sleep, interruptions, multitasking, interpersonal debts, the expectations to bring when dealing with institutions, don't tough out period pain. Outcome type: energy/time.
A 6B 17C 2
Lifespan 2Time and energy 233 contested
This section keeps only the accounts of energy and time. It looks at four things: how much attention you have to use each day, how fast you react, how many mistakes you make, and how long it takes you to get back after being interrupted. All the numbers are measured directly, from experiments or from records that followed a group of people over time (cohort studies). These numbers are not converted into lifespan, nor into mortality. Most items in this section have an evidence grade of B. That is simply how this field is; it does not mean the advice doesn't hold. Item 19 is about the expectations to bring when dealing with police officers, doctors and counter clerks; it cites regulations, not experiments, so its grade is recorded as C. Item 15 (treat pessimistic thoughts as a symptom) and Item 22 (treat “others demand that I be perfect” as a symptom) cite figures on mortality and suicide risk; their outcome type differs from the other items in this section, so they are not compared with them.
The items in this section are grouped by topic below; the numbers in parentheses are item numbers.
Attention at work: turn off nonessential notifications (Item 1), handle messages in batches each day (Item 5), block out interruptions of even a few seconds (Item 6), do one thing at a time (Item 7), use earplugs in an open-plan office (Item 12), don't keep piling on weekly work hours (Item 13).
Sleep and staying up late: get up at a fixed time (Item 2), sleep a full 7 to 8 hours every night (Item 3), no caffeine after two in the afternoon (Item 4), no screens in the hour before bed (Item 8), go to bed on time (Item 9), no major decisions late at night (Item 10), nap for 10 minutes when you're sleepy in the afternoon (Item 11).
Thoughts in your head: switch to a hands-on task when you keep replaying something (Item 14), treat pessimistic thoughts as a symptom (Item 15), divide your estimate by two when you feel you've embarrassed yourself (Item 21), treat “others demand that I be perfect” as a symptom (Item 22).
Anger and low mood: leave the scene first when you're angry (Item 17), the few things to do first when you're feeling low (Item 18), don't vent anger by smashing things (Item 23), write down what's on your mind over several days in a row (Item 24).
People and social media: learn to turn down requests you don't want to take on (Item 16), see police officers and doctors as people doing a job by the rules (Item 19), set limits on apps for scrolling your peers' updates (Item 20).
Pain in your body: for period pain, take a painkiller like ibuprofen (Item 25).
Sources are reproduced exactly as they appear in the Chinese original, including Chinese titles of laws and quoted statutory text. Original Chinese text of this section at upstream commit bb25081, on GitHub.
Turn off nonessential notifications, and keep your phone out of sight while you work
Value for cost High
In plain terms
When your phone goes off, even if you don't look at it or touch it, your score on attention tasks drops. The drop is about the same as when you actually take a call or send a text. There is another case too: with the phone just lying on the desk, even if you hold back from looking at it, your brain has less spare capacity to work with. The more someone depends on their phone, the bigger the drop.
Cost
Free. A few minutes to change some settings; it takes almost no willpower.
No moneyDone in passingNo willpowerBenefit size medium
Benefit
Merely receiving a notification, without looking at or touching the phone, made performance on attention tasks noticeably worse. The decline was comparable in size to actually making or taking calls or sending texts. This was done in the lab, and the participants were college students. Two other experiments showed that even when people successfully resist looking at their phone, having their own phone lying nearby reduces the brain's available spare capacity. People with strong phone dependence dropped the most.
Evidence grade
B
Notes
These are all single tasks in the lab. What builds up when this happens day after day, no one has measured. For jobs that truly require being on call, you can keep notifications on for just a few contacts.
Sources
Stothart, Mitchum & Yehnert (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. https://doi.org/10.1037/xhp0000100；Ward, Duke, Gneezy & Bos (2017). Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research. https://doi.org/10.1086/691462
The fifth of people with the most irregular schedules had body clocks about 2.5 hours later than the most regular fifth. That amounts to living in jet lag for the long term. Students with more regular schedules also got better grades.
Cost
Free. You have to sleep in less on weekends, which takes a little willpower.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
61 college students kept records for 30 days. The most irregular fifth had body clocks (circadian rhythms) about 2.5 hours later than the most regular fifth. Specifically, the time melatonin starts to be released was 00:08 versus 21:32. The sleepiest period, when they most wanted to sleep, was also about 1.8 hours later. The more regular the schedule, the better the academic performance (correlation coefficient r = 0.37).
Evidence grade
B
Notes
There is only this one study, it had few people, and it had no control groups either. The researchers estimated that the difference in body clocks came mainly from differences in how much light people got each day. So a fixed wake-up time has to go together with getting light in the morning to have an effect. For how to make up after an occasional late night, see Section 2, Item 38 (catch up on sleep the very next night after staying up late): the way to do it is to go to bed early that night, not to sleep until noon the next day.
Sources
Phillips et al. (2017). Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sleep/wake timing. Scientific Reports. https://doi.org/10.1038/s41598-017-03171-4
Sleep a full 7 to 8 hours every night; don't treat 6 hours as enough
Value for cost Standard
In plain terms
Sleep only 6 hours a night for two weeks in a row, and your brain's performance falls to the level of going two days and two nights without shutting your eyes. The trouble is that you can't feel it. Self-rated sleepiness rises over the first few days and then stops rising; people sleeping 6 hours rate themselves just as sleepy as people sleeping 4 hours. So the statement “6 hours of sleep is enough for me” is itself a symptom of not getting enough sleep.
Cost
Free. You have to set aside 1 to 2 more hours a day for sleep. Going to bed a bit earlier takes a little willpower.
No moneyTime every daySome willpowerBenefit size large
Benefit
48 healthy adults aged 21 to 38 slept only 6 hours a night for 14 days in a row. Their mental performance declined cumulatively, falling to the level of 2 nights in a row with no sleep at all. Yet their self-rated sleepiness rose only in the first few days and barely increased after that. The 6-hour group and the 4-hour group did not differ in self-rated sleepiness. This was a controlled trial that randomly divided people into two groups (RCT). Separately, short-term sleep loss (acute sleep deprivation, no more than 48 hours) has an effect size of g = -0.776 on simple lapses of attention and g = -0.125 on reasoning accuracy. This part pools 70 studies.
