Immortality: A Practical Guide

Im­mor­tal­ity: A Prac­ti­cal Guide


This ar­ti­cle is about how to in­crease one’s own chances of liv­ing for­ever or, failing that, liv­ing for a long time. To be clear, this guide defines death as the long-term loss of one’s con­scious­ness and defines im­mor­tal­ity as never-end­ing life. For those who would like less lengthy in­for­ma­tion on de­creas­ing one’s risk of death, I recom­mend read­ing the sec­tions “Can we be­come im­mor­tal,” “Should we try to be­come im­mor­tal,” and “Cry­on­ics,” in this guide, along with the ar­ti­cle Lifestyle In­ter­ven­tions to In­crease Longevity.

This ar­ti­cle does not dis­cuss how to treat spe­cific dis­ease you may have. It is not in­tended as a sub­sti­tute for the med­i­cal ad­vice of physi­ci­ans. You should con­sult a physi­cian with re­spect to any symp­toms that may re­quire di­ag­no­sis or med­i­cal at­ten­tion.

When read­ing about the effect sizes in sci­en­tific stud­ies, keep in mind that many sci­en­tific stud­ies re­port false-pos­i­tives and are bi­ased,101 though I have tried to min­i­mize this by max­i­miz­ing the qual­ity of the stud­ies used. Meta-analy­ses and sci­en­tific re­views seem to typ­i­cally be of higher qual­ity than other study types, but are still sub­ject to bi­ases.114

Cor­rec­tions, crit­i­cisms, and sug­ges­tions for new top­ics are greatly ap­pre­ci­ated. I’ve tried to write this ar­ti­cle tersely, so feed­back on do­ing so would be es­pe­cially ap­pre­ci­ated. Apolo­gies if the ar­ti­cle’s font type, size and color isn’t stan­dard on Less Wrong; I made it in google docs with­out be­ing aware of Less Wrong’s stan­dard and it would take too much work chang­ing the style of the en­tire ar­ti­cle.


  1. Can we be­come im­mor­tal?

  2. Should we try to be­come im­mor­tal?

  3. Rel­a­tive im­por­tance of the differ­ent topics

  4. Food

    1. What to eat and drink

    2. When to eat and drink

    3. How much to eat

    4. How much to drink

  5. Exercise

  6. Carcinogens

    1. Chemicals

    2. Infections

    3. Radiation

  7. Emo­tions and feelings

    1. Pos­i­tive emo­tions and feelings

    2. Psy­cholog­i­cal distress

    3. Stress

    4. Anger and hostility

  8. So­cial and per­son­al­ity factors

    1. So­cial status

    2. Giv­ing to others

    3. So­cial relationships

    4. Conscientiousness

  9. In­fec­tious diseases

    1. Den­tal health

  10. Sleep

  11. Drugs

  12. Blood donation

  13. Sitting

  14. Sleep apnea

  15. Snoring

  16. Exams

  17. Genomics

  18. Aging

  19. Ex­ter­nal causes of death

    1. Trans­port accidents

    2. Assault

    3. In­ten­tional self harm

    4. Poisoning

    5. Ac­ci­den­tal drowning

    6. Inan­i­mate me­chan­i­cal forces

    7. Falls

    8. Smoke, fire, and heat

    9. Other ac­ci­den­tal threats to breathing

    10. Elec­tric current

    11. Forces of nature

  20. Med­i­cal care

  21. Cryonics

  22. Money

  23. Fu­ture advancements

  24. References

Can we be­come im­mor­tal?

In or­der to po­ten­tially live for­ever, one never needs to make it im­pos­si­ble to die; one in­stead just needs to have one’s life ex­pec­tancy in­crease faster than time passes, a con­cept known as the longevity es­cape ve­loc­ity.61 For ex­am­ple, if one had a 10% chance of dy­ing in their first cen­tury of life, but their chance of death de­creased by 90% at the end of each cen­tury, then one’s chance of ever dy­ing would be be 0.1 + 0.12 + 0.13… = 0.11… = 11.11...%. When ap­plied to risk of death from ag­ing, this akin to one’s re­main­ing life ex­pec­tancy af­ter jump­ing off a cliff while be­ing af­fected by grav­ity and jet propul­sion, with grav­ity be­ing akin to ag­ing and jet propul­sion be­ing akin to anti-ag­ing (re­ju­ve­na­tion) ther­a­pies, as shown be­low.

The num­bers in the above figure de­note plau­si­ble ages of in­di­vi­d­u­als when the first re­ju­ve­na­tion ther­a­pies ar­rive. A 30% in­crease in healthy lifes­pan would give the users of first-gen­er­a­tion re­ju­ve­na­tion ther­a­pies 20 years to benefit from sec­ond-gen­er­a­tion re­ju­ve­na­tion ther­a­pies, which could give an ad­di­tional 30% in­crease if life span, ad in­fini­tum.61

As for causes of death, many deaths are strongly age-re­lated. The pro­por­tion of deaths that are caused by ag­ing in the in­dus­trial world ap­proaches 90%.53 Thus, I sup­pose post­pon­ing ag­ing would dras­ti­cally in­crease life ex­pec­tancy.

As for efforts against ag­ing, the SENS Re­search foun­da­tion and Science for Life Ex­ten­sion are char­i­ta­ble foun­da­tions for try­ing to cure ag­ing.54, 55 Ad­di­tion­ally, Cal­ico, a Google-backed com­pany, and Ab­bVie, a large phar­ma­ceu­ti­cal com­pany, have each com­mit­ted fund $250 mil­lion to cure ag­ing.56

I spec­u­late that one could ad­di­tion­ally de­crease risk of death by be­com­ing a cy­borg, as me­chan­i­cal bod­ies seem eas­ier to main­tain than biolog­i­cal ones, though I’ve found no ar­ti­cles dis­cussing this.

Similar to be­com­ing a cy­borg, an­other po­ten­tial method of de­creas­ing one’s risk of death is mind up­load­ing, which is, roughly speak­ing, the trans­fer of most or all of one’s men­tal con­tents into a com­puter.62 How­ever, there are some con­cerns about the trans­fer cre­at­ing a copy of one’s con­scious­ness, rather than be­ing the same con­scious­ness. This is­sue is made very ap­par­ent if the mind-up­loaded pro­cess leaves the origi­nal mind in­tact, mak­ing it seem un­likely that one’s con­scious­ness was trans­ferred to the new body.63 Eliezer Yud­kowsky doesn’t seem to be­lieve this is an is­sue, though I haven’t found a cita­tion for this.

With re­gard to con­scious­ness, it seems that most in­di­vi­d­u­als be­lieve that the con­scious­ness in one’s body is the “same” con­scious­ness as the one that was in one’s body in the past and will be in it in the fu­ture. How­ever, I know of no ev­i­dence for this. If one’s con­scious­ness isn’t the same of the one in one’s body in the fu­ture, and one defined death as one’s con­scious­ness per­ma­nently end­ing, then I sup­pose one can’t pre­vent death for any time at all. Sur­pris­ingly, I’ve found no ar­ti­cles dis­cussing this pos­si­bil­ity.

Although cur­ing ag­ing, be­com­ing a cy­borg, and mind up­load­ing may pre­vent death from dis­ease, they still seem to leave one­self vuln­er­a­ble to ac­ci­dents, mur­der, suicide, and ex­is­ten­tial catas­tro­phes. I spec­u­late that these prob­lems could be solved by giv­ing an ar­tifi­cial su­per­in­tel­li­gence the abil­ity to take con­trol of one’s body in or­der to pre­vent such deaths from oc­cur­ring. Of course, this pos­si­bil­ity is cur­rently un­available.

Another po­ten­tial cause of death is the Sun ex­pand­ing, which could ren­der Earth un­in­hab­it­able in roughly one billion years. Death from this could be pre­vented by coloniz­ing other planets in the so­lar sys­tem, al­though even­tu­ally the sun would ren­der the rest of the so­lar sys­tem un­in­hab­it­able. After this, one could po­ten­tially in­habit other stars; it is ex­pected that stars will re­main for roughly 10 quin­til­lion years, al­though some the­o­ries pre­dict that the uni­verse will be de­stroyed in a mere 20 billion years. To con­tinue sur­viv­ing, one could po­ten­tially go to other uni­verses.64 Ad­di­tion­ally, there are ideas for space-time crys­tals that could pro­cess in­for­ma­tion even af­ter heat death (i.e. the “end of the uni­verse”),65 so per­haps one could make one­self com­posed of the space-time crys­tals via mind up­load­ing or an­other tech­nique. There could also be other meth­ods of sur­viv­ing the con­ven­tional end of the uni­verse, and life could po­ten­tially have 10 quin­til­lion years to find them.

Yet an­other po­ten­tial cause of death is liv­ing in a com­puter simu­la­tion that is ended. The prob­a­bil­ity of one liv­ing in a com­puter simu­la­tion ac­tu­ally seems to not be very im­prob­a­ble. Nick Bostrom ar­gues that: least one of the fol­low­ing propo­si­tions is true: (1) The frac­tion of hu­man-level civ­i­liza­tions that reach a posthu­man stage is very close to zero; (2) The frac­tion of posthu­man civ­i­liza­tions that are in­ter­ested in run­ning an­ces­tor-simu­la­tions is very close to zero; (3) The frac­tion of all peo­ple with our kind of ex­pe­riences that are liv­ing in a simu­la­tion is very close to one.

