RD Research Services, LLC | Compliance Expertise for IRBs & Institutions

Tag: technology

  • Sharpenย Yourย Review:ย Practicalย Adviceย forย Reviewers

    Sharpenย Yourย Review:ย Practicalย Adviceย forย Reviewers

    Modified on March 5, 2026 to remove the “subscribe” option. This blog has been retired and replaced by the S.P.I.R.I.T. newsletter.

    Good morning, good afternoon, and good evening, Compliance Rockstars, Clinical Research Professionals, Ethics Enthusiasts, and Investigators from around the globe! 230+ subscribers and counting!

    I hope everyone has been doing well! It feels good to be writing to you all again. I absolutely love knowledge sharing and learning from others. As Iโ€™ve also mentioned in a previous post, writing has always been beneficial towards my mental health. Whatโ€™s even better is knowing that blog posts can help folks in the research and compliance space to think differently. Or perhaps, enhance their current thinking on a particular subject.

    In essence, I love to help people and explain complex topics in an easy format.

    I would like to share top strategies to sharpen how to review your institution’s protocols. As a general reminder, these are my own interpretations. Any legal information discussed within this post should be discussed with your institution.

    Sharpen your pencils folks and let’s begin:

    Research the research

    Have you ever been confused by an investigator’s proposed research protocol?

    Perhaps the investigator is using jargon only well known in their field. The same can go for undefined acronyms. Just because the investigator knows their proposed research so well, doesn’t mean the reviewer has that same level of knowledge. The easy route would be to send back as one of your revisions the following:

    “Avoid using technical jargon and define acronyms. This should be written in a way that anyone can understand.”

    Of course, this can be one of your revisions…but what I’m proposing here is to dig deeper. See what you can learn about the proposed research yourself.

    With everyone extremely busy, it can feel like you’re trying to just pump out reviews as fast as you can. Let me provide an example:

    Maybe your first go-to resource is seeing how other IRBs have reviewed similar research proposals.

    Sure, you can do this. This is a great way to see how other reviewers at other institutions review similar projects. Consider the following hypothetical example:

    • The researcher would like to see how well their device can measure the average calories participants burn in a day.
    • Their device is a wearable (i.e., a wristband).
    • They want to compare this device with a device currently on the market (e.g., Fitbit).
    • Their device is also linked to an app they created that will be downloaded to a participant’s phone.
    • From the app, the researcher would like to obtain physiological and geospatial data.

    At first glance, you may think…wow there are a lot of moving parts here. You may automatically wonder if FDA regulations apply in addition to the Common Rule. Or maybe, your head is spinning because you have no clue where to start. I urge you to think about the study design itself:

    1. You know the study team plans to compare their device to a Fitbit.
    2. You may already know that Fitbit has an app that is downloaded to a user’s phone.

    By researching the research, I propose to take the following action…

    1. Read through the Fitbit instruction manual,
    2. Review the instructions provided:
      • How to install the Fitbit app on your phone,
      • How to care for the Fitbit,
      • How to troubleshoot for an error, etc.
    3. Any potential risks listed by using the Fitbit,
    4. Any permissions required to ensure the Fitbit app accurately tracks your data

    By understanding the mechanism behind the proposed research, you will start to notice your questions coming together:

    1. Did the study team provide instructions on how to use the device?
    2. Did the study team consider all potential risks and how to mitigate these risks?
    3. Did the study team indicate the specific data measures that’ll be collected from the device’s app?
    4. Did the study team list what permissions are required from the participant’s phone to successfully use the app?
    5. Did the study team consider if their app must be integrated with other apps on the participant’s phone (e.g., the Health app)?

    By researching the research, you will find yourself asking the more critical questions. You should always consider what others are doing and regulations but try to put yourself in the researcher’s shoes. Help the researcher make their proposal that much more thought-out! By working together, we can help the researcher improve their protocol design. As a reviewer, you will start to think about the more in-depth questions to ensure adequate participant protections.

    The split-screen hack

    As a reviewer, we know we can have countless documents to review. This can be a daunting task in itself.

    On top of that, it can take a lot of time to ensure consistencies within various sections of the protocol.

    I would argue that the protocol is the most important document in a researcher’s proposal. The protocol how many important sections ranging from objectives to data collection methods, and so forth. I propose to use the “split-screen” function in Microsoft Word when reviewing the protocol document.

    What is the “split” function in Microsoft Word?

    By enabling this function, it essentially “splits” the document in half horizontally:

    • You will notice two scroll bars on the upper and lower half of the screen.
    • On the upper half of the screen, you can look at a certain section of the document.
    • On the lower part of the screen, you can look at a completely different section within the same document!

    Why would this be helpful, you may ask?

    This is particularly useful with long protocol documents. Let’s take the example of reviewing a study’s objective against the data collection methods. In a perfect world, these two sections within the protocol would be consistent with one another. Sometimes, they aren’t. Also, depending on the protocol, the objectives could be the first section of the protocol. Then, the data collection methods could be somewhere in the middle of the protocol.

    In lieu of scrolling back and forth in hopes you catch the inconsistencies; this ensures you absolutely catch any inconsistencies!

    This hack is a time saver.

