Support a Guideline Evidence Request With Neutral Sources

A committee member reaches out with a specific question. They are reviewing a guideline and want to know what the evidence actually shows before they vote on a change. This is one of the more comfortable requests a medical science liaison gets, because the ask is for information rather than opinion. But comfortable does not mean simple. The moment you say more than the data supports, you stop supporting a guideline review and start steering one, and that line is easy to cross without noticing.

Why This Conversation Goes Wrong

The risk here is not that you will lie or fabricate anything. It is subtler than that. A committee member under time pressure often wants a shortcut. They want you to tell them what the answer should be, not just what the studies say. They might ask something like "does the data support keeping this drug first line" or "should we change our protocol." Answering that directly, even with good intentions, puts you in the position of recommending a guideline change instead of informing one.

There is a second failure mode that is just as common. You want to be helpful, so you hand over everything you have, including a promising abstract from a conference poster or an early phase study that has not been peer reviewed. The committee member does not always know how to weigh that material against a large randomized trial. If you do not flag the difference clearly, you have handed them an evidence base that looks more solid than it is. Guideline committees rely on liaisons who are precise about what a source can and cannot support. Lose that precision once and your materials get treated with more suspicion going forward, not less.

How to Prepare

Before you respond to the request, get clear on four things. First, identify the exact clinical question under review. A guideline covers many populations and endpoints, and a generic answer wastes the committee's time. Second, ask about the publication window that matters to them. Some committees only want evidence from the last three to five years, others want the full history including the original pivotal trials. Third, find out whether they want primary source papers, a summary evidence table, or both. Fourth, check what your medical affairs team has already approved for this exact use, since committee evidence requests usually need to go through the same scientific exchange channels as any other unsolicited request.

Pull together your source list before the call if you can, sorted by evidence strength. Mark which studies are peer reviewed and published, which are preprints or conference presentations, and which are company sponsored versus independent. This sorting work is the same discipline covered in Assemble a Cited Data Package for Review at a Health Systemand it applies just as much to a guideline committee as to a formulary review. A committee member who gets a dated, sourced, clearly labeled package trusts it more than one who gets a folder of PDFs with no context.

What to Say

Keep your language focused on the evidence and hand the interpretation back to the committee every time you feel the pull to do otherwise.

The committee member opens with, "Given the newer trial data, do you think we should drop this agent from first line therapy?" Notice the trap. Answer that question directly and you have made the committee's decision for them.

You respond, "That decision belongs to your committee, and I want to support your review rather than direct it. I can give you the published trials on this question, including the newer one you mentioned, along with the original studies your current guideline is based on, and I can walk through how each was designed. Which patient population is your subcommittee focused on, first line general use or a specific subgroup?"

The member says they are looking at patients over sixty five with at least one prior treatment, and asks for everything published in the last four years plus the two pivotal trials that established the original recommendation.

You reply, "I can get you that full set. I will separate the peer reviewed publications from anything still in preprint or abstract form, and I will note where the newer trial used a different endpoint than the original studies, since that affects how directly they compare. I will have this to you before your next meeting. Is there anything on study design specifically that your group wants me to walk through, such as how the newer trial defined its primary endpoint?"

The member asks for a short written note on the endpoint difference alongside the sources. You confirm you will include it and repeat back the exact list, the population, the date range, and the endpoint note, so there is no ambiguity about what you are delivering.

Mistakes to Avoid

Do not answer a should question with an opinion, even a softened one like "personally, I'd lean toward." A committee member will remember that line more than the ten pages of citations you sent. Do not send unpublished or preprint data without a clear label, and do not let a member assume company sponsored research carries the same weight as an independent trial unless you say so out loud. Do not skip the step of confirming which sources are cleared for this kind of exchange. Not every internal analysis is meant for external committees. Do not let the request drift into recommending specific guideline language, even if the member asks you to draft wording. That is theirs to write, not yours. And do not treat this as a one and done task. If new data becomes available before the committee votes, tell them, since silence there reads as withholding.

How to Practice and Score This

A manager reviewing this conversation should check for a few concrete things. Did the liaison ask a clarifying question before sending anything, rather than just forwarding a pre built folder of papers. Did the liaison label study types clearly, peer reviewed versus preprint, sponsored versus independent. Did the liaison ever answer a direct should question, or did every version of that question get redirected back to the committee's own process. Did the liaison follow up with a written summary of exactly what was promised.

This is a good scenario to rehearse before it happens live, because the pressure to just answer the should question is stronger in the room than it feels in hindsight. DealSpeak lets a medical science liaison run this exact exchange against an AI counterpart that pushes back the way a real committee member does, asking leading questions and testing whether you hold the line. The scoring highlights any moment where language slides from informing into recommending, which is the exact failure point that is hard to catch on your own. A related scenario worth rehearsing alongside this one is in Guide an Independent Research Inquiry to the Right Processsince both situations depend on the same skill of answering a scientific question without drifting into advice on what someone else should decide.

Closing

A guideline evidence request is a chance to demonstrate exactly what makes a medical science liaison valuable, precise sourcing, honest labeling of evidence strength, and total restraint on the actual decision. Practice the moments where the pressure to just give an answer is highest, since that is where the conversation actually gets tested. You can find more scenarios like this one among the Pharmaceuticals practice conversations built for exactly this kind of exchange.