THE SIGNAL
|
||||||||||||||||||
|
||||||||||||||||||
|
Week of March 9–14, 2026 · Issue 001 · Andy Feltovich, CISSN · CSCS · StrongFirst Elite |
||||||||||||||||||
|
Each week, The Signal cuts through the noise in health, performance, and science:
The Signal analyzes each topic with the IICE Framework:
|
||||||||||||||||||
| ▶ What to Watch | ||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
| ✕ What to Ignore | ||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
| ↻ What’s “News” but Isn’t New | ||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
| ■ Deep Dive — Issue 001 | ||||||||||||||||||
|
GLP-1 Drugs and Addiction — The Unexpected Signal What a 600,000-person study means for the future of addiction medicine, the economics of drug development, and why the muscle loss story matters more than the weight loss headline. A retrospective cohort study of 606,434 U.S. veterans — the largest observational study of its kind — found that GLP-1 receptor agonists reduced the risk of new substance use disorders by 14–25% across alcohol, opioids, cocaine, and nicotine, and cut drug-related deaths in half among patients with existing addiction.[1] This is not a randomized controlled trial, but at 600,000 participants across multiple substances, the signal is hard to dismiss. Mechanistically, it makes sense: GLP-1 receptors are expressed in the brain’s reward circuitry, and the same dopaminergic dampening that reduces food cravings appears to reduce cravings for addictive substances.
A note on mechanism: GLP-1s are not a substitute for benzodiazepines in alcohol detox or methadone in opioid maintenance. They don’t touch withdrawal physiology. The signal is in what happens after detox: reduced craving, reduced relapse, reduced harm — across multiple substances simultaneously. That cross-substance effect is what makes this genuinely novel. Current addiction drugs (naltrexone, acamprosate) don’t work that way.
The incentives are clear: Novo Nordisk’s U.S. patent on semaglutide expires in 2032. The substance use disorder (SUD) treatment market is $42 billion annually — with only 2% of eligible patients currently prescribed any medication. An FDA approval for addiction wouldn’t just open a new revenue category — it would build a clinical moat.† Generic biosimilars can only copy approved indications. Every new pathology Novo and Lilly can lock in before the patent cliff forces generics to run their own expensive trials before they can compete. Eli Lilly is already running two Phase 3 trials of a dedicated GLP-1 addiction compound. The race is on. The GLP-1 revolution was never just a drug launch. It was a class launch — the first since statins redrew cardiovascular medicine in the 1990s — and unlike statins, it’s a class launch across multiple disease categories: obesity, diabetes, addiction, neurodegeneration. The scope and scale of such conquest makes the side effects — known and unknown — that much more troubling. One of the more well-known, short-term side effects is that approximately 40% of weight loss comes from lean muscle — not fat, as confirmed by multiple randomized controlled trials.[2, 3] The fix is not complicated: lifestyle modification as a first line of defense, strategic use of GLP-1 agonists only when necessary — and structured resistance training and nutrition — including adequate protein — always.
† In investment language, a moat is a durable competitive advantage that protects profits over time — the term originates with Warren Buffett and is formalized in H. Brilliant and J. J. Collins, Why Moats Matter: The Morningstar Approach to Stock Investing (Wiley, 2014). The strategic underpinning is M. E. Porter’s Five Forces framework, which identifies barriers to entry as one of the five forces that determine industry profitability — see M. E. Porter, On Competition (Harvard Business Review Press, Updated ed., 2008). See Further Reading below for both.
