient comes in for a visit, Upload the Source (Paper Source or Connect to E-Source), Lab Results, Test Results, AE logs, Conmed logs, and more into Sentinel. Drag-and-Drop PDF's or leverage our automation to auto-download your E-Source into auto-send it to Sentinel.
Real-Time Analysis and Intelligent Alerts
If Sentinel detects any discrepancies or potential issues that could lead to deviations or queries, it immediately generates intelligent alerts. For example...
If a Conmed is Prohibited, you will immediately know
If there is an eligibility issue or data that triggers a withdrawal criteria, you will immediately know
If there is an Adverse Event, you will immediately know
If drug compliance is out of range, you will immediately know
If a test measurement triggers an Adverse Event of Special Interest, you will immediately know
If a lab measurement has increased by your threshold of x% over the past 3 visits, you will immediately know
More...
Exclusive Beta Partner Program
Sentinel's Beta Partner Program (limited slots) is closing shortly. Participants in this program not only receive early-access and a lifetime discount, but get to influence the future development of this cutting-edge tool.
Interested in seeing Sentinel in action and learning how it can streamline your QC process? Reach out today to schedule a demo and secure your spot in our Beta Partner Program.
Comments