Evidence grade
A
Notes
“I don't feel sleepy” is not evidence that you have slept enough. People who sleep 6 hours for a long time have only gotten used to it in how they feel; their performance has not kept up. The part that pools many studies is about not sleeping all night. “Sleeping too little over the long term” is supported by only one randomized controlled trial. How sleep length relates to the risk of death is a separate account; see Section 2, Item 13 (sleep about 7 hours a night): there it is counted by mortality, and the 7-hour band has the lowest risk.
Sources
Van Dongen, Maislin, Mullington & Dinges (2003). The cumulative cost of additional wakefulness. Sleep. https://doi.org/10.1093/sleep/26.2.117；Lim & Dinges (2010). A meta-analysis of the impact of short-term sleep deprivation on cognitive variables. Psychological Bulletin. https://doi.org/10.1037/a0018883
Drink a large coffee (400 mg) 6 hours before bed, and that night instruments measure more than 1 hour less sleep. Yet the sleep diary you keep yourself shows no difference at all. An ordinary cup of coffee (107 mg) has to be finished more than 8.8 hours before bed for it not to affect your sleep. If you go to bed at eleven, two in the afternoon is the line.
Cost
Free. If you are used to drinking coffee in the afternoon, holding back from it takes a little willpower.
No moneyDone in passingSome willpowerBenefit size large
Benefit
Taking 400 mg of caffeine 6 hours before bed cut total sleep time, as measured by instruments, by 1.1 to 1.2 hours compared with a placebo. A placebo here is a control drink with no caffeine in it. But the sleep diaries the participants kept themselves reported no difference; they could not notice it. Pooling many studies (meta-analysis): caffeine reduced total sleep time by 45 minutes on average. The share of time in bed actually spent asleep (sleep efficiency) fell by 7%. A cup of coffee (107 mg) has to be finished more than 8.8 hours before bed for it not to affect sleep.
Evidence grade
A
Notes
The Drake study was done by only one lab, and the 400 mg it used is on the high side, about two or three cups of coffee. For people who metabolize caffeine slowly, the cutoff has to be even earlier. What cutting back on caffeine buys back is sleep time, which stacks with Item 3 (sleep a full 7 to 8 hours every night).
Sources
Drake, Roehrs, Shambroom & Roth (2013). Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. Journal of Clinical Sleep Medicine. https://doi.org/10.5664/jcsm.3170；Gardiner et al. (2023). The effect of caffeine on subsequent sleep: A systematic review and meta-analysis. Sleep Medicine Reviews. https://doi.org/10.1016/j.smrv.2023.101764
Switch email and messages to batch handling at a few fixed times a day
Value for cost Standard
In plain terms
The same group of people was compared across two weeks. In the week when email was limited to just 3 checks a day, daily stress was clearly lower, and they sent and received just as many emails. Without a limit, people checked an average of 15.5 times a day.
Cost
Free. It takes a little time to make clear to colleagues how often you will reply. Holding back from checking email at every moment takes a little willpower.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
124 adults, the same people compared before and after across two weeks (within-person comparison). In the week limited to checking email 3 times a day, daily stress was clearly lower than in the week with no limit (effect size Cohen's d = 0.37). The number of emails received and replied to did not differ between the two weeks. At the start, with no limit, people checked email an average of 15.5 times a day.
Evidence grade
B
Notes
There is only one study, the effect is moderate leaning small, and what it measured is stress, not the work that got done. No one has studied instant messages directly; here it is only inferred by analogy.
When doing work that needs sustained thought, keep out even interruptions of a few seconds
Value for cost Very high
In plain terms
A single interruption averaging 2.8 seconds doubles the chance of error in the steps that follow. A 4.4-second interruption triples it. Field tracking also found that it takes an average of 25 minutes to get back to the interrupted task, with two other things wedged in between. People who were interrupted worked faster, but their stress, frustration and sense of effort were all higher.
Cost
Free. Close the door, put on headphones, or hang up a “Do Not Disturb” sign.
No moneyDone in passingNo willpowerBenefit size large
Benefit
An interruption averaging 2.8 seconds doubled the step error rate on tasks done step by step in a fixed order (sequential tasks). A 4.4-second interruption tripled it. This was done in the lab. Another study was a field observation of 24 white-collar office workers (information workers). They switched work topics on average every 11 minutes. 57% of work topics were interrupted. Getting back to the interrupted task on the same day took an average of 25 minutes 26 seconds. In between, 2.26 other tasks were wedged in. There was also an experiment with 48 people: those who were interrupted worked faster, but reported higher stress, frustration, time pressure and effort.
Evidence grade
B
Notes
That field study split interruptions into two kinds: one where someone else comes to you, and one you open yourself. Closing the door only blocks the first kind. The second kind needs Item 1 (turn off nonessential notifications) and Item 5 (handle messages in batches) used along with it.
Sources
Altmann, Trafton & Hambrick (2014). Momentary interruptions can derail the train of thought. Journal of Experimental Psychology: General. https://doi.org/10.1037/a0030986；Mark, Gonzalez & Harris (2005). No task left behind? Examining the nature of fragmented work. CHI 2005. https://doi.org/10.1145/1054972.1055017；Mark, Gudith & Klocke (2008). The cost of interrupted work: More speed and stress. CHI 2008. https://doi.org/10.1145/1357054.1357072
Do one thing at a time; don't answer messages while sitting in a meeting
Value for cost Standard
In plain terms
As soon as you switch tasks, you react more slowly and make mistakes more easily. People who often have several screens open at once have not trained up the ability to do several things at once; they actually do worse on task-switching tests.
Cost
Free. Holding back from answering messages during meetings takes a little willpower.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
After switching tasks, reactions were noticeably slower and errors more likely. This part comes from a review of many studies. Another lab study had college students as participants. People who often use several kinds of media at the same time (heavy media multitaskers) were more easily pulled off track by irrelevant things and irrelevant memories. They actually did worse on task-switching tests.
Evidence grade
B
Notes
How much extra switching back and forth costs varies a lot with the work being done, so no single figure is given here. The Ophir study compared people at a single point in time, so it cannot tell apart “multitasking makes attention worse” from “people with poor attention like multitasking more.”
No glowing screens in the hour before bed; if you read, read a paper book or an e-ink screen
Value for cost Standard
In plain terms
Reading a glowing screen before bed means falling asleep about 10 minutes later than with a paper book. Melatonin is suppressed by 55%. The next day your body clock is pushed back more than 1.5 hours; you are sleepier in the morning and take longer to come fully awake.
