My Oura ring measured an average nighttime HRV of 71 last week. Five months ago, before I started anchoring my fasting window to my sleep schedule, it was 54. My resting heart rate during sleep dropped from 58 to 51. My fasting glucose, measured by continuous monitor, went from averaging 98 mg/dL to 88. I changed one thing: I stopped eating by 7 PM, three hours before my 10 PM bedtime, and extended my overnight fast to roughly 14 hours.
That is n=1. I know what n=1 means. I also know what my bloodwork and wearable data have looked like for half a year, and those numbers are not confused.
You have already seen the Northwestern trial. Thirty-nine people, 7.5 weeks, sleep-aligned fasting, 3.5% nighttime blood pressure dip, 5% nighttime heart rate dip, 90% adherence, no caloric restriction. The sample is small. The study duration is short. The criticism is valid. I am not here to argue that 39 people settled the science.
I am here to argue that waiting for the 10-year trial while ignoring everything we already know about circadian biology is a terrible risk calculation.
The Mechanism Is Not New. The Application Is.
The conversation around the Northwestern study has fixated on its sample size. Fair. But the underlying circadian biology is not sitting on a foundation of 39 people. It is sitting on decades of research and some genuinely large datasets.
The Vujović crossover trial at Brigham and Women's Hospital in 2022 rigorously controlled for calories, sleep, light, and activity, then shifted meals four hours later. The result: late eating increased hunger (p < 0.0001) and altered appetite-regulating hormones, increasing waketime and 24-h ghrelin:leptin ratio. Late eating decreased waketime energy expenditure (p = 0.002) and 24-h core body temperature. Same calories. Different timing. Measurably different metabolic output.
A 2025 twin study published in eBioMedicine found that later eating timing in relation to an individual internal clock is associated with lower insulin sensitivity. The researchers noted that shifting the main calorie intake to earlier circadian times may improve glucose metabolism, but genetic factors could influence the feasibility and effectiveness of eating-timing based interventions.
Then there is the nocturnal blood pressure literature, which is enormous. There is increasing evidence that the mean nocturnal BP level is the most sensitive predictor of cardiovascular morbidity and mortality. Several studies have shown that less nocturnal BP dipping was associated with poor prognosis irrespective of the 24-hour BP levels. We are not debating whether nocturnal dipping matters. It matters. The question is whether meal timing can improve it. Northwestern says yes.
And the NutriNet-Santé cohort of 103,000 adults? Having a later last meal of the day (later than 9PM compared to earlier than 8PM) was associated with a higher risk of cardiovascular outcomes, especially among women. That is not 39 people. That is 699,547 person-years of follow-up.
My Protocol and What It Actually Moved
I have been running a sleep-aligned eating window since late September 2025. The protocol: last meal by 7 PM, lights dimmed by 7:30, sleep by 10. First meal the next day between 9 and 11 AM, giving me a 14 to 16 hour overnight fast. No caloric restriction. I eat the same amount I always have.
What changed: overnight HRV up 31% (54 to 71). Resting sleep heart rate down 12% (58 to 51). Fasting glucose down 10% (98 to 88 mg/dL). Deep sleep percentage up from 17% to 23% as measured by Oura. My most recent lipid panel showed LDL at 98, down from 112 six months prior, though I cannot attribute that entirely to meal timing since I also increased my omega-3 intake.
I am not claiming this proves anything for you. But the convergence of my personal biomarkers with the mechanistic data is what keeps me on this protocol. Research shows that intermittent fasting significantly improves cardiovascular autonomic regulation in healthy adults, as evidenced by enhanced HRV and reduced resting heart rate. My data tracks with that finding. And studies on fasting have shown that increasing heart rate variability (RMSSD) accompanied by a more vascular than myocardial response following a 16 h fast.
The point is not that I am special. The point is that when a zero-risk behavioral intervention aligns with mechanistic evidence from multiple research groups across multiple study designs, I do not need n=10,000 to try it. I need n=10,000 to prove it. Different standard.
The Risk Calculation Nobody Is Doing
Every debate about this study frames it as "is this proven enough to recommend?" That is the wrong question for an individual optimizing their own health. The right question: what is the downside?
Not eating for three hours before bed costs you nothing. There is no supplement to buy, no drug interaction to worry about, no toxicity threshold. The worst case: you are slightly hungry at 8 PM for a week until you adapt. That is it. That is the entire downside.
The potential upside: improved nocturnal blood pressure dipping, better autonomic regulation during sleep, enhanced glucose metabolism the following day, and alignment of your peripheral organ clocks with your central circadian pacemaker. Several of these rhythms peak in the biological morning or early afternoon, implicating earlier in the daytime as optimal for food intake. Disruptions in these rhythms impair metabolism and influence the pathogenesis of metabolic diseases.
Meanwhile, poor cardiometabolic health is common. In 2018, only about 6.8% of U.S. adults met criteria for optimal cardiometabolic status. We are not doing well. The tools we have are not working fast enough. When a free, zero-risk intervention shows this much mechanistic coherence, the rational move is to run the experiment on yourself while the large trials catch up.
Yes, the NHANES analysis showing a 91% increased CVD mortality risk with 8-hour time-restricted eating is concerning. But that finding has serious confounding problems, and critically, the Northwestern protocol is not aggressive fasting. It is a 13 to 16 hour overnight window anchored to sleep. That is a meaningful distinction. The researchers themselves plan to take it to larger multi-center trials, and Grimaldi shared that they "also want to test this specifically in people with hypertension or diabetes, [who] might benefit most."
Dr. Phyllis Zee, the study's corresponding author, put it well: "It's not only how much and what you eat, but also when you eat relative to sleep that is important for the physiological benefits of time-restricted eating."
Protocol of the week: Stop eating three hours before your bedtime. Dim your lights at the same time. Do not change what you eat or how much. Aim for a 13 to 16 hour overnight fast. Track your resting heart rate and HRV using any wearable. Give it six weeks. If your numbers do not move, you lost nothing. If they do, the ROI on this protocol is insane.
Alex will want to see the multi-center replication before he tells anyone to bother. And I respect that. But my HRV has been climbing for five months and I am not waiting for the meta-analysis to keep going.