    You can access this function in Microsoft Word using the steps below:

    1. Under the “Review” ribbon, within the “Window” section select “Split”.
    2. When you no longer need to use this, simply select “Remove Split” following the same steps listed in Step 1.

    Streamline documenting common errors you see on protocols

    What type of research does your institution typically review? Perhaps SBER?

    Let’s roll with this example. Say within SBER, you notice many of the submission you review conduct surveys and interviews. Maybe these submissions include observations or focus groups as data collection methods. Now, I want you to take a moment and reflect. Are there common errors you see within protocols that you always need the researcher to address?

    If you said “Yes”, then I recommend streamlining common errors you see into a template.

    By this, I mean creating a document with standard reviewer comments. The document should contain standard verbiage that can be easily copied, pasted, and fine-tuned depending on the study you’re reviewing. It’s difficult to have standardized verbiage for proposed research, so I recommend starting with what you see the most. You can slowly build your list of common errors as you see different types of proposed research.

    This efficiency tip isn’t just for you. This is something that should be shared!

    I encourage you all to share these standard errors you commonly see within protocols with your team members. Help the team be efficient in the review process. You can even make it a living document where team members can comment on what they see in their reviews. It’s always important to have multiple sets of eyes for something like this. Someone else can think of something you didn’t even consider. It’s important to not only develop yourself, but also help your team thrive! Dive into success together.

    Promote transparency in your communication with investigators

    Has anyone ever told you to do something, and you ask why this action is necessary?

    I know I have! I personally love to understand how things work. So, if someone says, “Hey, go do this”, I need to know why I have to. I need to understand why this particular step is important in the process. The same goes for researchers. They are inquisitive in nature. Therefore, if you’re asking a researcher to make a specific change in their protocol, tell them why this change matters.

    Say you’re reviewing a consent form within the researcher’s application, and you notice there isn’t a statement about participant risk.

    The researcher comes back and says, “Well, there isn’t any major risk to my study. I don’t see a point in saying this.”

    This is where the “why” to your “ask” comes into play. Sure, you state that addressing risk is a regulatory requirement (i.e., basic element of consent). You should also state that the participant should be aware if risks are present. The participant should have all the details necessary for them to make an informed decision to participate. If there are no foreseeable risks or discomforts, then the participant should be aware of this. This can impact the participant’s decision-making process. By explaining these concepts to the researcher, they will be more inclined to incorporate this statement into the consent form. They will also be mindful of this in future submissions.

    Reflect on oddball situations

    Have you ever had a submission that was outside of the norm? Did the review process require additional steps that typically wouldn’t be required?

    Now – I want you to imagine you came across another submission of this nature. Do you remember everything you did when you came across this oddball review the first time? Did you do your due diligence and document the process? Maybe you intended to but were caught up in another task and oops…you forgot. Here, I’d like to promote the IRB Precedent Tool. This is a structured way of documenting these oddball situations.

    You can incorporate key details of:

    • The situation itself,
    • Steps you took during the review process,
    • Any regulatory information that assisted you in making a determination, etc.

    Unsure what I’m talking about? You can read more about this brilliant approach here: Steps toward a System of IRB Precedent: Piloting Approaches to Summarizing IRB Decisions for Future Use

    Stay current with regulations and developments in the field

    This…I must admit is a tough tip to implement.

    With all the regulatory updates and publications coming out like clockwork, this can feel overwhelming. You can attempt this via:

    • Signing up for various agency newsletters (e.g., FDA if you have mainly biomedical research). Another great one is the Office of Human Research Protections (OHRP).
    • Rely on newsletters from various organizations such as PRIM&R for these types of updates.
    • Check the Federal Register daily for any new rules or notices.
    • Check agency websites daily for any regulatory updates.
    • Review journals on a daily basis for developments in the field (e.g., Ethics and Human Research).

    You can split the proposed tips above if you have multiple team members. One person (or multiple) can focus on regulatory updates while others focus on recent publications.

    But what about those with small HRPP offices where there could be only 1-2 people?

    Time for a little self-promotion. An excellent starting point would be to follow along this blog’s monthly posts (even if you aren’t in a small HRPP):

    • Research Compliance Chronicle (see the first edition here): a comprehensive review of regulatory updates from the previous month including:
      • Agency-specific news, policy updates, and new rules
      • Presidential actions (i.e., executive orders)
      • Proposed bills that can potentially impact research and compliance
    • RAC Digest (see the first edition here): a comprehensive review of recent developments in the research administration and compliance fields from the previous month including updates from journals such as:
      • Accountability in Research,
      • AJOB Empirical Bioethics,
      • JAMA, and many more!

    Leverage experiences from senior personnel

    Knowledge sharing is a hallmark in an optimal institution’s review committee.

    You might remember what it was like when you first started your career as a research reviewer. Remember being overwhelmed – thinking – how could I possibly learn all this? Well, this is where knowledge sharing comes into play. Senior personnel are the hidden gems within a review committee. Why? They’ve likely seen many types of proposed research throughout their career. Not only common types of research projects, but also oddball scenarios. Leverage their expertise and learn from them. Explain your thought process on a particular scenario. Senior personnel will likely provide insights that you didn’t initially think about.

