Lesson 1 of 4 · 4 min read
Picking the right study
Most research posts fail before they're written, because they start with an interesting paper instead of a patient's question. Flip it.
Start with the question. Take one from your question jar: "Does collagen powder do anything for skin?" Now go find the evidence.
Search where the evidence lives. PubMed for papers, your specialty society for guidelines, the Cochrane Library for systematic reviews. Filter by the last 5 to 10 years unless the classic study is still the answer.
Pick by strength. Guidelines and systematic reviews first, then randomised trials, then cohort studies. A single small study or a mouse study is a "promising, not proven" post, never a "this works" post.
Four quick checks. How many people? Who were they, and are they like your patients? What exactly was measured, and does it matter to a patient? Who paid for it?
One study, one post. If you need three studies to make the point, the post is probably two posts.
Your action step: Pick one patient question and find one guideline or review that answers it. Run the four checks.
Want me to make your first post for you, free, in your specialty?
Drop your Instagram handle. One post, on your page's look, so you can judge it on your own page. No card.
Reply POST to any of my course emails with your Instagram handle, or message me on Instagram. Lesson 2 of 4 · 5 min read
Plain-language translation
Patients trust what they understand. Your job isn't to dumb it down, it's to translate.
One finding. Write the result in one sentence a 14-year-old could repeat. If you can't, you haven't found the finding yet.
Numbers people can picture. "Out of 100 people who did X, 8 got Y, compared with 12 who didn't." Natural frequencies are easier to understand than percentages.
Absolute, not just relative. "Cuts risk by 50%" can mean 2 in 1,000 down to 1 in 1,000. Give both, or give the absolute numbers.
Swap the jargon. Hypertension becomes high blood pressure, and topical becomes on the skin. If you must use the term, explain it right there on the slide.
Say what it doesn't show. "This was in adults over 50, so we don't know yet about younger people." That one line builds more trust than any credential.
Structure for a carousel. Slide 1 hook. Slide 2 the question. Slide 3 what the study did. Slide 4 what it found, in numbers. Slide 5 what it doesn't show. Slide 6 what to do. Slide 7 the source.
Your action step: Rewrite your study's main result as one plain sentence plus one 'out of 100' number.
Want me to make your first post for you, free, in your specialty?
Drop your Instagram handle. One post, on your page's look, so you can judge it on your own page. No card.
Reply POST to any of my course emails with your Instagram handle, or message me on Instagram. Lesson 3 of 4 · 3 min read
Sourcing under the post
A source line turns "a doctor's opinion" into "a doctor showing the evidence". It also protects you.
Where to put it. Last slide of the carousel and the end of the caption. Short form: first author, journal, year. Example: Smith et al., JAMA Dermatology, 2024.
Make it checkable. Put the DOI or PubMed ID in the caption, since links aren't clickable in captions. Patients rarely check, but the ones who do become your biggest fans.
Screenshot with care. A cropped screenshot of the abstract's key line on a slide reads as proof. Keep it small and readable, and never crop it so it says something the paper doesn't.
Match the claim to the source. Every sentence on the slides should be traceable to the source or to standard guidance. If a line isn't, cut it or soften it.
Date it. Evidence changes. If a newer guideline comes out, update or archive the post.
Your action step: Write the source line and the caption reference for your post, with the DOI or PMID.
Want me to make your first post for you, free, in your specialty?
Drop your Instagram handle. One post, on your page's look, so you can judge it on your own page. No card.
Reply POST to any of my course emails with your Instagram handle, or message me on Instagram. Lesson 4 of 4 · 4 min read
Staying compliant (and when to hand it off)
Good medical content is also safe medical content. These rules hold in most places. Your own board or regulator has the final word, so check their advertising guidance once.
No guarantees. Never promise results: no "you'll get rid of", no "guaranteed", no "best doctor in". Say what the evidence shows and that results vary.
Patient privacy. No patient faces, names, stories or images that could identify them without written consent. In the US that's HIPAA; elsewhere similar laws apply. Before-and-after photos need explicit consent and are restricted in some countries.
Education, not advice. Add a line like "General information, not medical advice. Talk to your doctor about your situation." Don't diagnose people in the comments; invite them to book instead.
Disclose relationships. If a brand paid you or gave you a product, say so clearly at the start.
Stay in your lane. Post about what you treat. It's safer, and it's more convincing.
Want this done for you? This is exactly how Haelyx works. Every claim gets a real paper, every post is checked before you see it, and nothing goes live without your OK. Your first post is free, so you can judge the quality yourself.
Your action step: Run your post through the Source-Check Checklist from your email before you publish it.
Want me to make your first post for you, free, in your specialty?
Drop your Instagram handle. One post, on your page's look, so you can judge it on your own page. No card.
Reply POST to any of my course emails with your Instagram handle, or message me on Instagram. Want us to do this for you every week?Haelyx makes your posts and reels from your own words and the research. Nothing goes live without your OK.See the plans →