Dispense As Written

Gregory Castelli
  • 0.0 (0)
  • Medicine
  • Updated Weekly

Clinical trial breakdowns and evidence-based medicine for busy clinicians. I am an Associate Professor and Clinical Pharmacist. I review and translate the latest medical research into practical takeaways you can use. Every week, I cover new studies, examine controversial evidence, and answer your questions. No industry influence. No clickbait conclusions. Just honest analysis. WHAT YOU'LL FIND HERE: → Weekly clinical trial breakdowns → Evidence-based medicine concepts explained → Viewer Q&A on practice controversies PERFECT FOR: Physicians | Pharmacists | and All Medical Professions

  1. 5d ago

    Tirzepatide Is No Worse Than Dulaglutide. But What If Dulaglutide Wasn't Good to Begin With?

    SURPASS-CVOT made headlines for proving tirzepatide is non-inferior to dulaglutide for cardiovascular outcomes. But before you let that result drive your clinical decisions, there's a question nobody asked: what exactly was dulaglutide's CV evidence built on? The REWIND trial — the landmark Lancet 2019 RCT that put dulaglutide on the CV map — is the foundation. And when you look closely, the foundation is thinner than the guidelines suggest.9,901 patients, a 5.4-year follow-up, a MACE composite that barely cleared statistical significance after an interim alpha adjustment — and a North American subgroup that trended the wrong direction. The composite lived and died on a single endpoint: non-fatal stroke. CV death, MI, and all-cause mortality were all non-significant. REWIND earns a Castelli Coefficient of 5.5 / 10. Technically positive. Practically thin. And now you have the receipts.KEY FINDINGS• Primary composite MACE outcome: 12.0% dulaglutide vs. 13.4% placebo — HR 0.88 (95% CI 0.79–0.99); p=0.026 (adjusted threshold 0.0467 after interim alpha spend)• Absolute risk reduction: 1.4% over 5.4 years — NNT ~60• Non-fatal stroke drove the result: HR 0.76, p=0.017 — CV death (NS), MI (NS), all-cause mortality (NS)• USA + Canada subgroup: HR 1.14 (95% CI 0.89–1.47) — trending in the wrong direction in ~21% of the trial population• Only 31.5% had established cardiovascular disease — most of the trial was primary prevention• GI adverse events: 47.4% dulaglutide vs. 34.1% placeboLINKS & RESOURCESTrial citation: Gerstein HC et al. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet 2019; 394: 121–30. https://doi.org/10.1016/S0140-6736(19)31149-3SURPASS-CVOT episode: https://youtu.be/CLCpeJXUKNQ?si=biScjU7oXRQtSGezDispense As Written (DAW) is an independent, unsponsored evidence-based medicine channel hosted by Gregory Castelli, PharmD, FCCP, BCPS, BC-ADM, CDCES — Clinical Pharmacist and Associate Professor. Produced by Dispense As Written, LLC.DAW breaks down clinical trials, guidelines, and pharmacology with the same critical appraisal framework used to train clinicians and residents — without pharmaceutical sponsorships, without the fluff.📧 dispensingevidence@gmail.com📱 Instagram / TikTok: @gregcastellipharmd🐦 X / Twitter: @gregcastellirx💼 LinkedIn: gregcastellipharmdThe content produced by Dispense As Written, LLC is intended for educational and informational purposes only and does not constitute medical advice. The views expressed in this video are those of Dispense As Written, LLC and do not represent the views of any employer, academic institution, or affiliated organization. This content should not be used as a substitute for professional medical judgment. Always refer to primary literature, institutional guidelines, and clinical expertise when making patient care decisions. Dispense As Written, LLC has no financial relationships with pharmaceutical manufacturers. No content on this channel has been sponsored, reviewed, or approved by any drug manufacturer or commercial entity.#evidencebasedmedicine #dispenseaswritten #primarycare #familymedicine #internalmedicine #meded #clinicalpharmacy

  2. Aug 10

    Tirzepatide Has Better Metabolics. So Why Doesn't It Win?