    Consult with subject-matter experts

    Having a diverse network of subject-matter experts to consult can aid the review process.

    Engage with subject-matter experts when the particular topic within the proposed research is out of your purview. Similar to the tip above, these folks can provide the specific knowledge needed to ensure a comprehensive review. An example of this could include security concerns about an app the research proposes to use in their project. Building this diverse network of these folks can give you a shoulder to lean on to ensure adequate participant protection.

    I hope you found this post useful!

  • April 2025: RAC Digest

    April 2025: RAC Digest

    Modified on March 5, 2026 to remove the “subscribe” option. This blog has been retired and replaced by the S.P.I.R.I.T. newsletter.

    Good morning, good afternoon, and good evening, Compliance Rockstars, Clinical Research Professionals, Ethics Enthusiasts, and Investigators! 200+ blog subscribers and counting!

    Welcome to the RAC Digest!

    RAC stands for “Research Administration and Compliance”.

    The RAC Digest will feature select publications from the previous month from journals related to research administration and research compliance.

    As a general reminder, these are my own interpretations. Any legal information discussed within this post should be discussed with your institution.

    Let’s get started:

    Expanding Roles for Research Administration and Research Development Professionals: A Team Science Coaching Program

    • Addressing Interdisciplinary Challenges:
      • Interdisciplinary research teams often face issues like:
        • Misaligned goals,
        • Leadership struggles, and
        • Communication barriers.
      • The program aims to mitigate these challenges by introducing structured team coaching.
    • Training RA and RD Professionals as Coaches:
      • The initiative trained RA and RD staff to serve as team coaches, equipping them with skills to:
        • Support team formation,
        • Collaboration, and
        • The development of research outputs.
    • Utilization of Team Science Tools:
      • Coaches employed tools informed by the Science of Team Science (SciTS) to facilitate effective team dynamics and project progression.
    • Natural Extension of RA and RD Roles:
      • Many found the coaching role to be a seamless addition to their existing responsibilities, although interpretations varied among individuals.
    • Focus on Teamwork and Taskwork:
      • Coaches addressed both interpersonal dynamics (teamwork) and project-related tasks (taskwork), ensuring comprehensive support for research teams.
    • Support for Diverse and Virtual Teams:
      • The coaching model proved particularly beneficial for teams integrating diverse perspectives and collaborating virtually over extended periods.
    • Implications for Research Administration:
      • The program highlights the potential for RA and RD staff to play a more active role in facilitating interdisciplinary research.
      • This suggests shift towards more integrated support structures within research institutions.

    The Representative Studies Rubric: A Tool for Diversity in Clinical Trials

    • The Representative Studies Rubric (RSR) was created to promote diversity in clinical trials.
      • This 12-item questionnaire that assessed whether protocols included or excluded underrepresented groups such as:
        • People of all ages,
        • Women (cis and trans),
        • Gender nonbinary individuals,
        • People who inject drugs, and
        • Pregnant people.
      • It also evaluated the use of inclusive statistical practices, community engagement, and stigmatizing language.
    • Many current HIV trials still exclude underrepresented populationsโ€”often without justification.
      • The retrospective review of 47 NIH-funded HIV/AIDS trials showed frequent unjustified exclusions of:
        • Transgenders,
        • Gender nonbinary people, and
        • People who inject drugs.
    • Pregnant people are frequently excluded, limiting critical safety and efficacy data.
      • This occurred without clear scientific rationale of exclusion.
    • Only half of the studies included specific enrollment goals for diverse populations.
    • As a result, the NIH-funded HIV/AIDS Clinical Trials Networks mandated RSR use for future protocols to proactively promote diversity.

    If your institution can’t access the publication, you can download the article here: The Representative Studies Rubric: A Tool for Diversity in Clinical Trials

    “Dear Editor, may I speak with you?”

    • Many journals prohibit appeals and limit communication after manuscript rejection, preventing authors from understanding editorial decisions.
    • Factors such as an author’s prestige, institutional affiliation, or editorial board membership can unfairly influence publication decisions.
    • Restricting the number of peer-review rounds can prevent authors from fully addressing feedback and improving their work.
    • There is a call for advocacy for greater editor-author communication to foster fairness and mutual understanding in the review process.
    • Authors should investigate journal policies before submission, while editors should embrace transparent, responsive practices in the process.

    If your institution can’t access the publication, you can download the article here: “Dear editor, may I speak with you?”

    Perceptions of network-level ethics in an engineering research center: Analysis of ethical issues & practices reported by scientific & engineering participants

    • There were significant variations in how authorship was assigned e.g., some labs used structured discussions.
      • A major concern from participants was unequal contributions to the research.
    • Many engineers deferred ethical considerations to biomedical partners, viewing ethics as outside their domain.
    • Ethics and regulatory concerns (e.g., FDA requirements) were not deeply integrated into daily research decisions.
      • Some participants feared over-regulation could suppress innovation.
    • While collaboration was a central goal, early-stage efforts were often informal or idea-based.
    • Participants wanted clearer, network-wide policies on data sharing, authorship, and ethics training.
    • Many participants admitted limited understanding of ethics beyond lab-level compliance expressed desire for more training.