|
||||||||||||||||||
|
Sleep Wearables: Good Servant — Poor Master Sleep trackers are useful tools. They are not polysomnography. Here is what they actually measure, what the science says about what they get right, and why knowing the difference matters more than your sleep score. We’re entering the second stage of sleep wearables. The first stage was making scientifically validated metrics commercially feasible. The second stage will be interpretation and wisdom — knowing not just how to use the metrics, but when. But first, a whirlwind recap of how we got here. The Gold Standard You Don’t Have Electronic sleep measurement isn’t new. The discipline was pioneered by Nathaniel Kleitman in 1937 with the use of electroencephalography to record brain activity during sleep.[4] Since then, sleep scientists have progressively added more measuring devices to more body parts to better triangulate and taxonomize the physiological states that are collectively known as “sleep”:
That bundle of laboratory measurements is collectively known as polysomnography (PSG) — the gold standard — which allows a trained practitioner to score every 30-second epoch of your night into one of five stages: Wake, N1, N2, N3 (slow-wave/deep), and REM. William Dement, Kleitman’s student and the originator of that nomenclature, joked that had he known those terms would persist, he would have thought of something sexier. Your Oura Ring, WHOOP strap, or Apple Watch does none of that. The breakthrough that made wearables feasible came in the 1970s with wrist actigraphy — motion detection — followed by skin temperature and photoplethysmography (PPG), the optical heart rate sensor on the back of your ring. Wearables combine those measurements with proprietary algorithms to infer what sleep stage you might be in — not the same league as PSG. As Brandon Marcello puts it: the gap in error is the difference between a DEXA body composition scan and asking someone across the room to guess your body fat percentage.[6] What Do Sleep Wearables Actually Do? The following is adapted from an article by sleep physician and FeltovichFit Podcast guest Dr. Muhammad Usama.[7, 8] For total sleep time — whether you were asleep or awake — consumer wearables perform reasonably well. Oura Ring Gen 3 and Gen 4 show approximately 94% sensitivity for sleep detection. WHOOP is in the same range. The wheels come off with specificity — distinguishing quiet wakefulness from actual sleep. Fitbit Inspire 2 clocked 94% sensitivity but only 13% specificity in one validation, meaning it logged nearly every quiet period as sleep. If you lie awake at 3 AM for 40 minutes and your tracker doesn’t know it, your reported sleep total is fiction. A 2025 meta-analysis of 24 studies found that popular wearables underestimate total sleep time by roughly 17 minutes and sleep efficiency by 5%, while overestimating wake-after-sleep-onset by about 13 minutes compared to PSG. Four-stage scoring (light, deep, REM, wake) is where the wheels fall off. The best-performing consumer devices show 75–80% agreement with PSG. Oura Ring Gen 3 showed stage-scoring accuracy between 75–91% depending on the stage — the best peer-reviewed numbers in the consumer space. WHOOP showed approximately 64% overall agreement with PSG on four-stage scoring. The Apple Watch Series 8 could distinguish sleep versus wake adequately but fared poorly on estimating sleep stages.
The Fatigue Science ReadiBand — a technology representing $37 million in DoD-funded research and 25 years of development — achieves approximately 93% agreement with PSG, making it the most validated device in the consumer-adjacent market. However, it’s expensive: $649 for the device and a one-year subscription as of this writing, plus $599 for a 12-month renewal (to which I respectfully declined) — and the app isn’t as good as Oura’s. Orthosomnia: The Word You Need to Know Knowledge is knowing a tomato is a fruit. Wisdom is knowing not to put it in a fruit salad. You can know your deep sleep percentage down to a decimal point and make yourself worse. Clinicians have a name for this: orthosomnia — a preoccupation with optimizing tracker data to the point where the monitoring itself disrupts sleep. The patient extends time in bed to chase a higher efficiency score, which — counterproductively — typically lowers efficiency. The anxiety about the score impairs sleep onset. The vigilance required to monitor your own wakefulness is neurologically incompatible with falling asleep. If you are the kind of person who optimizes everything, you are at elevated risk for this. Wearables are an adjunct to sleep hygiene — not a substitute or scorecard for it. Three Questions That Don’t Require a Wearable Before you interpret a single score from any device, answer these honestly:
If you answer yes to all three, your sleep is functioning. The wearable can add texture. If you answered no to any of them, no device will fix that. And the device may, in fact, distract you from fixing it.
The IICE read: Wearables are converging to the theoretical limit of what can be achieved without full PSG in a sleep lab under a trained practitioner. Fatigue Science is the “best” and offers customization — useful if, for example, you own a heavy equipment company and need to ensure operators are rested for safety and liability purposes. Otherwise, Oura is sufficient and comes with a better price and app. Assess the score in conjunction with how you’re feeling and behaving, and understand that wearables can only approximate sleep stages. If a wearable says you’re sleeping well but you wake up exhausted, believe your body. The wearable watches you sleep. It doesn’t sleep for you. Last, it bears repeating: do not try to diagnose or treat obstructive sleep apnea (OSA) or any other sleep disorder with a wearable.