    Your patient has type 2 diabetes, established cardiovascular disease, and has been on tirzepatide for two years. Their A1c is down, their weight is down, and their triglycerides are better than they've been in a decade. The question every clinician should be asking: is any of that metabolic improvement actuallytranslating into fewer heart attacks and strokes? SURPASS-CVOT is the first randomized cardiovascular outcomes trial for tirzepatide — and the answer is more complicated than the headlines suggest. In this episode we break down SURPASS-CVOT, a double-blind noninferiority trial comparing tirzepatide to dulaglutide in over 13,000 patients with type 2 diabetes and atherosclerotic cardiovascular disease. Tirzepatide was noninferior— meaning no worse than — dulaglutide on 3-point MACE (CV death, MI, stroke), but did not achieve superiority. Despite dramatically better A1c reduction, weight loss, and triglyceride lowering, the primary cardiovascular endpoint was essentially a tie. All-cause mortality did tell a different story, and that'swhere this paper gets interesting. We score this one a Castelli Coefficient of 7.5 / 10 — a well-designed trial that may not have taught us as much as we hoped. Key Findings ●     Primary endpoint (3-pointMACE): HR 0.92 (95.3% CI 0.83–1.01) —noninferior (p=0.003), not superior (p=0.09) ●     All-cause mortality: HR 0.84 (95% CI 0.75–0.94) — tirzepatide group 8.6% vs.dulaglutide 10.2% ★exploratory ●     MACE-4 (+revascularization): HR 0.88 (95% CI0.81–0.96) — statistically significant ●     Metabolic advantage —real, but didn't move the MACE needle: A1c−1.66% vs. −0.88%; weight −11.6% vs. −4.8%; triglycerides −24.2% vs. −10.2% ●     GI adverse events: 42.5% with tirzepatide vs. 35.9% with dulaglutide — morenausea and diarrhea ●     Generalizability caveat: Only 14.7% of patients enrolled in North America — the NAsubgroup HR was 1.01, the weakest of all regions   Links & Resources ●     Full paper: Nicholls SJ, et al. Cardiovascular Outcomes withTirzepatide versus Dulaglutide in Type 2 Diabetes. N Engl J Med2025;393:2409–20. https://doi.org/10.1056/NEJMoa2505928 ●     REWIND trial (dulaglutideCV outcomes vs. placebo): Gerstein HC, etal. Lancet 2019;394:121–30 ●     EBM Primers —Noninferiority Trials Explained: Watch on@dispenseaswritten [https://youtu.be/SpeVLfGrvFk?si=dzK-6OHunFv5eT8J]   About GregoryCastelli, PharmD, FCCP, BCPS, BC-ADM, CDCES Clinical Pharmacist | AssociateProfessor Produced by Dispense As Written, LLC   📺YouTube: @dispenseaswritten 📸 Instagram/TikTok: @gregcastellipharmd 🐦X/Twitter: @gregcastellirx 💼 LinkedIn: gregcastellipharmd 📧dispensingevidence@gmail.com   Legal Disclaimer ⚠ LEGAL DISCLAIMER — DO NOT MODIFY The content in this video is produced by Dispense As Written, LLC for educational purposes only and does not constitute medical advice. The views expressed are those of Dispense As Written, LLC and do not represent the views of any affiliated academic institution, employer, or health system. This content is not intended to diagnose, treat, cure, or prevent any disease. Always consult a qualified healthcare professional for clinical decision-making. Dispense As Written, LLC has no pharmaceutical sponsorships and maintains full editorial independence.

  3. Aug 3

    Psilocybin: Hype or Signal?