The ar­gu­ment for this is here.100

If one does die, one could po­ten­tially be re­vived. Cry­on­ics, dis­cussed later in this ar­ti­cle, may help in this. Ad­di­tion­ally, I sup­pose one could pos­si­bly be re­vived if fu­ture in­tel­li­gences con­tinu­ally cre­ate new con­scious in­di­vi­d­u­als and even­tu­ally cre­ate one of them that have one’s “own” con­scious­ness, though con­scious­ness re­mains a mys­tery, so this may not be plau­si­ble, and I’ve found no ar­ti­cles dis­cussing this pos­si­bil­ity. If the prob­a­bil­ity of one’s con­scious­ness be­ing re­vived per unit time does not ap­proach or equal zero as time ap­proaches in­finity, then I sup­pose one is bound to be­come con­scious again, though this sce­nario may be un­likely. Again, I’ve found no ar­ti­cles dis­cussing this pos­si­bil­ity.

As already dis­cussed, in or­der to be live for­ever, one must ei­ther be re­vived af­ter dy­ing or pre­vent death from the con­scious­ness in one’s body not be­ing the same as the one that will be in one’s body in the fu­ture, ac­ci­dents, ag­ing, the sun dy­ing, the uni­verse dy­ing, be­ing in a simu­la­tion and hav­ing it end, and other, un­known, causes. Keep in mind that adding ex­tra de­tails that aren’t guaran­teed to be true can only make events less prob­a­ble, and that peo­ple of­ten don’t ac­count for this.66 A spread­sheet for es­ti­mat­ing one’s chance of liv­ing for­ever is here.

Should we try to be­come im­mor­tal?

Be­fore de­cid­ing whether one should try to be­come im­mor­tal, I sug­gest learn­ing about the cog­ni­tive bi­ases scope in­sen­si­tivity, hy­per­bolic dis­count­ing, and bias blind spot if you don’t know cur­rently know about them. Also, keep in mind that one study found that sim­ply in­form­ing peo­ple of a cog­ni­tive bias made them no less likely to fall prey to it. A study also found that peo­ple only par­tially ad­justed for cog­ni­tive bi­ases af­ter be­ing told that in­form­ing peo­ple of a cog­ni­tive bias made them no less likely to fall prey to it.67

Many ar­ti­cles ar­gu­ing against im­mor­tal­ity are found via a quick google search, in­clud­ing this, this, this, and this. This ar­ti­cle along with its com­ments dis­cusses counter-ar­gu­ments to many of these ar­gu­ments. The Fable of the Dragon Tyrant pro­vides an ar­gu­ment for cur­ing ag­ing, which can be ex­tended to be an ar­gu­ment against mor­tal­ity as a whole. I sug­gest read­ing it.

One can also eval­u­ate the util­ity of im­mor­tal­ity via de­ci­sion the­ory. As­sum­ing in­di­vi­d­u­als re­ceive a finite amount of util­ity per unit time such that it is never less than some above-zero con­stant, liv­ing for­ever would give in­finitely more util­ity than liv­ing for a finite amount of time. Us­ing these as­sump­tions, in or­der to max­i­mize util­ity, one should be will­ing to ac­cept any finite cost to be­come im­mor­tal. How­ever, the situ­a­tion is com­pli­cated when one con­sid­ers the po­ten­tial of be­com­ing im­mor­tal and re­ceiv­ing an in­finite pos­i­tive util­ity un­in­ten­tion­ally, in which case one would re­ceive in­finite ex­pected util­ity re­gard­less of if one tried to be­come im­mor­tal. Ad­di­tion­ally, if one both has the chance of re­ceiv­ing in­finitely high and in­finitely low util­ity, one’s ex­pected util­ity would be un­defined. In­finite util­ities are dis­cussed in “In­finite Ethics” by Nick Bostrom.

For those in­ter­ested in de­creas­ing ex­is­ten­tial risk, liv­ing for a very long time, albeit not nec­es­sar­ily for­ever, may give one more op­por­tu­nity to do so. This idea can be gen­er­al­ized to many goals one has in life.

On whether one can in­fluence one’s chances of be­com­ing im­mor­tal, stud­ies have shown that only roughly 20-30% of longevity in hu­mans is ac­counted for by ge­netic fac­tors.68 There are mul­ti­ple ac­tions one can to in­crease one’s chances of liv­ing for­ever; these are what the rest of this ar­ti­cle is about. Keep in mind that you should con­sider con­tin­u­ing read­ing this ar­ti­cle even if you don’t want to try to be­come im­mor­tal, as the ar­ti­cle pro­vides in­for­ma­tion on liv­ing longer, even if not for­ever, as well.

Rel­a­tive im­por­tance of the differ­ent topics

The figure be­low gives the rel­a­tive fre­quen­cies of pre­ventable causes of death.


Some causes of death are ex­cluded from the graph, but are still large causes of death. Most no­tably, 440,000 deaths in the US, roughly one sixth of to­tal deaths in the US are es­ti­mated to be from pre­ventable med­i­cal er­rors in hos­pi­tals.2

Risk calcu­la­tors for car­dio­vas­cu­lar dis­ease are here and here. Though they seem very sim­plis­tic, they may be worth look­ing at and can prob­a­bly be com­pleted quickly.

Here are the fre­quen­cies of causes of deaths in the US in year 2010 based off of an­other clas­sifi­ca­tion:

  • Heart dis­ease: 596,577

  • Cancer: 576,691

  • Chronic lower res­pi­ra­tory dis­eases: 142,943

  • Stroke (cere­brovas­cu­lar dis­eases): 128,932

  • Ac­ci­dents (un­in­ten­tional in­juries): 126,438

  • Alzheimer’s dis­ease: 84,974

  • Di­a­betes: 73,831

  • In­fluenza and Pneu­mo­nia: 53,826

  • Nephri­tis, nephrotic syn­drome, and nephro­sis: 45,591

  • In­ten­tional self-harm (suicide): 39,518



What to eat and drink

Keep in mind that the re­la­tion­ship be­tween health and the con­sump­tion of types of sub­stances aren’t nec­es­sar­ily lin­ear. I.e. some sub­stances are benefi­cial in small amounts but harm­ful in large amounts, while oth­ers are benefi­cial in both small and large amounts, but con­sum­ing large amounts is no more benefi­cial than con­sum­ing small amounts.

Recom­men­da­tions from The Nutri­tion Source

The Nutri­tion Source is part of the Har­vard School of Public Health.

Its recom­men­da­tions:

  • Make ½ of your “plate” con­sist of a va­ri­ety of fruits and a va­ri­ety of veg­eta­bles, ex­clud­ing pota­toes, due to pota­toes’ nega­tive effect on blood sugar. The Har­vard School of Public Health doesn’t seem to spec­ify if this is based on calories or vol­ume. It also doesn’t ex­plain what it means by plate, but pre­sum­ably ½ of one’s plate means ½ solid food con­sumed.

  • Make ¼ of your plate con­sist of whole grains.

  • Make ¼ of your plate con­sist of high-pro­tein foods.

  • Limit red meat con­sump­tion.

  • Avoid pro­cessed meats.

  • Con­sume mo­noun­sat­u­rated and polyun­sat­u­rated fats in mod­er­a­tion; they are healthy.

  • Avoid par­tially hy­dro­genated oils, which con­tain trans fats, which are un­healthy.

  • Limit milk and dairy prod­ucts to one to two serv­ings per day.

  • Limit juice to one small glass per day.

  • It is im­por­tant to eat seafood one or two times per week, par­tic­u­larly fatty (dark meat) fish that are richer in EPA and DHA.

  • Limit diet drink con­sump­tion or con­sume in mod­er­a­tion.

  • Avoid sug­ary drinks like soda, sports drinks, and en­ergy drinks.3


The bot­tom line is that sat­u­rated fats and es­pe­cially trans fats are un­healthy, while un­sat­u­rated fats are healthy and the types of un­sat­u­rated fats omega-3 and omega-6 fatty acids fats are es­sen­tial. The pro­por­tion of calories from fat in one’s diet isn’t re­ally linked with dis­ease.

Sat­u­rated fat is un­healthy. It’s gen­er­ally a good idea to min­i­mize sat­u­rated fat con­sump­tion. The lat­est Die­tary Guidelines for Amer­i­cans recom­mends con­sum­ing no more than 10% of calories from sat­u­rated fat, but the Amer­i­can Heart As­so­ci­a­tion recom­mends con­sum­ing no more than 7% of calories from sat­u­rated fat. How­ever, don’t de­crease nut, oil, and fish con­sump­tion to min­i­mize sat­u­rated fat con­sump­tion. Foods that con­tain large amounts of sat­u­rated fat in­clude red meat, but­ter, cheese, and ice cream.

Trans fats are es­pe­cially un­healthy. For ev­ery 2% in­crease of calories from trans-fat, risk of coro­nary heart dis­ease in­creases by 23%. The Fed­eral In­sti­tute for Medicine states that there are no known re­quire­ments for trans fats for bod­ily func­tions, so their con­sump­tion should be min­i­mized. Par­tially hy­dro­genated oils con­tain trans fats, and foods that con­tain trans fats are of­ten pro­cessed foods. In the US, prod­ucts can claim to have zero grams of trans fat if they have no more than 0.5 grams of trans fat. Prod­ucts with no more than 0.5 grams of trans fat that still have non-neg­ligible amounts of trans fat will prob­a­bly have the in­gre­di­ents “par­tially hy­dro­genated veg­etable oils” or “veg­etable short­en­ing” in their in­gre­di­ent list.

Un­sat­u­rated fats have benefi­cial effects, in­clud­ing im­prov­ing choles­terol lev­els, eas­ing in­flam­ma­tion, and sta­bi­liz­ing heart rhythms. The Amer­i­can Heart As­so­ci­a­tion has set 8-10% of calories as a tar­get for polyun­sat­u­rated fat con­sump­tion, though eat­ing more polyun­sat­u­rated fat, around 15%of daily calories, in place of sat­u­rated fat may fur­ther lower heart dis­ease risk. Con­sum­ing un­sat­u­rated fats in­stead of sat­u­rated fat also pre­vents in­sulin re­sis­tance, a pre­cur­sor to di­a­betes. Mo­noun­sat­u­rated fats and polyun­sat­u­rated fats are types of un­sat­u­rated fats.