Cost
Free. Putting your phone down an hour before bed means fighting habit over the long term, which takes willpower.
No moneyDone in passingLots of willpowerBenefit size medium
Benefit
In the lab, the same people did both conditions in turn (crossover trial). Reading a light-emitting e-book before bed meant falling asleep about 10 minutes more slowly than reading a paper book (25.7 minutes versus 15.8 minutes). Evening melatonin was suppressed by 55%. The next day's body clock (circadian rhythm) was delayed by more than 1.5 hours; the time melatonin starts to be released was 22:31 versus 21:01. The next morning they were sleepier and took longer to become fully alert.
Evidence grade
B
Notes
The sample was small, and it was done in the lab under tightly controlled conditions. At everyday phone brightness and length of use, the effect may not be this large. How much a phone's night mode helps, this study did not test.
Sources
Chang, Aeschbach, Duffy & Czeisler (2015). Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS. https://doi.org/10.1073/pnas.1418490112
Go to bed on time; don't stay up late for games, short videos or porn
ContestedValue for cost High
In plain terms
In 2025, people in China slept an average of 6.97 hours at night and did not fall asleep until 0:10 on average. People who use a phone or tablet before bed are more than twice as likely to get too little sleep, and 2.7 times as likely to be sleepy during the day. When the device is just in the bedroom and not being used, the numbers are about the same. So the easiest fix is to keep the phone out of the bedroom.
Cost
Free. The hard part is fighting the “just one more episode” habit over the long term. Keeping your phone out of the bedroom takes the least effort.
No moneyDone in passingLots of willpowerBenefit size large
Benefit
In 2025, residents of China slept an average of 6.97 hours at night, and the average time of falling asleep was 0:10 (Chinese Sleep Research Society). People who used portable devices before bed were about 2.17 times as likely to get too little sleep (OR 2.17). About 1.46 times as likely to have poor sleep quality (OR 1.46). About 2.72 times as likely to be sleepy during the day (OR 2.72). When the device was merely in the bedroom and not used, the figures were about the same: about 1.79, 1.53 and 2.27 times (OR 1.79/1.53/2.27). This part pools 20 surveys taken at a single point in time, covering 125198 children and adolescents. Another survey taken at a single point in time: among 844 adults aged 18 to 94, those who still used their phone after turning off the lights fell asleep more slowly, spent a smaller share of their time in bed actually asleep, and were more tired during the day.
Evidence grade
B
Notes
Contested. For the cost of staying up late itself, see Item 3 of this section (sleep a full 7 to 8 hours every night). As for what effect porn has on ordinary people's energy and mental capacity, there is at present no reliable evidence that can say whether porn is what causes it. One analysis that pooled many studies concluded that so-called “porn problems” come more from a mismatch between a person's beliefs and their own behavior than from the content itself. Another study pooled data from 280 men. The result was that time spent viewing was unrelated to erectile function during sex with a partner. This item is aimed only at the way it eats into sleep time. For the two figures 6.97 hours and 0:10, the Chinese Sleep Research Society's official website has not posted a release (checked September 2026); the Health News (健康报) report reposted by the Fujian Provincial Health Commission is taken as authoritative.
Don't make major decisions or send important messages late at night
Value for cost Standard
In plain terms
When people are well rested, they quickly learn to steer clear of high-risk options. After two nights in a row without sleep, they instead pick the high-risk ones again and again, and the older they are, the more so. How much a few hours less sleep matters, no one has measured, but still push important decisions to after you've slept.
Cost
Free. You just put things off until after you've slept.
No moneyDone in passingSome willpowerBenefit size small
Benefit
34 healthy people did the Iowa Gambling Task after 49.5 hours without sleep. This is an experiment that measures risky decision-making. When well rested, they quickly learned to avoid the high-risk options. After going without sleep, they instead chose the high-risk options more often. Older people were affected more. This is one experiment, and it is the only study.
Evidence grade
B
Notes
The study looked at the extreme case of two nights in a row without sleep. What happens when an ordinary person stays up until two in the morning has not been measured directly. The “don't send messages” half is the author's experience (grade C); sending it the next day is not too late.
When you're sleepy in the afternoon, nap for 10 minutes, not half an hour
Value for cost High
In plain terms
A 10-minute nap perks you up right away, and the effect lasts two and a half hours. A 5-minute nap is basically useless. With a 20-minute nap, you have to wait half an hour before it starts to work. After a 30-minute nap, you first go through a groggy spell, and during it you perform worse than if you hadn't slept at all.
Cost
Free. It takes 10 minutes, plus a place where you can lie down or put your head down on a desk.
No moneyDone in passingNo willpowerBenefit size medium
Benefit
After too little sleep the night before, a 10-minute nap immediately improved sleepiness, fatigue, vigor and mental performance, and part of the benefit lasted beyond 155 minutes. A 5-minute nap gave almost no benefit. The benefit of a 20-minute nap appeared only after 35 minutes. A 30-minute nap was followed first by a period of sleep inertia: right after waking, performance was actually worse, and it improved only later.
Evidence grade
B
Notes
There is only one lab study, and the participants were all healthy young people. If you can't fall asleep, don't force it. Whether lying down with your eyes closed for 10 minutes helps, no research supports. Napping longer than an hour is instead linked with higher mortality and higher risk of coronary heart disease; see Section 2, Item 37 (keep naps within half an hour).
Sources
Brooks & Lack (2006). A brief afternoon nap following nocturnal sleep restriction: Which nap duration is most recuperative? Sleep. https://doi.org/10.1093/sleep/29.6.831
When working in an open-plan office, use earplugs or find a quiet room for work that takes memory
Value for cost Standard
In plain terms
When background noise rises from 39 decibels to 51 decibels, you remember fewer words, you are more tired, and you feel less like working. During breaks, watching a nature film with sound gets your energy back better than going on listening to office noise.
Cost
Earplugs cost a few dozen yuan. Booking a quiet meeting room takes a little time.
A little moneyDone in passingNo willpowerBenefit size small
Benefit
Students were set to work in a simulated open-plan office. When background noise rose from 39 decibels to 51 decibels (dB, LAeq), they remembered fewer words, rated themselves as more tired, and felt less like working. During breaks, watching a nature film with sound brought their energy ratings back better than listening only to the sound of a river or to office noise.