    A single 25 mg dose of psilocybin. One supervised session. And then — for the next six weeks — a clinically substantial reduction in depression scores, with no pill to refill and no daily adherence to manage. That's the claim from a Phase 2 RCT published in JAMA, and the numbers are hard to dismiss. The question isn't whether the signal is real. The question is whether you can trust it. This video breaks down the trial design, the primary outcome data, the safety profile, and the methodological problems that keep this from landing as a practice-changer. Centralized blinded raters, an active niacin comparator, and a 6-week follow-up make this the best-designed psilocybin trial to date. Sponsor-controlled data management, near-certain blinding failure, a 9-to-1 dropout differential, and a sustained remission rate that didn't reach significance in the primary analysis are the reasons the Castelli Coefficient lands at 7 — not higher.📊 KEY FINDINGS• MADRS reduction from baseline to day 43: −19.1 pts (psilocybin) vs −6.8 pts (niacin) — between-group difference −12.3 pts (95% CI −17.5 to −7.2)• Effect appeared by day 8 (−12.0 pt difference) and held flat through day 43 — no attenuation across any timepoint• Sustained response (≥50% reduction at all 4 post-dose visits): 42% psilocybin vs 11% niacin (OR 5.6; 95% CI 1.9–16.7)• Sheehan Disability Scale at day 43: −2.31 mean difference (95% CI −3.50 to −1.11) — real-world functional improvement in work, social, and home domains• Sustained remission (MADRS ≤10 at all 4 visits): 25% vs 9% — did not reach statistical significance in the primary ITT analysis (p = .05)• Severe adverse events: 8% psilocybin vs 0% niacin; no serious treatment-emergent adverse events; no suicidal behavior in either arm🔗 LINKS & RESOURCESFull paper: Raison CL, et al. Single-Dose Psilocybin Treatment for Major Depressive Disorder: A Randomized Clinical Trial. JAMA. 2023;330(9):843–853. DOI: 10.1001/jama.2023.14530⚕️ ABOUT DISPENSE AS WRITTENEvidence-based medicine without the fluff. Every video ends with the Castelli Coefficient — a verdict that incorporates study design, statistical validity, and clinical applicability. No pharmaceutical sponsorships. No conflicts of interest. Just the data. Gregory Castelli, PharmD, FCCP, BCPS, BC-ADM, CDCESAssociate ProfessorDispense As Written, LLC 📱 FIND USYouTube: @dispenseaswrittenInstagram / TikTok: @gregcastellipharmdX / Twitter: @gregcastellirxEmail: dispensingevidence@gmail.com⚠️ LEGAL DISCLAIMERThe content provided in this video is for educational and informational purposes only. It is intended for healthcare professionals and medical students as a supplement to — not a substitute for — clinical judgment, professional training, and the individualized care of patients. Nothing in this video constitutes medical advice, and it should not be used as the basis for any clinical decision regarding an individual patient. Always consult current clinical guidelines, institutional protocols, and your own clinical judgment when making treatment decisions. Gregory Castelli, PharmD, is speaking in his personal capacity as an educator. The views expressed do not represent the official positions of any affiliated institution. This channel has no affiliation with, and has received no funding or compensation from, any pharmaceutical manufacturer, device company, insurance company, or other commercial entity with a financial interest in the content discussed. Use of trade names is for identification purposes only and does not constitute endorsement. Produced by Dispense As Written, LLC.