    If your institution can’t access the publication, you can download the article here: Perceptions of network-level ethics in an engineering research center: Analysis of ethical issues & practices reported by scientific & engineering participants

    SPIRIT 2025 Statement: Updated Guideline for Protocols of Randomized Trials

    • SPIRIT 2025 is an updated guideline for writing protocols of randomized trials.
      • The revised checklist within the guidance integrated contemporary issues like open science, patient involvement, and transparency in reporting.
    • Several other items were revised for clarity and completeness such as:
      • Data sharing,
      • Conflicts of interest,
      • Blinding, and
      • Statistical methods.
    • SPIRIT 2025 incorporates elements from CONSORT Harms 2022, SPIRIT-Outcomes 2022, and the TIDieR checklist.

    If your institution can’t access the publication, you can download the article here: SPIRIT 2025 Statement: Updated Guideline for Protocols of Randomized Trials

    Preserving Research Ethics Oversight Amid Decimation of the Research Enterprise

    • The following actions of the Trump administration has threatened the funding and functioning of IRBs:
      • Cutting grants,
      • Capping indirect costs, and
      • Reducing regulatory staff
        • All three of these components are essential for protecting research participants and maintaining public trust.
    • IRBs play a vital, legally mandated role in ensuring ethical conduct in human subjects research.
      • Their effectiveness depends on institutional support and federal guidance.
    • Massive cuts to OHRP and the termination of SACHRP weaken the federal governmentโ€™s ability to enforce ethical standards.

    If your institution can’t access the publication, you can download the article here: Preserving Research Ethics Oversight Amid Decimation of the Research Enterprise

    CONSORT 2025 Statement: Updated Guideline for Reporting Randomized Trials

    • CONSORT 2025 introduces a 30-item checklist for reporting randomized trials. Some new changes included:
      • Reporting on data sharing,
      • Conflicts of interest, and
      • Patient/public involvement.
    • Items from related CONSORT extensions (e.g., on harms, outcomes, nonpharmacological treatments) and guidelines like TIDieR were incorporated.
    • Revisions to existing items emphasize transparency about protocol access and changes made after trial commencement.
    • The checklist aligns more closely with SPIRIT 2025 providing consistent guidance from protocol design to trial reporting.

    If your institution can’t access the publication, you can download the article here: CONSORT 2025 Statement: Updated Guideline for Reporting Randomized Trials

    Generalizability of FDA-Approved AI-Enabled Medical Devices for Clinical Use

    • Only 56% of the 903 FDA-approved AI-enabled medical devices reported clinical performance studies at the time of approval, raising concerns about the depth of evaluation.
    • Fewer than one-third of clinical studies included sex-specific or age-specific data, limiting assessments of device performance across diverse patient populations.
    • Lastly, the study emphasized:
      • The need for continuous post-market monitoring,
      • Improved reporting standards, and
      • Stricter regulatory oversight to ensure safety, effectiveness, and generalizability of AI-enabled devices in real-world clinical settings.

    I hope you found this content useful!

  • ICH GCP E6(R3) vs. E6(R2): Guideline Differences

    ICH GCP E6(R3) vs. E6(R2): Guideline Differences

    Modified on March 5, 2026 to remove the “subscribe” option. This blog has been retired and replaced by the S.P.I.R.I.T. newsletter.

    Tasha Mohseni

    (Author)

    Adnan Shaikh

    (Author & Reviewer)

    Good morning, good afternoon, and good evening Compliance Rockstars, Clinical Researchers, Ethics Educators, and Investigators from around the globe!

    It’s hard to believe that January is coming to a close. It feels like just yesterday folks were sending their New Year wishes. It felt as though I was reviewing countless study submissions. I prefer to be on my toes than idle!

    ICYMI, TikTok went dark on January 18, 2025 only to return on January 20, 2025. You can read about my insights here:

    Now, back to today’s post. On January 6, 2025 the final version of the ICH GCP E6(R3) guidelines. Whether you’re a novice to clinical research (like me) or an expert, this is a HUGE deal. The last revision for ICH GCP E6(R2) was back in November 2016. Without writing this post, I could guess that there are significant differences. But I don’t like to guess…I like to know and understand why.

    For background, I primary review IRB studies related to social, behavioral, and education research (SBER). I am extremely interested in learning more about clinical research ethics and clinical trials in general.

    Therefore, I would like to thank Adnan Shaikh for his willingness to review and co-author this post!

    In this post, we plan to describe in detail the differences between the E6(R2) version and the E6(R3) version. As a general disclaimer, these are our own interpretations of these guidelines. If you have any questions about these guidelines, you should always consult with your institution.

    Tasha Mohseni’s POV on General Guideline Differences

    Aside from document length, the table below summarizes the general differences between the guidelines:

    ICH GCP E6(R2) SummaryICH GCP E6(R3) Summary
    1. Published in November 2016.
    2. It emphasized updates needed to integrate modern technology and risk-based approaches into clinical trials.
    3. Key updates included:
    A. Adoption of risk-based monitoring.
    B. Enhanced guidance on electronic records and data management.
    C. Emphasis on quality management systems tailored to clinical trial complexities.    		1. Published in January 2025.
    2. Introduced broader revisions across clinical trial design, oversight, and ethical conduct, emphasizing:
    A. Greater focus on patient needs and preferences.
    B. Enhanced data reliability and transparency mechanisms.
    C. Encouragement of adaptive and innovative trial designs.
    D. Modernized guidance for handling big data and novel analytical tools.