Evidence grade
B
Notes
There is only one lab study. The most disruptive part of noise is speech you can make out. White-noise headphones help with this, but this study did not test headphones.
Sources
Jahncke, Hygge, Halin, Green & Dimberg (2011). Open-plan office noise: Cognitive performance and restoration. Journal of Environmental Psychology. https://doi.org/10.1016/j.jenvp.2011.07.002
Don't keep piling on weekly work hours: past roughly 49 hours, each extra hour adds less and less extra output
Value for cost High
In plain terms
Records from British munitions factories in World War I show that up to 49 hours a week, output rose in step with the hours worked. Beyond 49 hours, each extra hour added less and less output. A 70-hour week produced almost the same as a 56-hour week; the extra 14 hours were worked for nothing.
Cost
You may earn somewhat less overtime pay. You may also have to put in effort to negotiate with your boss.
A little moneyDone in passingSome willpowerBenefit size large
Benefit
Output records of workers in British munitions factories during World War I: up to 49 hours a week, output was proportional to hours worked. Beyond 49 hours, output began to grow more slowly. It peaked at about 63 hours. Output at 70 hours was almost no different from output at 56 hours; the extra 14 hours amounted to working for nothing.
Evidence grade
B
Notes
The data come from factories a hundred years ago; the work was physical, and the workers were mostly women. This threshold may not be the same when carried over to mental work, and the study's author also says it may be higher or lower for other kinds of work. What is reliable is the direction: past a certain point, working more yields almost no more output. The specific number 49 is not reliable.
When you notice yourself replaying the same upsetting thing over and over, switch to a task that needs your hands
Value for cost Standard
In plain terms
Replaying the same upsetting thing over and over will not come up with an answer. It only deepens depression, nails negative thoughts in more firmly, leaves you less able to solve problems and less able to actually act, and wears out the people around you who are willing to listen to you. When you notice you are going around in circles, go do something that needs your hands.
Cost
Free. The hard part is first noticing that you are replaying it over and over (this is called rumination); the noticing itself takes practice.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
A review of many studies says that replaying the same upsetting thing over and over (ruminative thinking) worsens depression. It also strengthens negative thinking and weakens the ability to solve problems. It gets in the way of actually taking action, and it wears out the people around you who are willing to listen to you. It is associated with anxiety, binge eating, heavy drinking and self-harm.
Evidence grade
B
Notes
The review gives only a direction, no numbers. “Switching to another task” is the author's own practice. All the literature can hold up is one sentence: diverting your attention is better than going on circling in place. As for what task to switch to, no one has compared.
Treat thoughts like “things are sure to get worse” as a symptom, not as a fact
ContestedValue for cost Standard
In plain terms
In a twin study that followed people for 20 years, each notch higher in pessimism came with about 13% higher odds of dying during that period, and about 20% higher odds of dying of cardiovascular disease. Conversely, high optimism scores did not bring a matching benefit. So what needs handling is pessimism, not “not enough positive energy.”
Cost
Free. The hard part is seeing this kind of thought as a symptom rather than as a fact.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
2978 Australian twins aged over 50 were followed for an average of 20 years, during which 1068 of them died. Each 1 standard deviation higher on the pessimism scale meant about 13% higher risk of death during that period (hazard ratio 1.134). The confidence interval for this figure is 95% CI 1.065–1.207, P=8.85×10⁻⁵. Cardiovascular death was about 20% higher (1.196, 1.045–1.368, P=0.0093). There was no association with cancer death. Optimism scores also had no clear association with mortality.
Evidence grade
B
Notes
Contested. This was only observed, not randomized. Some will say “in fact people get sick first and then become pessimistic”; the authors counter that if that were so, cancer death should also be associated. But there may still be factors that were not measured. Optimism did not protect, so there is no need to force yourself to “think positive.” What to do is stop taking pessimistic expectations as accurate predictions of the future.
Sources
Whitfield JB, Zhu G, Landers JG, Martin NG (2020). Pessimism is associated with greater all-cause and cardiovascular mortality, but optimism is not protective. Scientific Reports, 10, 12609. https://doi.org/10.1038/s41598-020-69388-y
Cut back on relationships that drain you, and learn to turn down requests you don't want to take on
Value for cost Standard
In plain terms
The harm done by one person who drains you is larger, and more consistent, than the good done by one person who supports you. So subtract first, then add.
Cost
Free. In the short term there will be social friction, and you may lose some relationships. Saying no out loud takes considerable willpower.
No moneyDone in passingLots of willpowerBenefit size medium
Benefit
A survey of 120 widowed women aged 60 to 89: unpleasant interactions were linked with psychological well-being more consistently, and more strongly, than pleasant ones. By this result, the harm done by one person who drains you is greater than the good done by one person who supports you.
Evidence grade
C
Notes
The people surveyed were a special group, and it was a one-time questionnaire, so it cannot show what came first. What it measured is well-being, not energy. Putting it in this section is the author's inference. There is no objective standard for which relationships count as “draining”; you can only judge for yourself.
Sources
Rook (1984). The negative side of social interaction: Impact on psychological well-being. Journal of Personality and Social Psychology. https://doi.org/10.1037/0022-3514.46.5.1097
When you're angry, leave the scene first; treat the other person as weather, not as an enemy
Value for cost Standard
In plain terms
Reinterpreting what just happened, or putting yourself in the other person's position for a moment, both genuinely bring your emotions down, and the second works better. Forcing it down so it doesn't show does not work. Also, people who punished the person who offended them actually spent longer afterward replaying it; people who didn't retaliate moved on faster. People overestimate in advance how good revenge will feel.
Cost
Free. In the moment, hold back from talking back, turn and walk away, and decide afterward whether to deal with it. This takes a bit of practice.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
Pooling 306 experimental comparisons (meta-analysis): reinterpreting what happened (cognitive reappraisal) had an effect size on emotion of d = 0.36. Seeing it from the other person's position, d = 0.45. Diverting attention, d = 0.27. Forcing feelings down had no effect (d = -0.04). Three other experiments showed that people who punished an offender actually spent longer afterward replaying thoughts of that person; people who did not punish “moved on” faster. People overestimated in advance how good revenge would feel.