Omega-3 fatty acids (omega-3 fats) are a type of un­sat­u­rated fat. There are two main types: Marine omega-3s and alpha-linolenic acid (ALA). Omega-3 fatty acids, es­pe­cially marine omega-3s, are healthy. Though one can make most needed types of fats from other fats or sub­stances con­sumed, omega-3 fat is an es­sen­tial fat, mean­ing it is an im­por­tant type of fat and can­not be made in the body, so they must come from food. Most amer­i­cans don’t get enough omega-3 fats.

Marine omega-3s are pri­mar­ily found in fish, es­pe­cially fatty (dark mean) fish. A com­pre­hen­sive re­view found that eat­ing roughly two grams per week of omega-3s from fish, equal to about one or two serv­ings of fatty fish per week, de­creased risk of death from heart dis­ease by more than one-third. Though fish con­tain mer­cury, this is in­signifi­cant the pos­i­tive health effects of their con­sump­tion (for the con­sumer, not the fish). How­ever, it does benefit one’s health to con­sult lo­cal ad­vi­sories to de­ter­mine how much lo­cal fresh­wa­ter fish to con­sume.

ALA may be an es­sen­tial nu­tri­ent, and in­creased ALA con­sump­tion may be benefi­cial. ALA is found in veg­etable oils, nuts (es­pe­cially walnuts), flax seeds, flaxseed oil, leafy veg­eta­bles, and some an­i­mal fat, es­pe­cially those from grass-fed an­i­mals. ALA is pri­mar­ily used as en­ergy, but a very small amount of it is con­verted into marine omega-3s. ALA is the most com­mon omega-3 in west­ern diets.

Most Amer­i­cans con­sume much more omega-6 fatty acids (omega-6 fats) than omega-3 fats. Omega-6 fat is an es­sen­tial nu­tri­ent and its con­sump­tion is healthy. Some sources of it in­clude corn and soy­bean oils. The Nutri­tion Sources stated that the the­ory that omega-3 fats are healthier than omega-6 fats isn’t sup­ported by ev­i­dence. How­ever, in an image from the Nutri­tion Source, seafood omega-6 fats were ranked as healthier than plant omega-6 fats, which were ranked as healthier than mo­noun­sat­u­rated fats, al­though such a rank­ing was to the best of my knowl­edge never stated in the text.3


There seems to be two main de­ter­mi­nants of car­bo­hy­drate sources’ effects on health: nu­tri­tion con­tent and effect on blood sugar. The bot­tom line is that con­sum­ing whole grains and other less pro­cessed grains and de­creas­ing re­fined grain con­sump­tion im­proves health. Ad­di­tion­ally, mod­er­ately low car­bo­hy­drate diets can in­crease heart health as long as pro­tein and fat comes from health sources, though the type of car­bo­hy­drate at least as im­por­tant as the amount of car­bo­hy­drates in a diet.

Glycemic in­dex and is a mea­sure of how much food in­creases blood sugar lev­els. Con­sum­ing car­bo­hy­drates that cause blood-sugar spikes can in­crease risk of heart dis­ease and di­a­betes at least as much as con­sum­ing too much sat­u­rated fat does. Some fac­tors that in­crease the glycemic in­dex of foods in­clude:

  • Be­ing a re­fined grain as op­posed to a whole grain.

  • Be­ing finely ground, which is why con­sum­ing whole grains in their whole form, such as rice, can be healthier than con­sum­ing them as bread.

  • Hav­ing less fiber.

  • Be­ing more ripe, in the case of fruits and veg­eta­bles.

  • Hav­ing a lower fat con­tent, as meals with fat are con­verted more slowly into sugar.

Vegeta­bles (ex­clud­ing pota­toes), fruits, whole grains, and beans, are healthier than other car­bo­hy­drates. Po­ta­toes have a nega­tive effect on blood sugar, due to their high glycemic in­dex. In­for­ma­tion on glycemic in­dex and the in­dex of var­i­ous foods is here.

Whole grains also con­tain es­sen­tial min­er­als such as mag­ne­sium, se­le­nium, and cop­per, which may pro­tect against some can­cers. Refin­ing grains takes away 50% of the grains’ B vi­tam­ins, 90% of vi­tamin E, and vir­tu­ally all fiber. Su­gary drinks usu­ally have lit­tle nu­tri­tional value.

Iden­ti­fy­ing whole grains as food that has at least one gram of fiber for ev­ery gram of car­bo­hy­drate is a more effec­tive mea­sure of health­ful­ness than iden­ti­fy­ing a whole grain as the first in­gre­di­ent, any whole grain as the first in­gre­di­ent with­out added sug­ars in the first 3 in­gre­di­ents, the word “whole” be­fore any grain in­gre­di­ent, and the whole grain stamp.3


Proteins are bro­ken down to form amino acids, which are needed for health. Though the body can make some amino acids by mod­ify­ing oth­ers, some must come from food, which are called es­sen­tial amino acids. The in­sti­tute of medicine recom­mends that adults get a min­i­mum of 0.8 grams of pro­tein per kilo­gram of body weight per day, and sets the range of ac­cept­able pro­tein in­take to 10-35% of calories per day. The In­sti­tute of Medicine recom­mends get­ting 10-35% of calories from pro­tein each day. The US recom­mended daily al­lowance for pro­tein is 46 grams per day for women over 18 and 56 grams per day for men over 18.

An­i­mal prod­ucts tend to give all es­sen­tial amino acids, but other sources lack some es­sen­tial amino acids. Thus, veg­e­tar­i­ans need to con­sume a va­ri­ety of sources of amino acids each day to get all needed types. Fish, chicken, beans, and nuts are healthy pro­tein sources.3


There are two types of fiber: sol­u­ble fiber and in­sol­u­ble fiber. Both have im­por­tant health benefits, so one should eat a va­ri­ety of foods to get both.94 The best sources of fiber are whole grains, fresh fruits and veg­eta­bles, legumes, and nuts.3


There are many micronu­tri­ents in food; get­ting enough of them is im­por­tant. Most healthy in­di­vi­d­u­als can get suffi­cient micronu­tri­ents by con­sum­ing a wide va­ri­ety of healthy foods, such as fruits, veg­eta­bles, whole grains, legumes, and lean meats and fish. How­ever, sup­ple­men­ta­tion may be nec­es­sary for some. In­for­ma­tion about sup­ple­ments is here.110

Con­cern­ing sup­ple­men­ta­tion, potas­sium, io­dine, and lithium sup­ple­men­ta­tion are recom­mended in the first-place en­try in the Quan­tified Health Prize, a con­test on de­ter­min­ing good min­eral in­take lev­els. How­ever, oth­ers sug­gest that potas­sium sup­ple­men­ta­tion isn’t nec­es­sar­ily benefi­cial, as shown here. I’m some­what skep­ti­cal that the sup­ple­ments are benefi­cial, as I have not found other sources recom­mend­ing their sup­ple­men­ta­tion. The sug­gested sup­ple­men­ta­tion lev­els are in the en­try.

Note that food pro­cess­ing typ­i­cally de­creases micronu­tri­ent lev­els, as de­scribed here. In gen­eral, it seems cook­ing, drain­ing and dry­ing foods siz­ably, tak­ing po­ten­tially half of nu­tri­ents away, while freez­ing and re­heat­ing take away rel­a­tively few nu­tri­ents.111

One micronu­tri­ent worth dis­cussing is sodium. Some sodium is needed for health, but most Amer­i­cans con­sume more sodium than needed. How­ever, recom­men­da­tions on ideal sodium lev­els vary. The US gov­ern­ment recom­mends limit­ing sodium con­sump­tion to 2,300mg/​day (one tea­spoon). The Amer­i­can Heart As­so­ci­a­tion recom­mends limit­ing sodium con­sump­tion to 1,500mg/​day (⅔ of a tea­spoon), es­pe­cially for those who are over 50, have high or ele­vated blood pres­sure, have di­a­betes, or are Afri­can Amer­i­cans3 How­ever, As RomeoStevens pointed out, the In­sti­tute of Medicine found that there’s in­con­clu­sive ev­i­dence that de­creas­ing sodium con­sump­tion be­low 2,300mg/​day effects mor­tal­ity,115 and some meta-analy­ses have sug­gested that there is a U-shaped re­la­tion­ship be­tween sodium and mor­tal­ity.116, 117

Vi­tamin D is an­other micronu­tri­ent that’s im­por­tant for health. It can be ob­tained from food or made in the body af­ter sun ex­po­sure. Most peo­ple who live farther north than San Fran­cisco or don’t go out­side at least fif­teen min­utes when it’s sunny are vi­tamin D defi­cient. Vi­tamin D defi­ciency is in­creases the risk of many chronic dis­eases in­clud­ing heart dis­ease, in­fec­tious dis­eases, and some can­cers. How­ever, there is con­tro­versy about op­ti­mal vi­tamin D in­take. The In­sti­tute of medicine recom­mends get­ting 600 to 4000 IU/​day, though it ac­knowl­edged that there was no good ev­i­dence of harm at 4000 IU/​day. The Nutri­tion Sources states that these recom­men­da­tions are too low and fail to ac­count for new ev­i­dence. The nu­tri­tion source states that for most peo­ple, sup­ple­ments are the best source of vi­tamin D, but most mul­ti­vi­tam­ins have too lit­tle vi­tamin D in them. The Nutri­tion Source recom­mends con­sid­er­ing and talk­ing to a doc­tor about tak­ing an ad­di­tional mul­ti­vi­tamin if the you take less than 1000 IU of vi­tamin D and es­pe­cially if you have lit­tle sun ex­po­sure.3

Blood pressure

In­for­ma­tion on blood pres­sure is here in the sec­tion ti­tled “Blood Pres­sure.”