    What Did ChatGPT Have to Say About The Guideline Differences?

    As you all know from my Show and Tell: How to Write a Clear SBER Protocol post, I have my own personal ChatGPT account.

    I can’t even quantify how long it took me to summarize the general differences for the two documents. Though this was interesting to review, I found it difficult to sift through the information. This might be because I haven’t had exposure in this area. I likely should have completed the GCP CITI training course to provide a foundation. Alas, here we are!

    Out of curiosity and efficiency, I decided to upload both guidelines to ChatGPT and asked for itemized differences. Below is a complete response from ChatGPT:

    “To provide a detailed, itemized comparison of the ICH E6(R2) and ICH E6(R3) documents, I will identify key differences across major sections and topics in both documents. This will include updates, additions, and revisions. Here’s an itemized list based on their content:

    Introduction and Scope

    • ICH E6(R2):
    • Introduced updates to integrate risk-based monitoring and electronic systems.
    • Focused on maintaining ethical principles while improving trial efficiency and data reliability.
    • ICH E6(R3):
    • Broader scope, emphasizing patient-centric approaches and advanced technologies like decentralized trials and real-world evidence.
    • Greater emphasis on stakeholder collaboration and global harmonization.

    Quality Management

    • ICH E6(R2):
    • Introduced a risk-based quality management system.
    • Focused on identifying and mitigating risks to critical data and processes.
    • ICH E6(R3):
    • Expands the quality management framework to include:
      • Adaptive approaches.
      • Continuous quality improvement throughout the trial lifecycle.
      • Integration of advanced analytics and real-world data for monitoring.

    Data Management

    • ICH E6(R2):
    • Introduced guidance on electronic systems and electronic data capture.
    • Added the requirement for an audit trail for all changes in electronic data.
    • ICH E6(R3):
    • Extends guidance on managing big data, artificial intelligence (AI), and machine learning (ML) tools.
    • Updates standards for data security, privacy, and transparency in decentralized trials.

    Trial Design

    • ICH E6(R2):
    • Focused on traditional trial designs, with some flexibility for centralized monitoring.
    • ICH E6(R3):
    • Encourages adaptive, innovative, and patient-centric trial designs.
    • Includes guidance for decentralized clinical trials (DCTs) and the use of real-world evidence (RWE).

    Investigator Responsibilities

    • ICH E6(R2):
    • Detailed investigator obligations, focusing on protocol adherence and oversight.
    • Introduced responsibility for supervising delegated tasks.
    • ICH E6(R3):
    • Expands the investigatorโ€™s role to ensure oversight in remote and virtual settings.
    • Adds accountability for new roles introduced in decentralized models (e.g., telemedicine).

    Monitoring

    • ICH E6(R2):
    • Introduced risk-based monitoring (RBM) for the first time.
    • Emphasized centralized monitoring over traditional on-site monitoring.
    • ICH E6(R3):
    • Refines RBM principles and promotes greater use of AI-driven monitoring systems.
    • Integrates remote monitoring for decentralized trials.

    Patient-Centric Approaches

    • ICH E6(R2):
    • Focused primarily on data integrity and participant safety.
    • ICH E6(R3):
    • Strong emphasis on patient engagement, including:
      • Simplified informed consent processes.
      • Participant-centered trial design.
      • Use of digital tools to enhance patient experience and access.

    Ethical Considerations

    • ICH E6(R2):
    • Rooted in the principles of the Declaration of Helsinki, with updates for vulnerable populations.
    • ICH E6(R3):
    • Expands ethical considerations to include:
      • Diversity and inclusion in clinical trial participation.
      • Enhanced protection for participants in decentralized and global trials.

    Use of Technology

    • ICH E6(R2):
    • Introduced electronic systems, requiring validation and audit trails.
    • ICH E6(R3):
    • Extends to newer technologies like AI, blockchain, and wearable devices.
    • Requires validation and traceability for advanced digital systems.

    Regulatory Harmonization

    • ICH E6(R2):
    • Harmonized standards for the EU, U.S., and Japan.
    • ICH E6(R3):
    • Broader harmonization efforts, including additional regions (e.g., Canada, Switzerland).
    • Greater emphasis on mutual regulatory acceptance of trial data.

    Risk-Based Approaches

    • ICH E6(R2):
    • Introduced risk-based approaches for trial management and monitoring.
    • ICH E6(R3):
    • Refines risk-based approaches with continuous risk assessment throughout the trial lifecycle.
    • Incorporates new methodologies to address emerging risks in decentralized and digital trials.

    Terminology and Glossary Updates

    • ICH E6(R2):
    • Updated glossary to include terms like “risk-based monitoring” and “certified copy.”
    • ICH E6(R3):
    • Expands glossary to include modern terms like “decentralized trials,” “real-world data,” and “validation of AI systems.”