Evidence grade
B
Notes
The part that pools many studies measured emotional intensity, not energy, and the effects are small to moderate. The revenge part has only one set of experiments, so it is graded B. The “empty boat” is a parable in the “Mountain Tree” (山木) chapter of the Zhuangzi: when an empty boat bumps into you, you don't get angry; you get angry only when someone is aboard, yet it is the same bump. This is an analogy, not evidence. If you have really suffered a loss, take the legal route in Section 8 and Section 9; it is much cheaper than blowing up on the spot.
Sources
Webb TL, Miles E, Sheeran P (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin. https://doi.org/10.1037/a0027600；Carlsmith KM, Wilson TD, Gilbert DT (2008). The paradoxical consequences of revenge. Journal of Personality and Social Psychology. https://doi.org/10.1037/a0012165
When you're feeling low, start with the few things that give the best value for cost: get moving, get sunlight, sleep on schedule, talk to someone, call 12356
Value for cost High
In plain terms
For depression, brisk walking and jogging work best, and the higher the intensity the more effective; for details see Section 22, Item 6 (when you're in a bad mood, go walk or run). Light exposure works very well for seasonal depression and also works for depression that is not seasonal. Online self-help cognitive behavioral courses work, and when symptoms are on the heavier side, it is better to have someone guide you through them. When things feel unbearable, call 12356 first.
Cost
Free, or very little money. A few sessions a week of brisk walking, jogging or strength training. Go outside in the daytime to get sunlight. Settle your sleep and waking times into a fixed routine. Make one phone call when things feel unbearable.
No moneyA few hoursSome willpowerBenefit size large
Benefit
218 randomized controlled trials with 14170 people were pooled (network meta-analysis). The control groups were people who were likewise doing some other activity. Compared with them, walking or jogging had an effect size on depression of g = -0.62 (95% confidence interval -0.80 to -0.45). Yoga -0.55, strength training -0.49. The higher the intensity, the better the effect. Pooling several light therapy studies: effect size 0.84 (0.60 to 1.08) for seasonal depression, 0.53 (0.18 to 0.89) for non-seasonal depression. One more analysis pooled the data of each patient from each study. It looked at internet-based cognitive behavioral therapy (online self-help courses). Both guided and unguided versions did better than the control groups. At moderate to severe levels (PHQ-9 above 9 points), the guided version did better. 12356 is the national mental-health assistance hotline, open no fewer than 18 hours a day.
Evidence grade
A
Notes
These are all supplementary measures for mild to moderate depression. People with moderate to severe depression or with suicidal thoughts should see a doctor, and call the 12356 mental-health hotline first (see Section 1, Item 25). Most exercise trials are small, and they cannot hide who is exercising the way drug trials hide who is taking the drug. The light therapy studies only go up to 2003, and they are small too. “Sleep on schedule” and “talk to someone” are the author's experience, with no literature cited for them separately. Don't turn “get moving” into a demand on people who are depressed; managing a ten-minute walk counts.
Treat police officers, doctors and counter clerks as people doing a job by the rules, not as characters: what moves things forward is paperwork and deadlines, not emotion
Value for cost Standard
In plain terms
Police officers, doctors and counter clerks are all people working under the rules of their post, governed by paperwork, deadlines and oversight. Whether the other person's manner is pleasant or not has basically nothing to do with whether your matter gets handled or how many days it takes. So get the receipt, note the deadline, appeal through the proper channel, and don't count on arguing it out until the other person agrees with you.
Cost
Free. What changes is your expectations and the order of your actions. The first time, you have to resist the urge to “argue it out until the other person agrees with you.”
No moneyDone in passingSome willpowerBenefit size small
Benefit
Change the moves you make in dealing with institutions: from “pleading, getting angry, retelling your grievance again and again” to “getting the paperwork, noting the deadlines, appealing through the proper channel.” What you save is the time spent on repeated communication and repeated replaying. This is a one-time change of outlook, not a time saving every day. What can be checked are the deadlines themselves, which are all written in documents. For public order cases, counting from the day the case is accepted, the handling period is no more than 30 days; for major or complex cases, it can be extended by 30 days with the approval of the public security authority one level up. In cases with a complainant where the decision is not to open a case, the notice of the decision not to open a case must be delivered within 3 days. The complainant can apply for reconsideration within 7 days, and the public security authorities (police) make a decision within 30 days. If the complainant does not accept it, they can apply to the next level up for review within a further 7 days.
Evidence grade
C
Notes
The reason for C: the judgment that “many people treat the person behind the window as a character, not as someone at work” has no literature to cite. Only the rules half can be cited: the deadlines, paperwork and channels of redress are all written down. All the figures cited in this item come from the two Ministry of Public Security regulations in Section 8, Item 39 (what happens after you report to the police). Following the procedure does nothing about “the other person has a bad attitude,” and it does not guarantee your matter will get done; all it changes is your expectations and your first move. Bank tellers and government service windows are the same: their operations leave a record, are subject to review, and mistakes bring accountability. So what to do is keep the slip, ask for the receipt and complain through the proper channel, rather than guessing blindly about “whether they'll cheat me.” Legitimate priorities are all out in the open; for example, the emergency department goes by how serious your condition is, not by who came first (Section 24, Item 7, emergency triage). If you want to thank a doctor who saved you, use a thank-you letter and a satisfaction rating; see Section 24, Item 11. Wanting to “give them a little something so they'll take more care”: toward people who handle cases and enforce the law, that is the crime of offering a bribe, and one the law explicitly says to punish more heavily; see Section 8, Item 40. If you run into obviously unequal treatment, complain when you should; don't treat it as an unwritten rule to copy.
Don't make “how other people are doing” your daily required reading: set limits on, or turn off, the apps where you scroll your peers' updates
Value for cost High
In plain terms
People who habitually compare themselves online with people doing better tend to be in a worse emotional state and have lower self-esteem. People who deactivated Facebook for four weeks became slightly happier, but their knowledge of current events also dropped a notch. People who limited three social platforms to 10 minutes a day each saw both loneliness and depression come down after three weeks.
Cost
Free. Setting limits on an app, or deactivating the account, takes a few minutes either way. The hard part is not opening them again afterward, which means fighting habit over the long term.