Choles­terol and triglycerides

In­for­ma­tion on op­ti­mal amounts of choles­terol and triglyc­erides are here.

The biggest in­fluences on choles­terol are fats and car­bo­hy­drates in one’s diet, and choles­terol con­sump­tion gen­er­ally has a far weaker in­fluence. How­ever, some peo­ple’s choles­terol lev­els rise and fall very quickly with the amount of choles­terol con­sumed. For them, de­creas­ing choles­terol con­sump­tion from food can have a con­sid­er­able effect on choles­terol lev­els. Trial and er­ror is cur­rently the only way of de­ter­min­ing if one’s choles­terol lev­els risk and fall very quickly with the amount of choles­terol con­sumed.


De­spite their ini­tial hype, ran­dom­ized con­trol­led tri­als have offered lit­tle sup­port for the benefit is sin­gle an­tiox­i­dants, though stud­ies are in­con­clu­sive.3

Die­tary refer­ence intakes

For the nu­mer­i­cally in­clined, the Die­tary Refer­ence In­take pro­vides quan­ti­ta­tive guidelines on good nu­tri­ent con­sump­tion amounts for many nu­tri­ents, though it may be harder to use for some, due to its quan­ti­ta­tive na­ture.


The Nutri­tion Source and SFGate state that wa­ter is the best drink,3, 112 though I don’t know why it’s con­sid­ered healthier than drinks such as tea.

Un­sweet­ened tea de­creases the risk of many dis­eases, likely largely due to polyphe­nols, and an­tiox­i­dant, in it. De­spite an­tiox­i­dants typ­i­cally hav­ing lit­tle ev­i­dence of benefit, I sup­pose polyphe­nols are rel­a­tively benefi­cial. All teas have roughly the same lev­els of polyphe­nols ex­cept de­caf­feinated tea,3 which has fewer polyphe­nols.96 Re­search sug­gests that pro­teins and pos­si­bly fat in milk de­crease the an­tiox­i­dant ca­pac­ity of tea.

It’s con­sid­ered safe to drink up to six cups of coffee per day. Un­sweet­ened coffee is healthy and may de­crease some dis­ease risks, though coffee may slightly in­crease blood pres­sure. Some peo­ple may want to con­sider avoid­ing coffee or switch­ing to de­caf, es­pe­cially women who are preg­nant or peo­ple who have a hard time con­trol­ling their blood pres­sure or blood sugar. The nu­tri­tion source states that it’s best to brew coffee with a pa­per filter to re­move a sub­stance that in­creases LDL choles­terol, de­spite con­sumed choles­terol typ­i­cally hav­ing a very small effect on the body’s choles­terol level.

Al­co­hol in­creases risk of dis­eases for some peo­ple3 and de­creases it for oth­ers.3, 119 Heavy al­co­hol con­sump­tion is a ma­jor cause of pre­ventable death in most coun­tries. For some groups of peo­ple, es­pe­cially preg­nant peo­ple, peo­ple re­cov­er­ing from al­co­hol ad­dic­tion, and peo­ple with liver dis­ease, al­co­hol causes greater health risks and should be avoided. The like­li­hood of be­com­ing ad­dicted to al­co­hol can be ge­net­i­cally de­ter­mined. Moder­ate drink­ing, gen­er­ally defined as no more than one or two drinks per day for men, can in­crease colon and breast can­cer risk, but these effects are offset by de­creased heart dis­ease and di­a­betes risk, es­pe­cially in mid­dle age, where heart dis­ease be­gins to ac­count for an in­creas­ingly large pro­por­tion of deaths. How­ever, al­co­hol con­sump­tion won’t de­crease car­dio­vas­cu­lar dis­ease risk much for those who are thin, phys­i­cally ac­tive, don’t smoke, eat a healthy diet, and have no fam­ily his­tory of heart dis­ease. Some re­search sug­gests that red wine, par­tic­u­larly when con­sumed af­ter a meal, has more car­dio­vas­cu­lar benefits than beers or spirits, but al­co­hol choice has still lit­tle effect on dis­ease risk. In one study, mod­er­ate drinkers were 30-35% less likely to have heart at­tacks than non-drinkers and men who drank daily had lower heart at­tack risk than those who drank once or twice per week.

There’s no need to drink more than one or two glasses of milk per day. Less milk is fine if cal­cium is ob­tained from other sources.

The health effects of ar­tifi­cially sweet­ened drinks are largely un­known. Oddly, they may also cause weight gain. It’s best to limit con­sum­ing them if one drinks them at all.

Su­gary drinks can cause weight gain, as they aren’t as filling as solid food and have high sugar. They also in­crease the risk of di­a­betes, heart dis­ease, and other dis­eases. Fruit juice has more calories and less fiber than whole fruit and is re­port­edly no bet­ter than soft drinks.3

Solid food

Fruits and veg­eta­bles are an im­por­tant part of a healthy diet. Eat­ing a va­ri­ety of them is as im­por­tant as eat­ing many of them.3 Fish and nut con­sump­tion is also very healthy.98

Pro­cessed meat, on the other hand, is shock­ingly bad.98 A meta-anal­y­sis found that pro­cessed meat con­sump­tion is as­so­ci­ated with a 42% in­creased risk of coro­nary heart dis­ease (rel­a­tive risk per 50g serv­ing per day; 95% con­fi­dence in­ter­val: 1.07 − 1.89) and 19% in­creased risk of di­a­betes.97 De­spite this, a bit of red meat con­sump­tion has been found to be benefi­cial.98 Con­sump­tion of well-done, fried, or bar­be­cued meat has been as­so­ci­ated with cer­tain can­cers, pre­sum­ably due to car­cino­gens made in the meat from be­ing cooked, though this link isn’t defini­tive. The amount of car­cino­gens in­creases with in­creased cook­ing tem­per­a­ture (es­pe­cially above 300ºF, in­creased cook­ing time, char­ring, or be­ing ex­posed to smoke.99

Eat­ing less than one egg per day doesn’t in­crease heart dis­ease risk in healthy in­di­vi­d­u­als and can be part of a healthy diet.3

Or­ganic foods have lower lev­els of pes­ti­cides than in­or­ganic foods, though the resi­dues of most or­ganic and in­or­ganic prod­ucts don’t ex­ceed gov­ern­ment safety thresh­old. Wash­ing fresh fruits and veg­eta­bles in recom­mended, as it re­moves bac­te­ria and some, though not all, pes­ti­cide resi­dues. Or­ganic foods prob­a­bly aren’t more nu­tri­tious than non-or­ganic foods.103

When to eat and drink

A ran­dom­ized con­trol­led trial found an in­crease in blood sugar vari­a­tion for sub­jects who skipped break­fast.6 In­creas­ing meal fre­quency and de­creas­ing meal size ap­pears to have some metabolic ad­van­tages, and doesn’t ap­pear to have metabolic dis­ad­van­tages.7 Note: old source; made in 1994 How­ever, Mayo Clinic states that fast­ing for 1-2 days per week may in­crease heart health.32 Per­haps it is op­ti­mal for health to fast, but to have high meal fre­quency when not fast­ing.

How much to eat

One’s weight gain is di­rectly pro­por­tional to the num­ber of calories con­sumed di­vided by the num­ber of calories burnt. Cen­ters for Disease Con­trol and Preven­tion (CDC) has guidelines for healthy weights and in­for­ma­tion on how to lose weight.

Some ad­vo­cate re­strict­ing weight to a greater ex­tent, which is known as calorie re­stric­tion. It’s un­known whether calorie re­stric­tion in­creases lifes­pan in hu­mans or not, but mod­er­ate calorie re­stric­tion with ad­e­quate nu­tri­tion de­creases risk of obe­sity, type 2 di­a­betes, in­flam­ma­tion, hy­per­ten­sion, car­dio­vas­cu­lar dis­ease, and metabolic risk fac­tors as­so­ci­ated with can­cer, and is the most effec­tive way of con­sis­tently in­creas­ing lifes­pan in a va­ri­ety of or­ganisms. The CR So­ciety has in­for­ma­tion on get­ting started on calorie re­stric­tion.4

How much to drink

Gen­er­ally, drink­ing enough to rarely feel thirsty and to have col­or­less or light yel­low urine is usu­ally suffi­cient. It’s also pos­si­ble to drink too much wa­ter. In gen­eral, drink­ing too much wa­ter is rare in healthy adults who eat an av­er­age Amer­i­can diet, al­though en­durance ath­letes are at a higher risk.10


A meta-anal­y­sis found the data in the fol­low­ing graphs for peo­ple aged over 40.


A weekly to­tal of roughly five hours of vi­gor­ous ex­er­cise has been iden­ti­fied by sev­eral stud­ies to be the safe up­per limit for life ex­pec­tancy. It may be benefi­cial to take one or two days off from vi­gor­ous ex­er­cise per week and to limit chronic vi­gor­ous ex­er­cise to ⇐ 60 min/​day.9 Based on the above, I my best guess for the op­ti­mal amount of ex­er­cise for longevity is roughly 30 MET-hr/​wk. Cal­is­then­ics burn 6-10 METs/​hr11, so an ex­am­ple ex­er­cise rou­tine to get this amount of ex­er­cise is do­ing cal­is­then­ics 38 min­utes per day and 6 days/​wk. Guides on how to ex­er­cise are available, e.g. this one.


