# Stress and energy: the practical brief *A short, honest translation of the evidence review. Not medical advice — see the last section for when that matters.* --- ## Start by ruling out the three things that are more common than "stress" Before attributing low energy to stress, three explanations are both more likely and more fixable. Each is routinely missed, and each has a specific reason why. **1. Chronic sleep restriction.** This is the big one. In the definitive study, people restricted to six hours a night for two weeks developed deficits equivalent to *two nights of total sleep deprivation* — and their subjective sleepiness ratings could not distinguish six hours from four. The feeling saturates while the impairment keeps accumulating. You cannot introspect your way to knowing whether you are sleep-deprived. > **What to do:** count actual sleep — time asleep, not time in bed — for two weeks. If it is under seven hours on most nights, that is a sufficient explanation for the fatigue and should be dealt with before anything else. **2. Depression.** Among people presenting to primary care with tiredness, **18.5%** turn out to be depressed. And exhaustion and depression correlate at r ≈ .80 once you correct for measurement error — they may not be separable constructs at all. "Burnout" is the more comfortable label; it is also the less useful one, because it points at nothing treatable. **3. A sleep disorder, usually apnoea.** Cheap to screen for, transformative to treat, and invisible to the person who has it. The red flags are snoring, witnessed pauses in breathing, daytime sleepiness disproportionate to hours slept, and waking unrefreshed. **One thing you can mostly stop worrying about:** in studies with control groups, serious physical disease was *no more common* in people presenting with fatigue than in people who weren't tired at all. A basic blood panel is cheap and reasonable. Extensive investigation without specific clinical findings has a low yield. --- ## The reframe that changes what to do The folk model says energy is a tank that stress drains. Two large preregistered studies across 23 and 36 laboratories killed that idea: willpower depletion produced effects of d = 0.04 and d = 0.06 — indistinguishable from nothing. What the evidence supports instead is that **stress raises the price of effort.** Effort is a cost your brain computes and weighs against expected reward. A day of demanding cognitive work measurably accumulates glutamate in the lateral prefrontal cortex and shifts your choices toward low-cost, short-delay options. The machinery still works. It just costs more to run. This matters because the two models recommend different things: | | If the tank is empty | If the price has gone up | |---|---|---| | **What helps** | Rest | Change what makes effort expensive — remove competing demands, restore genuine detachment, fix the low-control or low-reward features of the situation, protect the hardest work for the cheapest hours | | **What rest does** | Refills it | Lowers the price temporarily — which is exactly why holidays work and then fade within a month | It also removes the moral judgement. "I can't face it" is a price signal, not a character flaw, and "push through" is advice built on a model that failed replication. --- ## What actually works, in order **1. Fix the sleep problem.** CBT-I is the only intervention in the entire review with a *strong* recommendation from a guideline body, with an effect size around g = 0.98 on insomnia severity. Two caveats worth knowing: it improves sleep *quality* far more than sleep *duration* (g = 0.16 for total sleep time), so if the problem is hours rather than quality, it is the wrong tool. And **sleep hygiene alone is explicitly recommended against** — the active ingredients are stimulus control and sleep restriction, which are demanding and initially make you sleepier. The easy component is the one that gets promoted. **2. Move, three to five times a week.** Effect sizes of −0.42 to −0.62 for depression against *active* controls — meaning against something rather than against nothing. Either resistance or aerobic works. Two useful details: the largest returns come from the first increment above doing nothing, and benefit appears below public-health guideline volumes. One honest caveat: for *fatigue* specifically, placebo-controlled trials of chronic exercise found no effect. Exercise for mood on strong evidence, for energy on weaker evidence. **3. Get depression screened and treated if it's there.** Highest base rate of any treatable cause, strongest treatment evidence in the review. **4. Behavioural activation.** SMD 0.67, and it still works self-guided (0.36) — the highest evidence-to-effort ratio available. It is also the intervention that best fits the effort-price model: you schedule valued and rewarding activity on a plan rather than on how you feel, which changes the cost–benefit structure rather than trying to top up a tank. **5. Protect detachment — structurally, not aspirationally.** Psychological detachment from work is the best-supported recovery construct (r ≈ −0.42 with fatigue). But here is the finding that governs everything else: > **The recovery paradox.** Job demands predict *lower* detachment, not higher. The people who most need recovery are least able to initiate it — because activation keeps work thoughts accessible, because depletion makes effortful recovery harder to start, and because connectivity removes the boundary. This is why "you just need to look after yourself better" fails as advice. Recovery has to be a structural feature — boundaries that hold on a bad day without requiring discipline — or it will not happen when it is most needed. **And one distinction worth internalising:** detachment reduces fatigue (ρ = −0.39) but barely produces *vigour* (ρ = +0.14). What generates vigour is **control** (0.31) and **mastery** (0.29). Collapsing on the sofa and learning something hard are not interchangeable. You need both, for different jobs. --- ## Worth doing, modest returns - **Caffeine cut-off ~9 hours before bed.** Recovers ~45 minutes of sleep. (The popular "six hours" rule traces to a 12-person manufacturer-funded study whose result was non-significant.) - **Less alcohol.** Costs ~2.8% of your REM and delays REM onset ~18 minutes, while leaving total sleep time unchanged — which is precisely why nobody notices. - **Slow-paced breathing**, about six breaths a minute, 10–20 minutes. Effect g ≈ −0.35. Cheap, low-risk, small. - **Real breaks, and longer ones.** They improve vigour (d = 0.36). They do *not* measurably improve performance, and longer breaks did better than shorter — the opposite of the popular claim. Take them for recovery, not output. ## Don't bother - **Sleep hygiene as your whole plan** — recommended against. - **Ashwagandha and rhodiola.** The ashwagandha literature contains *no negative trials at all* (Egger's p < 0.001) and it carries a real liver-injury signal. The confirmatory rhodiola RCT found placebo was *better*. - **Supplements if you're not deficient.** Vitamin D, omega-3 and B vitamins are null in large trials; the omega-3 arm of one megatrial found a small *increase* in depression risk. - **Resilience training and wellness apps.** A 32,974-person randomised trial found 2 of 80 outcomes significant, both self-reported. Apps show g = 0.09 against a placebo app, with median real-world 30-day retention of 3.3%. - **Waiting for a holiday to fix it.** The effect is real (d = 0.43) and gone within a month. --- ## When to involve a clinician Persistent unexplained fatigue, or functioning that is substantially impaired, warrants a proper assessment. A competent one covers full blood count, thyroid function, HbA1c or glucose, ferritin, renal and liver function, coeliac serology where indicated, screening for obstructive sleep apnoea, and structured screening for depression and anxiety. If you are being offered a salivary cortisol panel and supplements for "adrenal fatigue," that is not a recognised diagnosis — a systematic review of 58 studies concluded it "is still a myth." --- ## The thing worth knowing in advance Among people who reached the clinical threshold for stress-related exhaustion, seven years later **only 16% reported full recovery**. Forty-six per cent still reported extreme fatigue and 73% reduced stress tolerance. Depression and anxiety in the same cohort resolved within months; the exhaustion did not. But **87% were not on sick leave**, and at ten years 83% were working or studying. Functional recovery arrives long before symptomatic recovery. Seventy-three per cent had changed workplace — against a population base rate of 25.5% — which reads less like defeat than like people making an accurate adjustment. Two conclusions follow, and they point the same way. Crossing that threshold has consequences that outlast the sick note by years, so the strongest argument in all of this evidence is for acting well before you get there. And if you are already past it, the trajectory afterwards is not the cliff it feels like from the inside. --- *Sources for every figure here are in the full review. Compiled 26 July 2026.*