  4. Jul 27

    Does Intermittent Fasting Work? Cochrane Finally Has Answers | 2026

    Intermittent fasting has been one of the most hyped dietary interventions of the past decade — promoted on podcasts, social media, and wellness blogs as a metabolic cure-all. But what does the highest-quality evidence actually say? In this episode, Gregory Castelli, PharmD, FCCP, BCPS, BC-ADM, CDCES, breaks down a 2026 Cochrane systematic review and meta-analysis that analyzed 22 randomized controlled trials and 1,995 participants to answer that question once and for all.Against regular dietary advice, intermittent fasting produced a mean weight loss difference of just −0.33% (95% CI −0.92 to +0.26) — a result that earned low-certainty evidence from Cochrane. Against no intervention, IF did produce more weight loss (MD −3.42%), but fell short of the 5% threshold considered clinically meaningful. No study in the review reported diabetes outcomes, patient satisfaction, or mortality. The Castelli Coefficient for this review: 6 out of 10. The methodology was rigorous — the evidence base wasn't.• 22 RCTs, 1,995 participants — outpatient settings across North America, Australia, Asia, and Europe (2016–2024)• vs. Regular Dietary Advice: MD −0.33% weight loss (95% CI −0.92 to +0.26); RR 0.98 for ≥5% reduction — LOW certainty• vs. No Intervention: MD −3.42% weight loss (95% CI −4.95 to −1.90) — MODERATE certainty, but below 5% MCID threshold• Quality of life: SMD +0.11 vs. dietary advice (95% CI −0.27 to +0.49) — LOW certainty; no meaningful difference• Adverse events: RR 1.45 vs. dietary advice (I² = 78%) — VERY LOW certainty• Zero studies reported participant satisfaction, diabetes status, or overall comorbidity outcomesGaregnani LI, Oltra G, Ivaldi D, Burgos MA, Andrenacci PJ, Rico S, Boyd M, Radler D, Escobar Liquitay CM, Madrid E. Intermittent fasting for adults with overweight or obesity. Cochrane Database of Systematic Reviews 2026, Issue 2. Art. No.: CD015610. DOI: 10.1002/14651858.CD015610.pub2Read the full review:https://doi.org/10.1002/14651858.CD015610.pub2Gregory Castelli, PharmD, FCCP, BCPS, BC-ADM, CDCES is a clinical pharmacist and Associate Professor with an extensive background in evidence-based medicine teaching at the residency and graduate level.Dispense As Written (DAW) — "Evidence-Based Medicine Without the Fluff." Every episode breaks down clinical trials, teaches EBM methodology, and gives you a defensible, patient-oriented bottom line. No pharmaceutical sponsorships. No substitutions for profit.Produced by Dispense As Written, LLCYouTube: @dispenseaswrittenInstagram / TikTok: @gregcastellipharmdX / Twitter: @gregcastellirxEmail: dispensingevidence@gmail.comThe content in this video is produced by Dispense As Written, LLC and is intended for educational purposes only. It does not constitute medical advice and should not be used as a substitute for the advice of a qualified healthcare professional. The views expressed are those of the host and do not represent those of any institution or employer.#DispenseAsWritten#EvidenceBasedMedicine #MedEd #primarycare #familymedicine #internalmedicine

  5. Jul 6

    Guidelines Can't See Your Patient

    Four episodes. One guideline. One honest answer to the question nobody asks at the end of a guideline review: what do we actually do now? This finale recaps the four key takeaways from the 2026 ACC/AHA Dyslipidemia series, addresses the non-statin evidence gap, and closes with the argument that matters most — the one about the patient sitting across from you. 📊 Four Takeaways: • COR 1 + LOE C-EO = a vote, not a trial. They carry the same label. • The LDL targets (70, 55) were inferred from achieved drug-trial levels — never prospectively tested as targets • Universal Lp(a) screening is COR 1 for a marker with no approved specific therapy and an outcomes trial still pending • The risk calculator was replaced with no outcomes evidence — 17M reclassified, 4.1M de-prescribed, 15.8M eligible patients still untreated 💊 Non-Statins in Brief: • Bempedoic acid (CLEAR Outcomes): strongest non-statin outcomes case — modest ARR in statin-intolerant patients • Inclisiran: COR 2a in the guideline; primary endpoints are LDL surrogates only • Evinacumab: COR 2b for homozygous familial hypercholesterolemia — affects 1 in 300,000 🏆 Series Final Castelli Coefficient: 6/10 "Read it. Use it. Know which parts have a foundation and which parts have a vote." 📺 Full Series: Ep 4 — 2026 Guideline Overview Ep 5 — LDL Targets: What the Evidence Actually Shows Ep 6 — Screen Everyone for Lp(a)? What the Guideline Doesn't Tell You Ep 7 — 17 Million Americans Lost Their Statin Prescription. Nobody Ran a Trial First. 📚 Series Sources: Blumenthal RS et al. 2026 ACC/AHA Dyslipidemia Guideline. JACC/Circulation. March 13, 2026. DOI: 10.1016/j.jacc.2025.11.016 Cannon CP et al. NEJM 2015;372:2387 (IMPROVE-IT) Sabatine MS et al. NEJM 2017;376:1713 (FOURIER) Schwartz GG et al. NEJM 2018;379:2097 (ODYSSEY OUTCOMES) Ray KK et al. NEJM 2020;382:1507 (ORION) Lee YJ et al. NEJM 2026 (Ez-PAVE). DOI: 10.1056/NEJMoa2600283 Ridker PM et al. NEJM 2024;391(22):2087 (Women's Health Study) Anderson TS et al. JAMA Intern Med. 2024;184(8):963 (PREVENT) Nissen SE et al. NEJM 2023;388:1353 (CLEAR Outcomes — bempedoic acid) 👤 Gregory Castelli, PharmD, FCCP, BCPS, BC-ADM, CDCES Associate Professor | Director of Academic and Clinical Pharmacy Dispense As Written, LLC 🎙️ YouTube: @dispenseaswritten 📸 Instagram/TikTok: @gregcastellipharmd 🐦 X/Twitter: @gregcastellirx 📧 dispensingevidence@gmail.com ⚖️ LEGAL DISCLAIMER: This content is produced by Dispense As Written, LLC and is intended for educational purposes for healthcare professionals only. It does not constitute medical advice and should not be used to guide individual patient care decisions. The information presented reflects the evidence available at the time of filming and may not incorporate subsequent publications or guideline updates. Always consult current clinical guidelines, peer-reviewed literature, and your own professional judgment when making clinical decisions. Dispense As Written, LLC operates independently of Gregory Castelli's academic and institutional roles. No pharmaceutical sponsorship, funding, or promotional consideration was received in the production of this content. #DispenseAsWritten #familymedicine #primarycare #EBM #MedEd