Car­cino­gens are can­cer-caus­ing sub­stances. Since can­cer causes death, de­creas­ing ex­po­sure to car­cino­gens pre­sum­ably de­creases one’s risk of death. Some foods are also car­cino­genic, as dis­cussed in the “Food” sec­tion.


Tobacco use is the great­est avoid­able risk fac­tor for can­cer wor­ld­wide, caus­ing roughly 22% of can­cer deaths. Ad­di­tion­ally, sec­ond hand smoke has been proven to cause lung can­cer in non­smok­ing adults.

Al­co­hol use is a risk fac­tor for many types of can­cer. The risk of can­cer in­creases with the amount of al­co­hol con­sumed, and sub­stan­tially in­creases if one is also a heavy smoker. The at­tributable frac­tion of can­cer from al­co­hol use varies de­pend­ing on gen­der, due to differ­ences in con­sump­tion level. E.g. 22% of mouth and orophar­ynx can­cer is at­tributable to can­cer in men but only 9% is at­tributable to al­co­hol in women.

En­vi­ron­men­tal air pol­lu­tion ac­counts for 1-4% of can­cer.84 Die­sel ex­haust is one type of car­cino­genic air pol­lu­tion. Those with the high­est ex­po­sure to diesel ex­haust are ex­posed to it oc­cu­pa­tion­ally. As for res­i­den­tial ex­po­sure, diesel ex­haust is high­est in homes near roads where traf­fic is heav­iest. Limit­ing time spent near large sources of diesel ex­haust de­creases ex­po­sure. Ben­zene, an­other car­cino­gen, is found in gasoline and ve­hi­cle ex­haust but ex­po­sure to it can also be cause by be­ing in ar­eas with un­ven­tilated fumes from gasoline, glues, solvents, paints, and art sup­plies. It can cause ex­po­sure from in­hala­tion or skin con­tact.86

Some oc­cu­pa­tions ex­po­sure work­ers to oc­cu­pa­tional car­cino­gens.84 A list of some of the oc­cu­pa­tions is here, all of which in­volve man­ual la­bor, ex­cept for hos­pi­tal-re­lated jobs.87


In­fec­tions are re­spon­si­ble for 6% of can­cer deaths in de­vel­oped na­tions.84 Many of the in­fec­tions are spread via sex­ual con­tact and shar­ing nee­dles and some can be vac­ci­nated against.85


Ioniz­ing ra­di­a­tion is car­cino­genic to hu­mans. Res­i­den­tial ex­po­sure to radon gas is es­ti­mated to cause 3-14% of lung can­cers, which is the largest source of radon ex­po­sure for most peo­ple 84 Be­ing ex­posed to radon and cigarette smoke to­gether in­creases one’s can­cer risk much more than they do sep­a­rately. There is much vari­a­tion radon lev­els de­pend­ing on where one lives and and radon is usu­ally higher in­side build­ings, es­pe­cially lev­els closer to the ground, such as base­ments. The EPA recom­mends tak­ing ac­tion to re­duce radon lev­els if they are greater than or equal to 4.0 pCi/​L. Radon lev­els can be re­duced by a qual­ified con­trac­tor. Re­duc­ing radon lev­els with­out proper train­ing and equip­ment can in­crease in­stead of de­crease them.88

Some med­i­cal tests can also in­crease ex­po­sure to ra­di­a­tion. The EPA es­ti­mates that ex­po­sure to 10 mSv from a med­i­cal imag­ing test in­creases risk of can­cer by roughly 0.05%. To de­crease ex­po­sure to ra­di­a­tion from med­i­cal imag­ing tests, one can ask if there are ways to shield parts of one’s body from ra­di­a­tion that aren’t be­ing tested and mak­ing sure the doc­tor perform­ing the test is qual­ified.89

Small doses of ioniz­ing ra­di­a­tion in­crease risk by a very small amount. Most stud­ies haven’t de­tected in­creased can­cer risk in peo­ple ex­posed to low lev­els of ioniz­ing ra­di­a­tion. For ex­am­ple, peo­ple liv­ing in higher al­ti­tudes don’t have no­tice­ably higher can­cer rates than other peo­ple. In gen­eral, can­cer risk from ra­di­a­tion in­creases as the dose of ra­di­a­tion in­creases and there is thought to be no safe level of ex­po­sure. Ul­travi­o­let ra­di­a­tion as a type of ra­di­a­tion that can be ioniz­ing ra­di­a­tion. Sun­light is the main source of ul­tra­vi­o­let ra­di­a­tion.84

Fac­tors that in­crease one’s ex­po­sure to ul­tra­vi­o­let ra­di­a­tion when out­side in­clude:

  • Time of day. Al­most ⅓ of UV ra­di­a­tion hits the sur­face be­tween 11AM and 1PM, and ¾ hit the sur­face be­tween 9AM and 5PM.

  • Time of year. UV ra­di­a­tion is greater dur­ing sum­mer. This fac­tor is less sig­nifi­cant near the equa­tor.

  • Alti­tude. High ele­va­tion causes more UV ra­di­a­tion to pen­e­trate the at­mo­sphere.

  • Clouds. Some­times clouds de­crease lev­els of UV ra­di­a­tion be­cause they block UV ra­di­a­tion from the sun. Other times, they in­crease ex­po­sure be­cause they re­flect UV ra­di­a­tion.

  • Reflec­tion off sur­faces, such as wa­ter, sand, snow, and grass in­creases UV ra­di­a­tion.

  • Ozone den­sity, be­cause ozone stops some UV ra­di­a­tion from reach­ing the sur­face.

Some tips to de­crease ex­po­sure to UV ra­di­a­tion:

  • Stay in the shade. This is one of the best ways to limit ex­po­sure to UV ra­di­a­tion in sun­light.

  • Cover your­self with cloth­ing.

  • Wear sun­glasses.

  • Use sun­screen on ex­posed skin.90

Tan­ning beds are also a source of ul­tra­vi­o­let ra­di­a­tion. Us­ing tan­ning booths can in­crease one’s chance of get­ting skin melanoma by at least 75%.91

Vi­tamin D3 is also pro­duced from ul­tra­vi­o­let ra­di­a­tion, al­though the Amer­i­can So­ciety for Clini­cal Nutri­tion states that vi­tamin D is read­ily available from sup­ple­ments and that the con­tro­versy about re­duc­ing ul­tra­vi­o­let ra­di­a­tion ex­po­sure was fueled by the tan­ning in­dus­try.92

There could be some risk of cell phone use be­ing as­so­ci­ated with can­cer, but the ev­i­dence is not strong enough to be con­sid­ered causal and needs to be in­ves­ti­gated fur­ther.93, 118

Emo­tions and feelings

Pos­i­tive emo­tions and feelings

A re­view sug­gested that pos­i­tive emo­tions and feel­ings de­creased mor­tal­ity. Pro­posed mechanisms in­clude pos­i­tive emo­tions and feel­ings be­ing as­so­ci­ated with bet­ter health prac­tices such as im­proved sleep qual­ity, in­creased ex­er­cise, and in­creased dietary zinc con­sump­tion, as well as lower lev­els of some stress hor­mones. It has also been hy­poth­e­sized to be as­so­ci­ated with other health-rele­vant hor­mones, var­i­ous as­pects of im­mune func­tion, and closer and more so­cial con­tacts.33 Less Wrong has a good ar­ti­cle on how to be happy.

Psy­cholog­i­cal distress

A meta-anal­y­sis was con­ducted on psy­cholog­i­cal stress. To mea­sure psy­cholog­i­cal stress, it used the GHQ-12 score, which mea­sured symp­toms of anx­iety, de­pres­sion, so­cial dys­func­tion, and loss of con­fi­dence. The scores range from 0 to 12, with 0 be­ing asymp­tomatic, 1-3 be­ing sub­clini­cally symp­tomatic, 4-6 be­ing symp­tomatic, and 7-12 be­ing highly symp­tomatic. It found the re­sults shown in the fol­low­ing graphs.

This as­so­ci­a­tion was es­sen­tially un­changed af­ter con­trol­ling for a range of co­vari­ates in­clud­ing oc­cu­pa­tional so­cial class, al­co­hol in­take, and smok­ing. How­ever, re­verse causal­ity may still partly ex­plain the as­so­ci­a­tion.30


A study found that in­di­vi­d­u­als with mod­er­ate and high stress lev­els as op­posed to low stress had haz­ard ra­tios (HRs) of mor­tal­ity of 1.43 and 1.49, re­spec­tively.27 A meta-anal­y­sis found that high per­ceived stress as op­posed to low per­ceived stress had a coro­nary heart dis­ease rel­a­tive risk (RR) of 1.27. The mean age of par­ti­ci­pants in the stud­ies used in the meta-anal­y­sis varied from 44 to 72.5 years and was sig­nifi­cantly and pos­i­tively as­so­ci­ated with effect size. It ex­plained 46% of the var­i­ance in effect sizes be­tween the stud­ies used in the meta-anal­y­sis.28

A cross-sec­tional study (which is a rel­a­tively weak study de­sign) not in the afore­men­tioned meta-anal­y­sis used 28,753 sub­jects to study the effect on mor­tal­ity from the amount of stress and the per­cep­tion of whether stress is harm­ful or not. It found that nei­ther of these fac­tors pre­dicted mor­tal­ity in­de­pen­dently, but but that taken to­gether, they did have a statis­ti­cally sig­nifi­cant effect. Sub­jects who re­ported much stress and that stress has a large effect on health had a HR of 1.43 (95% CI: 1.2, 1.7). Re­v­erse causal­ity may par­tially ex­plain this though, as those who have had nega­tive health im­pacts from stress may have been more likely to re­port that stress in­fluences health.83