    These itemized differences highlight the evolution of clinical trial guidelines from E6(R2) to E6(R3), reflecting advancements in technology, trial design, and ethical considerations. Let me know if you want a deeper analysis of any specific section!”

    At face value, this seems to be an impressive comparative analysis.

    You might be wondering why I said the word “seems”.

    I say this because I am acknowledging that I am not a subject-matter expert. I can’t even say that I have direct experience with this other than reading the guidelines themselves. I can’t necessarily validate if ChatGPT assessed the information correctly. I could have done more iterative prompting to receive an even more detailed response…but for what? This is why it is so important to have the expertise to verify the information is correct. Whether you become knowledgeable or collaborate with someone who has the expertise. Let’s see what Adnan Shaikh has to say about ChatGPT’s analysis!

    Adnan Shaikh’s POV on the ChatGPT Analysis (in his own words)

    Hey everyone! I hope you’re enjoying reading Tasha’s content. Here’s my take on the ICH GCP E6(R2) versus ICH GCP E6(R3).

    The adoption ICH GCP E6(R3) introduces a transformative update to the principles and practices of clinical research:

    • The ICH GCP E6(R2) guidelines were more focused
      • They adhered to the protocol and followed the traditional approach in clinical trial studies, that has an impact of โ€œone size fits all”
    • While ICH GCP E6(R3) is more modern and advanced approach for clinical trial, it encourages โ€œfit for purposeโ€
      • Which means that proportionality and risk-based approaches focus on the quality of clinical trials
        • This is critical and fundamental to the safety of participants and the reliability of participants results

    I humbly recognize that I am not as expert and that I am still learning

    However, to summarize the overall concept of ICH GCP E6(R3), it focuses on the following:

    • Emphasize proportionality risk-based approach,
    • Incorporates innovative technologies, and
    • Promotes adaptability to modern clinical trial design.

    In the recent times, AI is claimed to be one of the best tools in helping with clinical research.

    Now, I would like to share my POV on ChatGPT.

    It is helpful in many ways, for example:

    • Preparing document with little to no grammatical errors,
    • Providing information on particular drug molecules, and
    • Providing information on ICH GCP guidelines

    When I prompted ChatGPT about the difference between ICH GCP E6(R2) vs ICH GCP E6(R3), I received the following graphical explanation:

    Image of ChatGPT’s graphic representation of the differences between ICH GCP E6(R2) and ICH GCP E6(R3)

    As you can see, this further illustrates my point that ICH GCP E6(R3) highly focuses on modern adaptation. By implementing this concept, it can provide more flexibility in clinical research studies. In the era of AI and emerging technologies, itโ€™s important to adapt the modern techniques in clinical trials. It is also criticial to note that this must be done without compromising confidentiality. We must adhere to ethical standards for the responsible use of these emerging technologies.

    I am expressing my sincere gratitude to Tasha for involving me in this post. I ‘m looking forward to working with her on future projects!

    We hope you find the content useful and thought-provoking! What do you think about current guidelines? Please lead the discussion and leave a comment below!

    Again, thank you Adnan Shaikh for your invaluable expertise! For those who would like to learn more about him, you can follow him on LinkedIn and read his brief biography below.

    You can follow Adnan on LinkedIn, Instagram, and WhatsApp!

    Adnan Shaikha is a Pharm.D (Doctor of Pharmacy) candidate. He is completing his clinical pharmacy internship at New Civil Hospital. His degree will be from the Shree Dhanvantary Pharmacy College, Kim, Surat, Gujarat, India. He is expecting to graduate with his PharmD degree in June 2025. Additionally, he is doing a certification course on medical writing in clinical research.

    We wish you the best of luck on this exciting endeavor and soon-to-be new chapter in your life!

  • FDA Artificial Intelligence (AI) Guidance Highlights

    FDA Artificial Intelligence (AI) Guidance Highlights

    Modified on March 5, 2026 to remove the “subscribe” option. This blog has been retired and replaced by the S.P.I.R.I.T. newsletter.

    Tasha Mohseni's avatar

    Tasha Mohseni

    Good morning, good afternoon, and good evening IRBers, clinical research educators, and investigators from around the world!

    I hope everyone had a great first of full week back to work! Now, I can officially say, back to the grind. I should be used to how busy the start of the semester is with outreach training efforts. I was also busy reviewing submissions. We typically see an increase in submissions around this time since it is the start of Spring semester.

    Besides being busy with work, I was also working on the Show-and-Tell Series of blog posts. You can read about the series here:

    Advertisements

    Well, I wasn’t the only one busy this week. The FDA went on a guidance posting frenzy! Below is a list of guidance documents relevant to IRB and clinical research that were issued between Monday, January 6, 2025 – Friday, January 10, 2025:

    Though all these guidance documents should be reviewed, I plan to only deep dive into the following guidance documents:

    You may be wondering why I’m solely focusing on these two guidance documents.

    I’ll tell you why I’m into AI!