  6. Jun 29

    Same Patient. Same Lab. Different Calculator. Different Prescription

    The new 2026 cholesterol guideline just replaced the national heart risk calculator — and 17 million Americans no longer qualify for a statin because of it. 4.1 million of them are currently taking one. Here's what makes this worth examining closely: ↳ The new PREVENT calculator cuts mean 10-year ASCVD risk nearly in half nationally (8.0% → 4.3%) ↳ Black adults saw the largest drop: 10.9% → 5.1% ↳ Adults aged 70–75: 22.8% → 10.2% ↳ No randomized trial has established that PREVENT produces better cardiovascular outcomes than the old PCE ↳ The derivation dataset (Optum Data Warehouse) is proprietary — not publicly available for independent audit ↳ And while we debate calculators, 15.8 million already-eligible Americans aren't taking the statin they qualify for The calculator isn't the bottleneck. Access, cost, adherence, and follow-through are. In the latest episode of Dispense As Written, I break down what actually changed, what we don't know, and what the most important number in this paper really is — and it has nothing to do with which equation you use. Castelli Coefficient: 6/10. The therapy earns the score. The tool doesn't. 🎥 Full episode linked in comments. — Dispense As Written is an independent, unsponsored evidence-based medicine education platform. No pharmaceutical funding. No substitutions. ⚖️ LEGAL DISCLAIMER: This content is produced by Dispense As Written, LLC and is intended for educational purposes for healthcare professionals only. It does not constitute medical advice and should not be used to guide individual patient care decisions. Dispense As Written, LLC operates independently of Gregory Castelli's academic and institutional roles. No pharmaceutical sponsorship, funding, or promotional consideration was received in the production of this content. #EvidenceBasedMedicine #MedEd #PrimaryCare #FamilyMedicine #Dyslipidemia #DispenseAsWritten

  7. Jun 15

    Universal screening for a condition we can’t target specifically? 🤔

    The new 2026 ACC/AHA Dyslipidemia Guideline issued a Class 1 Recommendation for every adult in the US to get their Lipoprotein(a) measured once in a lifetime. But here is the catch: if you find it elevated, the guideline’s core recommendation is simply to optimize modifiable risk factors—the exact same care we should already be providing. With no completed outcomes trials yet proving that lowering Lp(a) reduces major cardiovascular events, this leaves primary care clinicians in a fascinating spot. Is this universal screening push brilliant proactive medicine, or is it market priming for an incoming class of therapeutics? In Part 3 of our critical appraisal series, we look at 30 years of data from the landmark Women's Health Study to separate the medicine from the marketing. 📺 Watch the full appraisal here: www.youtube.com/@DispenseAsWritten #EvidenceBasedMedicine #PrimaryCare #InternalMedicine #MedEd #Cardiology #LpaScreening

  8. Jun 8

    Are the New 2026 LDL Targets Backed by Evidence?