Anger and hostility

A meta-anal­y­sis found that af­ter fully con­trol­ling for be­hav­ior co­vari­ates such as smok­ing, phys­i­cal ac­tivity or body mass in­dex, and so­cioe­co­nomic sta­tus, anger and hos­tility was not as­so­ci­ated with coro­nary heart dis­ease (CHD), though the re­sults are in­con­clu­sive.34

So­cial and per­son­al­ity factors

So­cial status

A re­view sug­gested that so­cial sta­tus is linked to health via gen­der, race, eth­nic­ity, ed­u­ca­tion lev­els, so­cioe­co­nomic differ­ences, fam­ily back­ground, and old age.46

Giv­ing to others

An ob­ser­va­tional study found that stress­ful life events was not a pre­dic­tor for mor­tal­ity for those who en­gaged in un­paid helping be­hav­ior di­rected to­wards friends, neigh­bors, or rel­a­tives who did not live with them. This as­so­ci­a­tion may be due to giv­ing to oth­ers caus­ing one to have a sense of mat­ter­ing, op­por­tu­ni­ties for gen­er­a­tivity, im­proved so­cial well-be­ing, the emo­tional state of com­pas­sion, and the phys­iol­ogy of the care­giv­ing be­hav­ioral sys­tem.35

So­cial relationships

A large meta-anal­y­sis found that the odds ra­tio of mor­tal­ity of hav­ing weak so­cial re­la­tion­ships is 1.5 (95% con­fi­dence in­ter­val (CI): 1.42 to 1.59). How­ever, this effect may be a con­ser­va­tive es­ti­mate. Many of the stud­ies used in the meta-anal­y­sis used sin­gle item mea­sures of so­cial re­la­tions, but the size of the as­so­ci­a­tion was great­est in stud­ies that used more com­plex mea­sure­ments. Ad­di­tion­ally, some of the stud­ies in the meta-anal­y­sis ad­justed for risk fac­tors that may be me­di­a­tors of so­cial re­la­tion­ships’ effect on mor­tal­ity (e.g. be­hav­ior, diet, and ex­er­cise). Many of the stud­ies in the meta-anal­y­sis also ig­nored the qual­ity of so­cial re­la­tion­ships, but re­search sug­gests that nega­tive so­cial re­la­tion­ships are linked to in­creased mor­tal­ity. Thus, the effect of so­cial re­la­tion­ships on mor­tal­ity could be even greater than the study found.

Con­cern­ing cau­sa­tion, so­cial re­la­tion­ships are linked to bet­ter health prac­tices and psy­cholog­i­cal pro­cesses, such as stress and de­pres­sion, which in­fluence health out­comes on their own. How­ever, the meta-anal­y­sis also states that so­cial re­la­tion­ships ex­ert an in­de­pen­dent effect. Some stud­ies show that so­cial sup­port is linked to bet­ter im­mune sys­tem func­tion­ing and to im­mune-me­di­ated in­flam­ma­tory pro­cesses.36


A co­hort study with 468 deaths found that each 1 stan­dard de­vi­a­tion de­crease in con­scien­tious­ness was as­so­ci­ated with HR be­ing mul­ti­plied by 1.07 (95% CI: 0.98 – 1.17), though it gave no mechanism for the as­so­ci­a­tion.39 Although it ad­justed for sev­eral vari­ables, (e.g. so­cioe­co­nomic sta­tus, smok­ing, and drink­ing), it didn’t ad­just for drug use, risky driv­ing, risky sex, suicide, and vi­o­lence, which were all found by a meta-anal­y­sis to have statis­ti­cally sig­nifi­cant as­so­ci­a­tions with con­scien­tious­ness.40 Over­all, it seems to me that con­scien­tious­ness doesn’t seem to have a sig­nifi­cant effect on mor­tal­ity.

In­fec­tious diseases

Mayo clinic has a good ar­ti­cle on pre­vent­ing in­fec­tious dis­ease.

Den­tal health

A co­hort study of 5611 adults found that com­pared to men with 26-32 teeth, men with 16-25 teeth had an HR of 1.03 (95% CI: 0.91-1.17), men with 1-15 teeth had an HR of 1.21 (95% CI: 1.05-1.40) and men with 0 teeth had an HR of 1.18 (95% CI: 1.00-1.39).

In the study, men who never brushed their teeth at night had a HR of 1.34 (95% CI: 1.14-1.57) rel­a­tive to those who did ev­ery night. Among sub­jects who brushed at night, HR was similar be­tween those who did and didn’t brush daily in the morn­ing or day. The HR for men who brushed in the morn­ing ev­ery day but not at night ev­ery day was 1.19 (95% CI: 0.99-1.43).

In the study, men who never used den­tal floss had an HR of 1.27 (95% CI: 1.11-1.46) and those who some­times used it had an HR or 1.14 (95% CI: 1.00-1.30) com­pared to men who used it ev­ery day. Among sub­jects who brushed their teeth at night daily, not floss­ing was as­so­ci­ated with a sig­nifi­cantly in­creased HR.

Use of tooth­picks didn’t sig­nifi­cantly de­crease HR and mouth­wash had no effect.

The study had a list of other stud­ies on the effect of den­tal health on mor­tal­ity. It seems to us that al­most all of them found a nega­tive cor­re­la­tion be­tween den­tal health and risk of mor­tal­ity, al­though the study didn’t say their method­ol­ogy for se­lect­ing the stud­ies to show. I did a crude re­view of other liter­a­ture by only look­ing at their ab­stracts and found that five stud­ies found that poor den­tal health in­creased risk of mor­tal­ity and one found it didn’t.

Re­gard­ing pos­si­ble mechanisms, the study says that tooth­paste helps pre­vent den­tal caries and that den­tal floss is the most effec­tive means of re­mov­ing in­ter­den­tal plaque and de­creas­ing in­ter­den­tal gin­gi­val in­flam­ma­tion.38


It seems that get­ting too lit­tle or too much sleep likely in­creases one’s risk of mor­tal­ity, but it’s hard to tell ex­actly how much is too much and how lit­tle is too lit­tle.

One re­view found that the as­so­ci­a­tion be­tween amount of sleep and mor­tal­ity is in­con­sis­tent in stud­ies and that what as­so­ci­a­tion does ex­ist may be due to re­verse-causal­ity.41 How­ever, a meta-anal­y­sis found that the RR as­so­ci­ated with short sleep du­ra­tion (var­i­ously defined as sleep­ing from < 8 hrs/​night to < 6 hrs/​night) was 1.10 (95% CI: 1.06-1.15). It also found that the RR as­so­ci­ated with long sleep du­ra­tion (var­i­ously defined as sleep­ing for > 8 hrs/​night to > 10 hrs per night) com­pared with medium sleep du­ra­tion (var­i­ously defined as sleep­ing for 7-7.9 hrs/​night to 9-9.9 hrs/​night) was 1.23 (95% CI: 1.17 − 1.30).42

The Na­tional Heart, Lung, and Blood In­sti­tute and Mayo Clinic recom­mend adults get 7-8 hours of sleep per night, al­though it also says sleep needs vary from per­son to per­son. It gives no method of de­ter­min­ing op­ti­mal sleep for an in­di­vi­d­ual. Ad­di­tion­ally, it doesn’t say if its recom­men­da­tions are for op­ti­mal longevity, op­ti­mal pro­duc­tivity, some­thing else, or a com­bi­na­tion of fac­tors.43 The Har­vard Med­i­cal School im­plies that one’s op­ti­mal amount of sleep is enough sleep to not need an alarm to wake up, though it didn’t spec­ify the crite­ria for de­ter­min­ing op­ti­mal­ity ei­ther.45


None of the drugs I’ve looked into have a benefi­cial effect for the peo­ple with­out a spe­cial dis­ease or risk fac­tor. Notes on them are here.

Blood donation

A quasi-ran­dom­ized ex­per­i­ment with a val­idity near that of a ran­dom­ized trial pre­sum­ably sug­gested that blood dona­tion didn’t sig­nifi­cantly de­crease risk of coro­nary heart dis­ease (CHD). Ob­ser­va­tional stud­ies have shown much lower CHD in­ci­dence among donors, al­though the au­thors of the former ex­per­i­ment sus­pect that bias and re­verse cau­sa­tion played a role in this.29 That said, a re­view found that re­verse cau­sa­tion ac­counted for only 30% of the effect of blood dona­tion, though I haven’t been able to find the re­view. RomeoStevens sug­gests that the po­ten­tial benefits of blood dona­tion are high enough and the costs are low enough that blood dona­tion is worth do­ing.120


After ad­just­ing for amount of phys­i­cal ac­tivity, a meta-anal­y­sis es­ti­mated that for ev­ery one hour in­cre­ment of sit­ting in in­ter­vals 0-3, >3-7 and >7 h/​day to­tal sit­ting time, the haz­ard ra­tios of mor­tal­ity were 1.00 (95% CI: 0.98-1.03), 1.02 (95% CI: 0.99-1.05) and 1.05 (95% CI: 1.02-1.08) re­spec­tively. It pro­posed no mechanism for sit­ting time hav­ing this effect,37 so it might have been due to con­found­ing vari­ables it didn’t con­trol.

Sleep apnea

Sleep ap­nea is an in­de­pen­dent risk fac­tor for mor­tal­ity and car­dio­vas­cu­lar dis­ease.26 Symp­toms and other in­for­ma­tion on sleep ap­nea are here.