|
||||||||||||||||||
| ■ References · Further Reading · Additional Resources | ||||||||||||||||||
References — Issue 001 | ||||||||||||||||||
|
[1]
M. Cai, T. Choi, Y. Xie, and Z. Al-Aly, “Glucagon-like peptide-1 receptor agonists and risk of substance use disorders among US veterans with type 2 diabetes: cohort study,” BMJ, vol. 392, e086886, Mar. 2026. https://doi.org/10.1136/bmj-2025-086886
[2]
D. Pantazopoulos, E. Gouveri, D. Papazoglou, and N. Papanas, “GLP-1 receptor agonists and sarcopenia: Weight loss at a cost? A brief narrative review,” Diabetes Res. Clin. Pract., vol. 229, p. 112924, Nov. 2025. https://doi.org/10.1016/j.diabres.2025.112924
[3]
J. P. H. Wilding et al., “Once-Weekly Semaglutide in Adults with Overweight or Obesity,” N. Engl. J. Med., vol. 384, no. 11, pp. 989–1002, Mar. 2021. https://doi.org/10.1056/NEJMoa2032183
[4]
M. H. Kryger, T. Roth, and C. A. Goldstein, Kryger’s Principles and Practice of Sleep Medicine, 7th ed. Philadelphia: Elsevier, 2022. Hardcover Kindle
[5]
W. C. Dement, “Knocking on Kleitman’s Door: The View from 50 Years Later,” Sleep Medicine Reviews, vol. 7, no. 4, pp. 289–292, 2003. https://doi.org/10.1053/smrv.2003.0279
[6]
B. Marcello, “Sleep – The Only True ‘Fix-All’ for Health and Performance,” NSCA, May 15, 2020. [Online]. https://www.youtube.com/watch?v=WKt2LoHaCp4
[7]
A. Feltovich, “Sleep as a Superpower: Optimizing Rest, Wakefulness, and Longevity with Dr. Muhammad Usama,” FeltovichFit Podcast, 2025. Spotify
[8]
M. Usama, “Can Wearables Truly Transform Sleep Health, or Are We Just Watching Ourselves Sleep?” LinkedIn Pulse, Jun. 18, 2025. LinkedIn
|
||||||||||||||||||
Further Reading | ||||||||||||||||||
|
►
M. Walker, Why We Sleep, Scribner, 2018.
Definitive popular synthesis of sleep science. Note: some effect sizes have been criticized as overstated — read alongside the peer-reviewed literature.
Kindle
Audible
Hardcover
Paperback
►
H. Brilliant and J. J. Collins, Why Moats Matter: The Morningstar Approach to Stock Investing, Wiley, 2014.
Framework for the moat concept referenced in the GLP-1 article. Translates directly to pharmaceutical patent strategy and the economics of indication expansion.
Kindle
Audible
Hardcover
Paperback
►
M. E. Porter, On Competition, Harvard Business Review Press, Updated ed., 2008.
Porter’s Five Forces framework underlying the GLP-1 moat discussion. The barriers-to-entry and rivalry chapters apply directly to pharma patent strategy and indication expansion economics.
Kindle
Hardcover
►
S. W. Porges, The Polyvagal Theory: Neurophysiological Foundations of Emotions, Attachment, Communication, and Self-Regulation, W. W. Norton & Company, 2011.
The seminal academic source. Porges’ original theoretical framework — read this before citing polyvagal theory in any context. Understand what it actually claims vs. the consumer translation.
Kindle
Audible
Hardcover
Paperback
|
||||||||||||||||||
Additional Resources | ||||||||||||||||||
|
■
Fatigue Science — Manufacturer of the ReadiBand (about) | Purchase ReadiBand
■
PubMed / NCBI — Primary academic literature search. Start here before citing any study.
■
FeltovichFit Podcast — Spotify
|
||||||||||||||||||
|
||||||||||||||||||
Collegium of Order & Flow Frameworks for a Fallen World
The Collegium exists because no single organization:
“In the age of AI, siloed thinkers are obsolete. ΠΑΝΤΑ ΡΕΙ — Everything Flows
THE SIGNAL is produced by Andy Feltovich — CISSN · CSCS · StrongFirst Elite Manage subscription · Unsubscribe · Created with Perplexity Computer |
||||||||||||||||||