    The 2026 ACC/AHA Dyslipidemia Guideline sets LDL targets of less than 70 mg/dL for high-risk patients and less than 55 mg/dL for very high-risk patients. Those numbers will drive prescribing decisions for the next decade.Has anyone ever run a randomized trial where one group was treated to a target of 70 and another to 55, and compared what happened? When the guideline was published on March 13, 2026 — no. Two weeks later, a trial called Ez-PAVE changed that. In this episode, we go trial by trial through IMPROVE-IT, FOURIER, ODYSSEY OUTCOMES, and ORION — examining what each one actually showed — and what the guideline committee used to justify LDL thresholds that were never prospectively tested as targets. Key Findings: IMPROVE-IT: ARR 2.0%, HR 0.936 (CI 0.89–0.99); all-cause mortality HR 0.99; CV mortality HR 1.00; background statin was simvastatin 40mg — moderate intensity FOURIER: HR 0.85, no mortality benefit (CV death HR 1.05, all-cause HR 1.04); median follow-up 2.2 years; NNT 74 over 2 years at ~$14,000/year ODYSSEY OUTCOMES: All-cause mortality HR 0.85 is nominal — hierarchical testing stopped before it was formally tested; benefit concentrated in LDL ≥100 post-hoc subgroup; NNT 163 for mortality in full population ORION: Primary endpoints are LDL surrogates only — no cardiovascular outcomes data; received COR 2a in the guideline anyway Ez-PAVE (NEJM, March 28, 2026): First head-to-head target trial — open-label, 17 sites in South Korea, no mortality signal, published 15 days after the guideline Castelli Coefficient: 5.5/10 Sources:Cannon CP et al. N Engl J Med. 2015;372:2387. DOI: 10.1056/NEJMoa1410489Sabatine MS et al. N Engl J Med. 2017;376:1713. DOI: 10.1056/NEJMoa1615664Schwartz GG et al. N Engl J Med. 2018;379:2097. DOI: 10.1056/NEJMoa1801174Ray KK et al. N Engl J Med. 2020;382:1507. DOI: 10.1056/NEJMoa1912387Lee YJ et al. N Engl J Med. 2026. DOI: 10.1056/NEJMoa2600283 (Ez-PAVE)Blumenthal RS et al. 2026 ACC/AHA Dyslipidemia Guideline. March 13, 2026. Gregory Castelli, PharmD, FCCP, BCPS, BC-ADM, CDCESAssociate Professor | Director of Academic and Clinical PharmacyDispense As Written, LLC🎙️ YouTube: @dispenseaswritten📸 Instagram/TikTok: @gregcastellipharmd🐦 X/Twitter: @gregcastellirx📧 dispensingevidence@gmail.com LEGAL DISCLAIMER: This content is produced by Dispense As Written, LLC and is intended for educational purposes for healthcare professionals. It does not constitute medical advice and should not be used to guide individual patient care. Always consult current clinical guidelines and use professional judgment in clinical decision-making. Dispense As Written, LLC operates independently of Gregory Castelli's academic and institutional roles.#DispenseAsWritten #EvidenceBasedMedicine #EBM #MedEd #PrimaryCare #FamilyMedicine #InternalMedicine #Dyslipidemia #ldlcholesterol

See All (13)

About

Clinical trial breakdowns and evidence-based medicine for busy clinicians. I am an Associate Professor and Clinical Pharmacist. I review and translate the latest medical research into practical takeaways you can use. Every week, I cover new studies, examine controversial evidence, and answer your questions. No industry influence. No clickbait conclusions. Just honest analysis. WHAT YOU'LL FIND HERE: → Weekly clinical trial breakdowns → Evidence-based medicine concepts explained → Viewer Q&A on practice controversies PERFECT FOR: Physicians | Pharmacists | and All Medical Professions

Information

  • Creator
    Gregory Castelli
  • Years Active
    2026
  • Episodes
    13
  • Rating
    Clean
  • Copyright
    © Gregory Castelli
  • Show Website
    Dispense As Written