    • AI can optimize productivity. I canโ€™t tell you how many times this has saved me. For people like me who have trouble reading long documents, AI is great for summarizing key concepts. Of course, I will read all documents in their entirety. However, itโ€™s nice to have a general idea of what I plan to read. This way, if there are any concepts or terms Iโ€™m unfamiliar with, I can simply look them up. Then, when Iโ€™m reading the document in full, I wonโ€™t have to waste time looking up terms and concepts.
    • AI can help you be creative. I canโ€™t wait to share a post related to this! Before I think about automating a task, I ask ChatGPT if itโ€™s possible. Then, with my skills Iโ€™ve acquired over the years, I can attempt to act on my efficiency idea. I also love the DALL-E aspect. This is great for visual folks. I especially love to have DALL-E create flowcharts or diagrams. This helps me understand complex topics (such as the ethical codes in human subjects research).
    • AI can be used in any facet. Whether youโ€™re a writer, an artist, in IT, or even compliance, AI can be helpful anywhere! I will say that itโ€™s important to make your audience aware when AI was used. I always love to give credit where credit is due. I feel that AI can even make the least creative personโ€ฆa creator.

    Though I sound pro-AI, I do see there are downsides.

    • Privacy and confidentiality are major concerns. This likely goes without saying, but I will say it anyway. Compliance personnel such as myself likely know you shouldnโ€™t place any personal information into ChatGPT. Well, others may not know this. What about ways to withdraw your data? Can you do that in ChatGPT? I know there is a way to export data that was entered into the tool via your ChatGPT settings. But can ChatGPT unlearn data that has been withdrawn? Iโ€™m not sure, but I hope to learn more about this.
    • Machine learning bias is real. For those who may not be familiar with machine learning, this is a subset of AI. There isnโ€™t actual programming (e.g., with Python). It learns from datasets and makes inferences based on patterns. This is why it is called โ€œmachine learningโ€ because the tool learns over time. I feel this is GREAT for a highly specific function (such as a customer service chatbot). But what about when AI is being used for biomarker analysis or drug development? How can we ensure that we are applying the Belmont principle of Justice (subjects must be fairly selected)? How do we ensure we are fairly selecting datasets that are representative of the population of interest?
    • Compliance professionals are in an arms race in how to regulate the rapid use of AI in research. Again, this likely goes without saying it. You can probably start calling me โ€œCaptain Obviousโ€ now. Even as I learn about AI, it is hard to keep track of everything thatโ€™s going on. I follow federal agencies for guidance documents or if strategic plans are released discussing the ethical uses of AI. To me, the problem seems to be that everyone is developing their own guidance documents and best practices. It seems to me that something like this should have a best practice standard that agencies can adopt. I plan to learn about the EU AI Act in further detail as this seems like a great start. I recently completely GDPR trainings and felt this regulation really covered everything. We desperately need something like this in the United States.

    In this post, I plan to highlight key takeaways from these lengthy AI guidance documents (90 pages total).

    Then, I plan to analyze the documents even further as the request for comment is due Monday, April 7, 2025.

    If you are interested in leaving a public comment with me, please email me at tmohseni@renovationinirbeducation.com.

    Let’s make our voice count TOGETHER!

    Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making for Drug and Biological Products

    Per the FDA, this guidance provides recommendations to sponsors and other interested parties on the use of AI to produce informations or data intended to support regulatory decision-making regarding the safety, effectiveness, or quality of drugs. Though I am not an expert in drug development, I will say that I have intermediate understanding of AI. Letโ€™s see what the FDA has to say.

    The guidance provides a risk-based credibility assessment framework that may be used for establishing and evaluating the credibility of an AI model for a particular context of use (COU).

    The COU defines the specific role and scope of the AI model to address a specific question. As a former auditor, I also appreciate that the FDA has defined the word โ€œshouldโ€. The definition of โ€œshouldโ€ means that the FDA recommends actions within this guidance, but they arenโ€™t required. I remember carefully reviewing policies for words like โ€œshouldโ€, โ€œshallโ€, or โ€œmustโ€. Itโ€™s important for institutions to define this as well. This way, when folks are reviewing their institutionโ€™s policy or guidance, they know what is required versus what is recommended.

    A Risk-Based Credibility Assessment Framework

    This is a 7-step process:

    • Step 1: Define the question of interest that will be addressed by the AI model.
    • Step 2: Define the COU for the AI model.
    • Step 3: Assess the AI model risk.
    • Step 4: Develop a plan to establish credibility of AI model output within the COU.
    • Step 5: Execute the plan.
    • Step 6: Document the results of the credibility assessment plan and discuss deviations from the plan.
    • Step 7: Determine the adequacy of the AI model for the COU.

    Okay, so we know the steps. What do we do for each of these steps?

    Step 1 should describe the specific question, decision, or concern being addressed by the AI model. For step 2, the description of the COU should describe in detail what will be modeled and how model outputs will be used. It should also be notated on whether other information will be used in conjunction with the AI modelโ€™s output to answer the question of interest determined in step 1. Examples of other information include animal studies and/or clinical human research studies. In step 3, model risk is assessed by two factors: model influence and decision consequence. Model influence, like it sounds, compares data derived from the AI model to other evidence used to inform the question of interest in step 1. Decision consequence is the significance of an adverse outcome resulting from an incorrect decision concerning the question of interest in step 1. To appropriately assess these components of model risk, subject-matter expertise is strongly advised.