No moneyDone in passingLots of willpowerBenefit size large
Benefit
Pooling many studies (meta-analysis): 54 independent samples, 94 effect sizes, N = 36,583, using a three-level random-effects model. The average correlation between online upward social comparison and psychological maladjustment was r = 0.330 (95% CI 0.289–0.370, confidence interval). Broken down by specific outcome: negative emotions about social evaluation, such as fear of being looked down on, 0.438; anxiety 0.382. Depression 0.306, low well-being 0.268, low self-esteem 0.263. Age, cultural background and the year the data were collected did not affect the strength of this association. Studies that followed people over the long term also did not find a stronger association than surveys taken at a single point in time. Another randomized controlled trial had 2,743 American Facebook users as participants. Of them, 61% were willing to accept less than 102 US dollars to deactivate. These people were randomly assigned to a four-week deactivation group or a control group. Deactivation freed up 60 minutes a day for the typical participant. A subjective well-being index rose by 0.09 standard deviations. By component: life satisfaction 0.12, anxiety 0.10, depression 0.09, happiness 0.08. These still count after discounting what could be chance from making many comparisons. The authors gave a reference point: psychological interventions designed to make people more positive. Such interventions move subjective well-being by about +0.34 standard deviations and depression by about -0.23 standard deviations. So this effect is about 25% to 40% of a standard psychological intervention. The cost over the same period: a news knowledge index fell by 0.19 standard deviations, and political polarization fell by 0.16 standard deviations. After the experiment ended, the deactivation group still used Facebook 12 minutes/day less than the control group (0.31 standard deviations, a 23% drop). Yet another randomized controlled trial had 143 undergraduates as participants, randomly split into two groups. One group limited Facebook, Instagram and Snapchat to 10 minutes per platform per day; the other group used them as usual. After three weeks, loneliness and depression fell clearly in the limited group. In both groups, anxiety and fear of missing out were lower than at the start.
Evidence grade
A
Notes
The “correlation” half only shows that the two things occur together; it cannot be read as one causing the other, and the authors themselves say more long-term follow-up studies are needed. The randomized half can speak to cause and effect, but the participants were Facebook users before the 2018 US midterm elections, and applying it to platforms in China is forcing a fit. The Hunt study had only 143 undergraduates, and it is the only one. The effect sizes are all small; don't expect that turning off apps will solve emotional problems. What to block is the “seeing how other people are doing” kind of use; keeping in touch with family and friends, looking up information and job hunting are not included. If you really can't hold back, start by turning off push notifications, “people you may know” and home-page recommendations. For the overall screen-time accounting, see Section 4, Items 15 and 16 (a hard cap on short videos, no scrolling news). For what to do first when you're feeling low, see Item 18 of this section (get moving, get sunlight, sleep on schedule). For buying things to get back a sense of identity, see Section 6, Item 23. For spending more to “move up a notch,” see Section 6, Item 24.
Sources
Lei Y, Hu S, Sun Y, Zheng L (2026). "Looking up" linked to feeling down: a meta-analysis of online upward social comparison and psychological maladjustment. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2026.1825169；Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2020). The Welfare Effects of Social Media. American Economic Review, 110(3), 629–676. https://doi.org/10.1257/aer.20190658；Allcott H, Braghieri L, Eichmeyer S, Gentzkow M (2019). The Welfare Effects of Social Media. NBER Working Paper No. 25514（本条引用的具体数字按这一版逐字核对，刊出版的摘要不含数字）. https://www.nber.org/papers/w25514；Hunt MG, Marx R, Lipson C, Young J (2018). No More FOMO: Limiting Social Media Decreases Loneliness and Depression. Journal of Social and Clinical Psychology, 37(10), 751–768. https://doi.org/10.1521/jscp.2018.37.10.751
When you feel “everyone saw me embarrass myself,” divide that estimate by two
Value for cost Standard
In plain terms
When people wore an embarrassing T-shirt into a room with others in it, the wearers thought nearly half the people could recognize who was printed on the shirt; in fact it was under a quarter. When people wore a T-shirt they thought looked respectable, their estimate was six times the actual number. After chatting, people also generally underestimate how much the other person likes them. So when you feel “everyone remembers me embarrassing myself,” even cutting it in half still leaves it too high.
Cost
Free. Just ask yourself one more question when you're nervous: did they really see it?
No moneyDone in passingSome willpowerBenefit size small
Benefit
Experiment one: participants put on a T-shirt printed with Barry Manilow's face and walked into a room with other participants in it. Wearers estimated on average that 46% of the observers could correctly name the person on the shirt. The observers' actual accuracy was 23%; the estimate was exactly twice the actual figure. Subtracting the actual figure from the estimate session by session, the average overestimate was 23 percentage points (95% CI 9–38, confidence interval). After excluding the two sessions with only two and three observers, it was 27 percentage points (11–43). Experiment two: participants instead wore a T-shirt they chose themselves and considered positive. The average estimate was 48%, while the observers' actual accuracy was 8%, six times over. Another series of studies had three settings: strangers paired up to talk, first-year college roommates, and adult workshops. The result was that people systematically underestimate how much their conversation partner likes them, and also underestimate how much the partner enjoyed the conversation. The authors call this the “liking gap.” It was present in conversations of different lengths, and among first-year roommates it lasted for months.
Evidence grade
B
Notes
The reason for B: these are all lab and campus studies with not many people, and the settings are wearing clothes and chatting with strangers. There is no direct evidence for carrying them over to “being talked about by coworkers” or “saying the wrong thing in a meeting.” The figures only show the direction of the overestimate and its rough multiple; they cannot be applied as a coefficient. Nor is it true that other people's opinions never matter. In settings where you really are being evaluated, such as job interviews, thesis defenses, performance reviews and medical checkup reports, the evaluation is real, so prepare when you should. What to discount is only the everyday feeling that “everyone is looking at me”; what that part needs dealing with is the bias in your own estimate.
Sources
Gilovich T, Medvec VH, Savitsky K (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211–222. https://doi.org/10.1037/0022-3514.78.2.211；Boothby EJ, Cooney G, Sandstrom GM, Clark MS (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742–1756. https://doi.org/10.1177/0956797618783714
Treat “others demand that I be perfect” as a symptom, not as a fact
Value for cost Standard
In plain terms
People who constantly feel “the people around me demand that I be perfect, and if I can't do it I'll be rejected” are more prone to suicidal thoughts. “I demand that I myself be perfect” is much more weakly tied to suicidal thoughts, only about a third as strongly as the former. After subtracting the thoughts people already had, the former feeling is still linked with more such thoughts later, just not by much.
