Copy any prompt into a chatbot, fill in the brackets, and turn vague reassurance into real numbers. Use the answers to ask sharper questions at your next visit. These are starting points for a better conversation, not a substitute for one.
Tap Copy on any card, paste it into your AI assistant, and replace the blue brackets with your own details.
Take this risk statement I was given: [CLAIM]. Convert any relative risk into absolute risk. Tell me how many women out of 1,000 are affected with and without the intervention. Flag any place where the wording made the risk sound bigger or smaller than it is.
I was told to do this in pregnancy: [RECOMMENDATION]. Show what ACOG, NICE, and recent meta-analyses each say. Note where they agree, where they disagree, and where the advice runs ahead of the evidence. Name the strongest studies.
I was offered [TEST OR PROCEDURE]. List the questions to ask before I agree: the absolute benefit, the absolute risk, what happens if I wait, what happens if I decline, and the alternatives. Write them so I can read them aloud at my appointment.
Review this recommended item: [TEST]. Tell me if it is screening or diagnostic, what a positive result actually changes, the false positive rate, and whether routine use is backed by evidence or by habit. Explain in plain language.
For my situation: [STAGE OR CONDITION]. Give me the exact numbers to track: blood pressure thresholds, fetal movement counts, weight ranges, glucose targets. Tell me the specific value that should prompt a call, not just the words high or low.
Act as a maternal-fetal medicine specialist. Stress-test my birth plan: [PLAN]. Show where it could collide with hospital policy, where the evidence supports me, where it does not, and what I should negotiate in advance. Rank the points most likely to be challenged.
I am pregnant and considering [MEDICATION]. Summarize the evidence on safety in each trimester, the absolute risk of harm if known, and what is unknown versus proven dangerous. Separate established evidence from single studies.
For my stage of pregnancy: [WEEKS]. List the warning signs that need urgent attention and the symptoms that are normal. For each red flag, give the specific threshold and what to do. Separate true emergencies from common worries.
Here is the plan my care team proposed: [PLAN]. Build the strongest evidence-based case for a reasonable alternative. Identify the assumptions behind the original plan, where they may not apply to me, and the questions that would expose a weak recommendation.
How to use these well. AI can fabricate numbers, references, and doses that look correct. Ask it to separate evidence from inference and to say when it is unsure. Bring the answers to your doctor, midwife, or care team and check them together. The goal is a better conversation built on numbers you both can see, not a diagnosis from a chatbot.
