Your morning coffee stopped working years ago. You adjusted so gradually that you never noticed the moment it stopped giving you something and started preventing you from losing something. That distinction – between a substance that enhances performance and one that prevents withdrawal – is one most professionals never examine.

Caffeine's primary action is adenosine receptor antagonism. Adenosine accumulates during waking hours and produces the sensation of sleep pressure. Caffeine blocks the receptors and the signal is temporarily masked. But with chronic daily exposure the brain upregulates those receptors – it grows more of them and increases their sensitivity. After months of regular use, the morning coffee is not enhancing cognitive performance above the natural baseline. It is restoring performance to approximately where it would have been if you had never started.

Withdrawal is not caused by residual caffeine. It is caused by the adapted brain, now with excess receptors and no stimulant to block them. Days 3 to 5 are cognitively debilitating enough that most people interpret the experience as evidence they cannot function without caffeine, and the cycle resumes. By week 2, receptor downregulation has begun and a different quality of cognition becomes available. Almost nobody reaches week 2.

You can find a companion essay for this podcast episode at dmitrysokolovmd.com.

These drugs work. That needs to be said first, because what follows is going to be nuanced, and nuance is easily mistaken for scepticism. Semaglutide and tirzepatide produce weight loss at a scale no behavioural intervention has ever matched in clinical trials – 15 to 22 percent mean body weight reduction over 68 to 72 weeks.

What is in dispute, and almost never discussed honestly, is what that weight loss is actually made of and what happens when you stop. In the STEP 1 body-composition substudy, roughly 39 percent of the total weight lost was lean body mass. That includes water and glycogen, not only skeletal muscle – but the muscle component is not zero, and in a patient already losing 3 to 8 percent of muscle mass per decade, it matters. The STEP 1 extension trial followed participants for a year after discontinuation: without active lifestyle support, they regained two-thirds of the weight.

Used well, these drugs are a bridge – from a condition where change was impossible to one where change can be maintained. For some patients, the biology requires lifelong therapy, and there is no clinical shame in that. The panacea framing is wrong because it implies the drugs solve the problem alone. The curse framing is wrong because it dismisses genuinely effective pharmacology.

You can find a companion essay for this podcast episode at dmitrysokolovmd.com.

Average life expectancy in the UK has risen from roughly 50 years to over 80 in the last century. That is not a small thing. But the average person will spend their final 8 to 12 years living with significant functional limitation – unable to get off the floor unassisted, unable to carry their own shopping, unable to walk at a pace that would let them cross a road in the time the light gives them.

Medicine has become extraordinarily good at keeping people alive. It has not yet become equivalently good at keeping them capable. And the conversation about longevity, as it exists publicly, is almost always about duration. When people hear the word they hear "more years." The question that actually matters is what those years will contain – and the answer is determined not at 70, but at 40.

Muscle mass declines by 3 to 8 percent per decade after 30. VO₂max declines by roughly 10 percent per decade. Bone mineral density peaks at around 30 and falls thereafter. None of this is disease, and most of it will not show up on a blood test. It is the background rate at which capacity is being spent in every adult who is not actively working against it.

You can find a companion essay for this podcast episode at dmitrysokolovmd.com.