Cost
Free. The hard part is treating this kind of thought as something that needs handling, rather than as an accurate measurement of what others really demand.
No moneyDone in passingSome willpowerBenefit size small
Benefit
Pooling many studies (meta-analysis): 45 studies, N = 11,747, in populations of undergraduates, medical students, community adults and psychiatric patients, using a random-effects model. Below are the weighted average correlations with suicidal ideation. Perfectionistic concerns r = .28 (95% CI .24–.32, confidence interval; 42 studies, 7,936 people). Perfectionistic concerns is a composite that includes socially prescribed perfectionism, concern over mistakes, doubts about actions, the gap between reality and one's standards, and others. Within it, socially prescribed perfectionism counted on its own is also r = .28 (.25–.32, 30 studies, 3,640 people). Perfectionistic strivings r = .10 (.07–.13, 31 studies, 4,588 people); it is a composite of self-oriented perfectionism and personal standards. Parental criticism r = .20 (.11–.29). Now the relationship with suicide attempts. Perfectionistic concerns r = .12 (.07–.17, 15 studies, 5,275 people). Socially prescribed perfectionism r = .19 (.08–.29, 8 studies, 689 people). The part that followed people over time: after subtracting the suicidal ideation already present at the start, socially prescribed perfectionism still moved in the same direction as suicidal ideation at later follow-up, by a modest amount. Neither self-oriented nor other-oriented perfectionism showed an association.
Evidence grade
B
Notes
The reason for B: most of the included studies were one-time questionnaires, so they cannot show what came first. The correlation is moderate in strength (r about .28), roughly explaining only a few percent to a tenth of the differences. What was measured is suicidal ideation and suicide attempts, not death itself, so the size of the benefit is recorded as small under “surrogate endpoint only.” The authors themselves also say the group of results on perfectionistic concerns and suicide attempts looks like it has publication bias. Publication bias means studies whose results did not look good never got published. After the estimated missing studies are filled back in, the direction of the conclusion does not change. High standards are not the problem in themselves, and there is no need to lower your standards because of this. The difference lies in where the standards come from, and whether falling short of them makes you feel “I as a person am finished.” Like the pessimistic expectations in Item 15 (treat thoughts like “things are sure to get worse” as a symptom), it is something that can be handled as a symptom. If you have suicidal thoughts, first call 12356 and move lethal means out of reach; see Section 1, Item 25. For what you can do when someone close to you shows such thoughts, see Section 8, Item 15. For depression screening for children, see Section 30, Item 8.
Sources
Smith MM, Sherry SB, Chen S, Saklofske DH, Mushquash C, Flett GL, Hewitt PL (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. https://doi.org/10.1111/jopy.12333
When you're angry, don't vent by smashing things, hitting a punching bag or going for a run; first let your body slow down
Value for cost Standard
In plain terms
Ways of venting that rev the body up more, such as hitting a punching bag, throwing things or going for a run, overall do not get rid of anger. Ways that calm the body down, such as slow breathing, mindfulness and meditation, clearly push down anger and the urge to lash out physically.
Cost
Free. In the heat of anger, switch to a few minutes of something that slows your heartbeat: breathe slowly, sit down, be still. The hard part is that venting is exactly what you most want to do in that moment.
No moneyDone in passingSome willpowerBenefit size medium
Benefit
154 studies were pooled (meta-analysis), with 184 samples and 10,189 people in all. Ways of calming the body down, such as deep breathing, mindfulness and meditation, lowered anger and aggression. Effect size g = -0.63, 95% CI -0.82 to -0.43 (confidence interval). This result held steady among people of different sexes, races, ages and cultures. Ways of making the body more aroused, such as hitting a punching bag, jogging or cycling, were ineffective overall. Their effect size was g = -0.02, 95% CI -0.13 to 0.09. The authors' conclusion: venting anger and going for a run are not effective ways to cool down.
Evidence grade
A
Notes
What was measured is anger and aggression, not energy. Recording the outcome type as time is the author's classification: one fewer fight means one fewer mess to clean up. For methods you can use on the spot, see Section 22, Item 8 (cyclic sighing), and also Item 17 of this section (when you're angry, leave the scene first). Running is not useless. Over the long term it works for low mood; see Section 22, Item 6 (when you're in a bad mood, go walk or run). It just isn't suited to cooling down on the spot. In the heat of anger, don't drive, don't send messages, don't make decisions. If someone is hitting you, leave first and call the police. At that point what needs handling is not your emotions; see Section 8, Item 43 (if you are subjected to domestic violence).
Sources
Kjærvik SL, Bushman BJ (2024). A meta-analytic review of anger management activities that increase or decrease arousal: What fuels or douses rage? Clinical Psychology Review. https://doi.org/10.1016/j.cpr.2024.102414
If something is weighing on you, you can write down your thoughts and feelings over several days in a row, but don't expect it to cure anything
ContestedValue for cost Standard
In plain terms
Writing down upsetting things does a little good on average, but very little. The pooled conclusions are also inconsistent; two of the pooled analyses found no effect. It costs almost nothing, so you can try it. If emotional problems get serious, don't use it in place of seeing a doctor.
Cost
Free. The common method in the studies is to write 3 to 4 times, 15 to 20 minutes each time. Write about something that upsets you, and write your deepest thoughts and feelings.
No moneyDone in passingNo willpowerBenefit size small
Benefit
One analysis pooled 146 randomized studies (meta-analysis). People in the studies wrote down, or talked about, things that mattered to them. The average effect was positive and statistically significant, but very small, effect size r = .075. Another pooled analysis took in only randomized controlled trials; pooling 30 trials, there was no statistically significant effect on either physical health or mental health. Pooling 16 randomized trials in cancer patients and survivors, the psychological, physical and quality-of-life outcomes were all not statistically significant. For the psychological outcome, g = 0.04, 95% CI -0.06 to 0.14 (confidence interval).
Evidence grade
B
Notes
Contested. The supporting side is the pooled analysis of 146 studies, with a positive but very small effect. The opposing side is two pooled analyses that took in only randomized controlled trials, neither of which found a statistically significant effect. How many times and how long each time, as written in the Cost field, come from the earliest experimental design; the abstract of the Mogk pooled analysis states it. It costs almost nothing, doesn't take much time either, and you can do it on your own. When you don't want to talk to anyone, and have no one around to talk to either, you can use it. What it measures is not energy; recording the outcome type as time is the author's classification. If the writing makes you feel worse and worse, stop, and call 12356 instead; see Section 1, Item 25 (12356).