A meta-anal­y­sis found that self-re­ported ha­bit­ual snor­ing had a small but statis­ti­cally sig­nifi­cant as­so­ci­a­tion with stroke and coro­nary heart dis­ease, but not with car­dio­vas­cu­lar dis­ease and all-cause mor­tal­ity [HR 0.98 (95% CI: 0.78-1.23)]. Whether the risk is due to ob­struc­tive sleep ap­nea is con­tro­ver­sial. Only the ab­stract is able to be viewed for free, so I’m just bas­ing this off the ab­stract.31


The or­ga­ni­za­tion Su­san G. Komen, cit­ing a meta-anal­y­sis that used ran­dom­ized con­trol­led tri­als, doesn’t recom­mend breast self ex­ams as a screen­ing tool for breast can­cer, as it hasn’t been shown to de­crease can­cer death. How­ever, it still stated that it is im­por­tant to be fa­mil­iar with one’s breasts’ ap­pear­ance and how they nor­mally feel.49 Ac­cord­ing to the Me­mo­rial Sloan Ket­ter­ing Cancer Cen­ter, no study has been able to show a statis­ti­cally sig­nifi­cant de­crease in breast can­cer deaths from breast self-ex­ams.50 The Na­tional Cancer In­sti­tute states that breast self-ex­am­i­na­tions haven’t been shown to de­crease breast can­cer mor­tal­ity, but does in­crease biop­sies of be­nign breast le­sions.51

The Amer­i­can Cancer So­ciety doesn’t recom­mend tes­tic­u­lar self-ex­ams for all men, as they haven’t been stud­ied enough to de­ter­mine if they de­crease mor­tal­ity. How­ever, it states that men with risk fac­tors of tes­tic­u­lar can­cer (e.g. an un­de­scended tes­ti­cal, pre­vi­ous tes­tic­u­lar can­cer, of a fam­ily mem­ber who pre­vi­ously had tes­tic­u­lar can­cer) should con­sider self-ex­ams and dis­cuss them with a doc­tor. The Amer­i­can Cancer So­ciety also recom­mends hav­ing tes­tic­u­lar self-ex­ams in rou­tine can­cer-re­lated check-ups.52


Ge­nomics is the study of genes in one’s genome, and may help in­crease health by us­ing knowl­edge of one’s genes to have per­son­al­ized treat­ment. How­ever, it hasn’t proved to be use­ful for most; recom­men­da­tions rarely change af­ter knowl­edge from ge­nomic test­ing. Still, ge­nomics has much fu­ture po­ten­tial.102


Like I’ve said in the sec­tion “Can we be­come im­mor­tal,” the pro­por­tion of deaths that are caused by ag­ing in the in­dus­trial world ap­proaches 90%,53 but some or­ga­ni­za­tions and com­pa­nies are work­ing on cur­ing it.54, 55, 56

One could sup­port these or­ga­ni­za­tions in an effort to has­ten the de­vel­op­ment of anti-ag­ing ther­a­pies, al­though I doubt an in­di­vi­d­ual would have a no­tice­able im­pact on one’s own chance of death un­less one is very wealthy. That said, I have lit­tle knowl­edge in in­vest­ments, but I sup­pose in­vest­ing in com­pa­nies work­ing on cur­ing ag­ing may be benefi­cial, as if they suc­ceed, they may offer an enor­mous re­turn on in­vest­ment, and if they fail, one would prob­a­bly die, so los­ing one’s money may not be as bad. Cal­ico cur­rently isn’t a pub­lic stock, though.

Ex­ter­nal causes of death

Un­less oth­er­wise speci­fied, graphs in this sec­tion are on data col­lected from Amer­i­can cit­i­zens ages 15-24, as based off the Less Wrong cen­sus re­sults, this seems to be the most prob­a­ble de­mo­graphic that will read this. For this de­mo­graphic, ex­ter­nal causes cause 76% of deaths. Note that al­though this is true, one is much more likely to die when older than when aged 15-24, and older in­di­vi­d­u­als are much more likely to die from dis­ease than from ex­ter­nal causes of death. Thus, I think it’s more im­por­tant when young to de­crease risk of dis­ease than ex­ter­nal causes of death. The graph be­low shows the per­centage of to­tal deaths from ex­ter­nal causes caused by var­i­ous causes.


Trans­port accidents

Below are the rel­a­tive death rates of speci­fied means of trans­porta­tion for peo­ple in gen­eral:


Much in­for­ma­tion about pre­vent­ing death from car crashes is here. In­for­ma­tion on pre­vent­ing death from car crashes is here, here, here, and here.


Life­hacker’s “Ba­sic Self-Defense Moves Any­one Can Do (and Every­one Should Know)” gives a ba­sic in­tro­duc­tion to self defence.

In­ten­tional self harm

In­ten­tional self harm such as suicide, pre­sum­ably, in­creases one’s risk of death.47 Mayo Clinic has a guide on pre­vent­ing suicide. I recom­mend look­ing at it if you are con­sid­er­ing kil­ling your­self. Ad­di­tion­ally, if are are con­sid­er­ing kil­ling your­self, I sug­gest re­view­ing the po­ten­tial re­wards of achiev­ing im­mor­tal­ity from the sec­tion “Should we try to be­come im­mor­tal.”


What to do if a poi­son­ing occurs

CDC recom­mends stay­ing calm, di­al­ing 1-800-222-1222, and hav­ing this in­for­ma­tion ready:

  • Your age and weight.

  • If available, the con­tainer of the poi­son.

  • The time of the poi­son ex­po­sure.

  • The ad­dress where the poi­son­ing oc­curred.

It also recom­mends stay­ing on the phone and fol­low­ing the in­struc­tions of the emer­gency op­er­a­tor or poi­son con­trol cen­ter.18

Types of poisons

Below is a graph of the risk of death per type of poi­son.


Some types of poi­sons:

  • Medicine over­doses.

  • Some house­hold chem­i­cals.

  • Re­cre­ational drug over­doses.

  • Car­bon monox­ide.

  • Me­tals such as lead and mer­cury.

  • Plants12 and mush­rooms.14

  • Pre­sum­ably some an­i­mals.

  • Some fumes, gases, and va­pors.15

Re­cre­ational drugs

Us­ing recre­ational drugs in­creases risk of death.

Medicine over­doses and house­hold chemicals

CDC has tips for these here.

Car­bon monoxide

CDC and Mayo Clinic have tips for this here and here.


Lead poi­son­ing causes 0.2% of deaths wor­ld­wide and 0.0% of deaths in de­vel­oped coun­tries.22 Chil­dren un­der the age of 6 are at higher risk of lead poi­son­ing.24 Thus, for those who aren’t chil­dren, learn­ing more about pre­vent­ing lead poi­son­ing seems like more effort than it’s worth. No com­pletely safe blood lead level has been iden­ti­fied.23


MedlinePlus has an ar­ti­cle on mer­cury poi­son­ing here.

Ac­ci­den­tal drowning

In­for­ma­tion on pre­vent­ing ac­ci­den­tal drown­ing from CDC is here and here.

Inan­i­mate me­chan­i­cal forces

Over half of deaths from inan­i­mate me­chan­i­cal forces for Amer­i­cans aged 15-24 are from firearms. Many of the other deaths are from ex­plo­sions, ma­chin­ery, and get­ting hit by ob­jects. I sup­pose us­ing com­mon sense, pre­cau­tion, and stan­dard safety pro­ce­dures when deal­ing with such things is one’s best defense.


Again, I sup­pose com­mon sense and pre­cau­tion is one’s best defense. Ad­di­tion­ally, al­co­hol and sub­stance abuse is a risk fac­tor of fal­ling.72

Smoke, fire and heat

Own­ing smoke alarms halves one’s risk of dy­ing in a home fire.73 Again, com­mon sense when deal­ing with fires and items po­ten­tially caus­ing fires (e.g. elec­tri­cal wires and de­vices) seems effec­tive.

Other ac­ci­den­tal threats to breathing

Deaths from other ac­ci­den­tal threats to breath­ing are largely caused by stran­gling or chok­ing on food or gas­tric con­tents, and oc­ca­sion­ally by be­ing in a cave-in or trapped in a low-oxy­gen en­vi­ron­ment.21 Chok­ing can be caused by eat­ing quickly or laugh­ing while eat­ing.74 If you are chok­ing:

  • Force­fully cough. Lean as far for­wards as you can and hold onto some­thing that is firmly an­chored, if pos­si­ble. Breathe out and then take a deep breath in and cough; this may eject the for­eign ob­ject.

  • At­tract some­one’s at­ten­tion for help.75

Ad­di­tion­ally, chok­ing can be caused by vom­it­ing while un­con­scious, which can be caused by be­ing very drunk.76 I sug­gest ly­ing in the re­cov­ery po­si­tion if you think you may vomit while un­con­scious, so as to to de­crease the chance of chok­ing on vomit.77 Don’t for­get to use com­mon sense.

Elec­tric current

Elec­tric shock is usu­ally caused by con­tact with poorly in­su­lated wires or un­grounded elec­tri­cal equip­ment, us­ing elec­tri­cal de­vices while in wa­ter, or light­ning.78 Roughly ⅓ of deaths from elec­tric­ity are caused by ex­po­sure to elec­tric trans­mis­sion lines.21

Forces of nature

Deaths from forces of na­ture in (for Amer­i­cans ages 15-24) in de­scend­ing or­der of num­ber of deaths caused are: ex­po­sure to cold, ex­po­sure to heat, light­ning, avalanches or other earth move­ments, cat­a­clys­mic storms, and floods.21 Here are some tips to pre­vent these deaths:

  • When trav­el­ing in cold weather, carry emer­gency sup­plies in your car and tell some­one where you’re head­ing.79

  • Stay hy­drated dur­ing hot weather.80

  • Safe lo­ca­tions from light­ning in­clude sub­stan­tial build­ings and hard-topped ve­hi­cles. Safe lo­ca­tions don’t in­clude small sheds, rain shelters, and open ve­hi­cles.