    Step 4 describes what information should be in your credibility assessment plan. Below is a summarized list of information that should be considered:

    • Describe the datasets used for training and tuning the AI model and which model development activities were performed using these datasets
    • Describe how the development data have been or will be collected, processed, annotated, stored, controlled, and used for training and tuning the AI model
    • Describe how the development data is fit for the COU
    • Describe whether the development data are centralized
    • Describe how the AI model was trained
    • Specify if a pre-trained model was used
    • Describe the use of ensemble methods
    • Explain any calibration of the AI model
    • Describe the quality assurance and control procedures of computer softwares and how version changes were tracked (as well as code verification)
    • Describe the applicability of the test data to the COU to minimize data drift
    • Describe the agreement between the model prediction and the observed data
    • Provide rationale for the chosen model evaluation method
    • Describe any model limitations and biases

    For step 5, discussing the credibility assessment plan with the FDA prior to execution may be helpful. The last section of the document describes early engagement options s with the FDA. Step 6 should involved documenting the results and deviations from steps 1-4. Once this is complete, you can proceed to step 7. Step 7 is where you determine if the AI model is appropriate for the COU. Finally, the document concludes with life cycle maintenance of the credibility of the AI model output in certain COUs. This can be referred to as the management of changes to an AI model (whether incidentally or deliberately).

    Artificial Intelligence-Enabled Device Software Functions: Lifecycle Management and Marketing Submission Recommendation

    Per the FDA:

    This draft guidance, when finalized, will represent the current thinking of the FDA on this topic.

    Though this document will represent FDAโ€™s thoughts, I appreciate FDAโ€™s flexibility in approaches towards these recommendations. So long as the applicable statutes and regulations are met and it has been discussed with the FDA, you can use an alternative approach. The guidance provides recommendations on the contents of marketing submissions for devices that include AI-enabled device software functions including documentation and information that will support FDAโ€™s review. Similar to the previous guidance, the FDA defines the word โ€œshouldโ€ as โ€œsuggestedโ€ or โ€œrecommendedโ€. Nowโ€ฆletโ€™s get into this document!

    The FDA promotes a total product life cycle (TPLC) approach to the oversight of medical devices. You can read more about TPLC here: Total Product Life Cycle for Medical Devices. They also discussed the recent efforts made such as the 10 tenets of Good Machine Learning Practice (GMLP). The document further defines terminology used by the FDA versus the general AI community. For example, using the term โ€œvalidationโ€ to represent โ€œtrainingโ€ or โ€œtuningโ€ should be avoided in medical device marketing submissions. Instead, the word โ€œdevelopmentโ€ should be used. The FDA Digital Health and Artificial Intelligence Glossary โ€“ Educational Resource provides a compilation of commonly used AI Terms and how the FDA defines them.

    The next few sections within this guidance are what the FDA recommends including in marketing submissions. Each section provides a reason as to why it must be included, what must be included, and where to include it. Below is a general outline of what is recommended for submission:

    General Outline for Marketing Submissions

    • Device description
      • A statement that AI is used in the device
      • A description of device inputs and outputs
      • An explanation of how AI is used to achieve the deviceโ€™s intended use
      • A description of the intended users, their characteristics, and the level and type of training they are expected to have and/or receive
      • A description of its intended use environment(s)
      • A description of the intended workflow for the use of the device
      • A description of installation and maintenance procedures
      • A description of any calibration and/or configuration procedures
      • If the device can be configured by a user, then the submission should include information about:
        • All configurable elements of the AI-enabled device
        • How these elements and their settings can be configured
        • The potential impact of the configurable elements on user decision-making
      • If a device contains multiple connected applications with separate interfaces, then the device description should address all these applications
    • User Interface
      • A graphical representation of the device and its user interface
      • A written description of the device user interface
      • An overview of the operational sequence of the device and the userโ€™s expected interactions with the user interface
      • Examples of the output format
      • A demonstration of the device
    • Labeling
      • The following should be included at the age-appropriate reading level for the intended user:
        • Inclusion of AI
        • Model input
        • Model output
        • Automation
        • Model architecture
        • Model development data
        • Performance data
        • Device performance metrics
        • Performance monitroing
        • Limitations
        • Installation and use
        • Customization
        • Metrics and visualization
        • Patient and caregiver information
    • Risk assessment
      • Risk management file
    • Data management for both training and testing data
      • Data collection
      • Data processing and cleaning
      • Reference standard
      • Data annotation
      • Data storage
      • Management and independence of data
      • Representativeness
    • Model description and development
    • Performance validation
    • Device performance monitoring
    • Cybersecurity
      • Cybersecurity risk management report
      • How cybersecurity testing addresses the risks in the report
      • A security use case view(s) that covers the AI-enabled considerations for the Debi e
      • A description of controls
    • Publication submission summary
      • A statement that AI is used in the device
      • An explanation of how AI is used as part of the deviceโ€™s intended use
      • A description of the class of model and its limitations
      • A description of development and validation datasets
      • A description of the statistical confidence level of predictions
      • A description of how the model will be updated and maintained over time

    I hope you found this content enlightening and useful! I strive to provide my readers “food for thought”. What did you think of my interpretation of the guidance documents? Please leave a comment below and let’s get this discussion started!

    Advertisements