Sources
Frattaroli J (2006). Experimental disclosure and its moderators: a meta-analysis. Psychological Bulletin. https://doi.org/10.1037/0033-2909.132.6.823；Mogk C, Otte S, Reinhold-Hurley B, Kröner-Herwig B (2006). Health effects of expressive writing on stressful or traumatic experiences - a meta-analysis. Psycho-Social Medicine. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2736499/；Zachariae R, O'Toole MS (2015). The effect of expressive writing intervention on psychological and physical health outcomes in cancer patients--a systematic review and meta-analysis. Psycho-Oncology. https://doi.org/10.1002/pon.3802
For period pain, take a painkiller like ibuprofen instead of toughing it out; if the pain keeps you from work or school, or three months of medicine can't control it, get checked by a gynecologist
Value for cost Standard
In plain terms
For period pain, taking a drug like ibuprofen raises the share of people whose pain clearly eases from under two in ten to about half; it works better than acetaminophen, with slightly more side effects. If the pain keeps you from work or school, you have deep pain during sex, or three months of medicine still can't control it, go to gynecology and get checked for endometriosis. On average this disease takes 6.7 years to be diagnosed.
Cost
A box of ibuprofen costs ten-odd yuan, and you take it only on the days it hurts. If you need to go to gynecology, registration and an ultrasound cost some money and half a day.
A little moneyDone in passingNo willpowerBenefit size medium
Benefit
A Cochrane systematic review pooled 80 randomized controlled trials with 5820 women with primary dysmenorrhea. Primary dysmenorrhea means period pain where no other disease can be found. Compared with placebo, nonsteroidal anti-inflammatory drugs (NSAIDs) relieved pain more effectively (OR 4.37, 95% CI 3.76–5.09, confidence interval). Converted, 18% of those taking placebo got moderate or better relief, versus 45% to 53% of those taking these drugs. The price is somewhat more side effects (OR 1.29, 1.11–1.51). 10% of those taking placebo had side effects, versus 11% to 14% of those taking these drugs. The extra ones were mainly stomach upset (OR 1.58) and nervous system reactions (OR 2.74). Compared with acetaminophen, these drugs also relieved pain more effectively (OR 1.89, 1.05–3.43), though there were only 3 trials. Which drug of this kind is best: no difference was found. The UK's NICE guideline (updated in 2024) lists several situations in which endometriosis should be suspected. They are long-term pelvic pain; period pain that affects daily activities and quality of life; and deep pain during or after sex. Also painful bowel movements during periods, blood in the urine or painful urination during periods, and infertility together with these symptoms. Girls aged 17 and under are included too. The guideline recommends first trying acetaminophen or these drugs, for example for 3 months. If pain relief is not enough, refer for further investigation. A normal pelvic exam and ultrasound do not rule out the disease. One study surveyed 1418 women undergoing laparoscopy at 16 hospitals in 10 countries. From the first symptoms to diagnosis of endometriosis at surgery took an average of 6.7 years, with the delay mainly in the primary care stage. Diagnosed women lost an average of 10.8 hours of work a week, mostly from being at work but less productive. An online survey in the Netherlands asked 32748 women aged 15 to 45. 13.8% had taken time off because of their period, an average of 1.3 days a year. 80.7% were at work during their period but less productive, equivalent to 8.9 fewer days of work a year.
Evidence grade
A
Notes
People over 60, people who have had stomach bleeding, and people taking anticoagulants or hormones should ask a doctor before using these drugs. For details see Section 34, Item 3 (ibuprofen-type painkillers). The US ibuprofen label states that it may cause a severe allergic reaction, which is more likely in people allergic to aspirin, and the signs include an asthma attack. People with asthma should also ask a doctor before using it. If you might be pregnant, ask a doctor first, and after 20 weeks of pregnancy don't take it on your own; see Section 34, Item 5 (after 20 weeks of pregnancy). If you can't take these drugs, you can use acetaminophen instead; for its daily limit see Section 34, Item 1 (acetaminophen). For period pain, take the medicine only on the days of your period. The threshold in Section 34, Item 9 (don't take painkillers more than 15 days a month) applies to headaches, and the days you take medicine for period pain come nowhere near it anyway. It is graded A because the pain-relief effect comes from a large number of randomized trials. The authors rated the evidence quality of most comparisons as low, mainly because the trial methods were not clearly reported; 59% of the trials were funded by drug companies. Rating the benefit “moderate” is a judgment call: the review measured only pain, not fewer days off. In the Dutch survey, the losses from menstruation come to about 10 days a year, which, spread over each week, is on the order of hours. The endometriosis group were people already undergoing laparoscopy; it does not mean most people with period pain have this disease. The main beneficiary is you yourself.
Sources
Marjoribanks J, Ayeleke RO, Farquhar C, Proctor M (2015). Nonsteroidal anti-inflammatory drugs for dysmenorrhoea. Cochrane Database of Systematic Reviews, (7):CD001751. https://doi.org/10.1002/14651858.CD001751.pub3；National Institute for Health and Care Excellence (2017, updated 2024). Endometriosis: diagnosis and management (NG73), recommendations 1.3.1, 1.4.2, 1.4.3, 1.5.4. https://www.nice.org.uk/guidance/ng73/chapter/Recommendations；Nnoaham KE, Hummelshoj L, Webster P, et al.; World Endometriosis Research Foundation Global Study of Women's Health consortium (2011). Impact of endometriosis on quality of life and work productivity: a multicenter study across ten countries. Fertility and Sterility, 96(2):366-373.e8. https://doi.org/10.1016/j.fertnstert.2011.05.090；Schoep ME, Adang EMM, Maas JWM, De Bie B, Aarts JWM, Nieboer TE (2019). Productivity loss due to menstruation-related symptoms: a nationwide cross-sectional survey among 32 748 women. BMJ Open, 9(6):e026186. https://doi.org/10.1136/bmjopen-2018-026186；Haleon (2026). Advil (ibuprofen capsule, liquid filled), OTC drug label: Warnings. DailyMed. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1f01c10a-9434-91a4-2ee4-352315a6b610