  • Wait un­til there are no thun­der­storm clouds in the area be­fore go­ing to a lo­ca­tion that isn’t light­ning safe.81

Med­i­cal care

Since med­i­cal care is tasked with treat­ing dis­eases, re­ceiv­ing med­i­cal care when one has ill­nesses pre­sum­ably de­creases risk of death. Though nec­es­sary med­i­cal care may be es­sen­tial when one has ill­nesses, a re­view es­ti­mated that pre­ventable med­i­cal er­rors con­tributed to roughly 440,000 deaths per year in the US, which is roughly one-sixth of to­tal deaths in the US. It gave a lower limit of 210,000 deaths per year.

The fre­quency of deaths from pre­ventable med­i­cal er­rors varied across stud­ies used in the re­view, with a hos­pi­tal that was shown the put much effort into im­prov­ing pa­tient safety hav­ing a lower pro­por­tion of deaths from pre­ventable med­i­cal er­rors than that of oth­ers.57 Thus, I sup­pose that it would be benefi­cial to go to hos­pi­tals that are known for their ded­i­ca­tion to pa­tient safety. There are sev­eral rank­ings of hos­pi­tal safety available on the in­ter­net, such as this one. In­for­ma­tion on how to help pre­vent med­i­cal er­rors is found here and un­der the “What Con­sumers Can Do” sec­tion here. One rare med­i­cal er­ror is hav­ing a surgery be done on the wrong body part. The New York Times gives tips for pre­vent­ing this here.

Ad­di­tion­ally, I sup­pose it may be good to live rel­a­tively close to a hos­pi­tal so as to be able to quickly reach it in emer­gen­cies, though I’ve found no sources stat­ing this.

A com­mon form of med­i­cal care are gen­eral health checks. A com­pre­hen­sive Cochrane re­view with 182,880 sub­jects con­cluded that gen­eral health checks are prob­a­bly not benefi­cial.107 A meta-anal­y­sis found that gen­eral health checks are as­so­ci­ated with small but statis­ti­cally sig­nifi­cant benefits in fac­tor­ing re­lated to mor­tal­ity, such as blood pres­sure and body mass in­dex. How­ever, it found no sig­nifi­cant as­so­ci­a­tion with mor­tal­ity.109 The New York Times ac­knowl­edged that health checks are prob­a­bly not benefi­cial and gave some ex­pla­na­tion why gen­eral health checks are nonethe­less still com­mon.108 How­ever, CDC and MedlinePlus recom­mend get­ting rou­tine gen­eral health checks. The cited no stud­ies to sup­port their claims.104, 106 When I con­tacted CDC about it, it re­sponded, “Reg­u­lar health ex­ams and tests can help find prob­lems be­fore they start. They also can help find prob­lems early, when your chances for treat­ment and cure are bet­ter. By get­ting the right health ser­vices, screen­ings, and treat­ments, you are tak­ing steps that help your chances for liv­ing a longer, healthier life,” a claim that doesn’t seem sup­ported by ev­i­dence. It also stated, “Although CDC un­der­stands you are con­cerned, the agency does not com­ment on in­for­ma­tion from un­offi­cial or non-CDC sources.” I never heard back from MedlinePlus.


Cry­on­ics is the freez­ing of legally dead hu­mans with the pur­pose pre­serv­ing their bod­ies so they can be brought back to life in the fu­ture once tech­nol­ogy makes it pos­si­ble. Hu­man tis­sue have been cry­op­re­served and then brought back to life, al­though this has never been done on full hu­mans.59 The price of Cry­on­ics at least ranges from $28,000 to $200,000.60 More in­for­ma­tion on cry­on­ics is on LessWrong Wiki.


Cry­on­ics, med­i­cal care, safe hous­ing, and ba­sic needs all take money. Re­ju­ve­na­tion ther­apy may also be very ex­pen­sive. It seems valuable to have a rea­son­able amount of money and in­come.

Fu­ture advancements

Keep­ing up­dated on fur­ther ad­vance­ments in tech­nol­ogy seems like a good idea, as not do­ing so would pre­vent one from mak­ing use of fu­ture tech­nolo­gies. Keep­ing up­dated on ad­vance­ments on cur­ing ag­ing seems es­pe­cially im­por­tant, due to the mas­sive num­ber of ca­su­alties it in­flicts and the cur­rent work be­ing done to stop it. Up­dates on mind-up­load­ing seem im­por­tant as well. I don’t know of any very effi­cient method of keep­ing up­dated on new ad­vance­ments, but pe­ri­od­i­cally googling for ar­ti­cles about cur­ing ag­ing or Cal­ico and search­ing for new sci­en­tific ar­ti­cles on top­ics in this guide seems rea­son­able. As knb sug­gested, it seems benefi­cial to pe­ri­od­i­cally check on Fight Aging, a web­site ad­vo­cat­ing anti-ag­ing ther­a­pies. I’ll try to do this and up­date this guide with any new rele­vant in­for­ma­tion I find.

There is much un­cer­tainty ahead, but if we’re clever enough, we just might make it though al­ive.


  1. Ac­tual Causes of Death in the United States, 2000.

  2. A New, Ev­i­dence-based Es­ti­mate of Pa­tient Harms As­so­ci­ated with Hospi­tal Care.

  3. All pages in The Nutri­tion Source, a part of the Har­vard School of Public Health.

  4. Will calorie re­stric­tion work on hu­mans?

  5. The pages Get­ting Started, Tests and Bio­mark­ers, and Risks from The CR So­ciety.

  6. The causal role of break­fast in en­ergy bal­ance and health: a ran­dom­ized con­trol­led trial in lean adults.

  7. Low Glycemic In­dex: Lente Car­bo­hy­drates and Phys­iolog­i­cal Effects of al­tered food fre­quency. Pub­lished in 1994.

  8. Leisure Time Phys­i­cal Ac­tivity of Moder­ate to Vi­gor­ous In­ten­sity and Mor­tal­ity: A Large Pooled Co­hort Anal­y­sis.

  9. Ex­er­cis­ing for Health and Longevity vs Peak Perfor­mance: Differ­ent Reg­i­mens for Differ­ent Goals.

  10. Water: How much should you drink ev­ery day?

  11. MET-hour equiv­a­lents of var­i­ous phys­i­cal ac­tivi­ties.

  12. Poi­son­ing. NLM

  13. Car­cino­gen. Dic­

  14. Types of Poi­sons. New York Poi­son Center

  15. The Most Com­mon Poi­sons for Chil­dren and Adults. Na­tional Cap­i­tal Poi­son Cen­ter.

  16. Known and Prob­a­ble Hu­man Car­cino­gens. Amer­i­can can­cer so­ciety.

  17. Nutri­tional Effects of Food Pro­cess­ing. Nutri­tion­

  18. Tips to Prevent Poi­son­ings. CDC.

  19. Car­bon monox­ide poi­son­ing. Mayo Clinic.

  20. Car­bon Monox­ide Poi­son­ing. CDC.

  21. CDCWONDER. Query Cri­te­ria taken from all gen­ders, all states, all races, all lev­els of ur­ban­iza­tion, all week­days, dates 1999 – 2010, ages 15 – 24.

  22. Global health risks: mor­tal­ity and bur­den of dis­ease at­tributable to se­lected ma­jor risks.

  23. Na­tional Biomon­i­tor­ing Pro­gram Fact­sheet. CDC

  24. Lead poi­son­ing. Mayo Clinic.

  25. Mer­cury. Medline Plus.

  26. Snor­ing Is Not As­so­ci­ated With All-Cause Mor­tal­ity, In­ci­dent Car­dio­vas­cu­lar Disease, or Stroke in the Bus­sel­ton Health Study.

  27. Do Stress Tra­jec­to­ries Pre­dict Mor­tal­ity in Older Men? Lon­gi­tu­di­nal Find­ings from the VA Nor­ma­tive Aging Study.

  28. Meta-anal­y­sis of Per­ceived Stress and its As­so­ci­a­tion with In­ci­dent Coronary Heart Disease.

  29. Iron and car­diac is­chemia: a nat­u­ral, quasi-ran­dom ex­per­i­ment com­par­ing el­i­gible with dis­qual­ified blood donors.

  30. As­so­ci­a­tion be­tween psy­cholog­i­cal dis­tress and mor­tal­ity: in­di­vi­d­ual par­ti­ci­pant pooled anal­y­sis of 10 prospec­tive co­hort stud­ies.

  31. Self-re­ported ha­bit­ual snor­ing and risk of car­dio­vas­cu­lar dis­ease and all-cause mor­tal­ity.

  32. Is it true that oc­ca­sion­ally fol­low­ing a fast­ing diet can re­duce my risk of heart dis­ease?

  33. Pos­i­tive Affect and Health.

  34. The As­so­ci­a­tion of Anger and Hos­tility with Fu­ture Coronary Heart Disease: A Meta-An­a­lytic Re­view of Prospec­tive Ev­i­dence.

  35. Giv­ing to Others and the As­so­ci­a­tion Between Stress and Mor­tal­ity.

  36. So­cial Re­la­tion­ships and Mor­tal­ity Risk: A Meta-an­a­lytic Re­view.

  37. Daily Sit­ting Time and All-Cause Mor­tal­ity: A Meta-Anal­y­sis.

  38. Den­tal Health Be­hav­iors, Den­ti­tion, and Mor­tal­ity in the Elderly: The Leisure World Co­hort Study.

  39. Low Con­scien­tious­ness and Risk of All-Cause, Car­dio­vas­cu­lar and Cancer Mor­tal­ity over 17 Years: White­hall II Co­hort Study.

  40. Con­scien­tious­ness and Health-Re­lated Be­hav­iors: A Meta-Anal­y­sis of the Lead­ing Be­hav­ioral Con­trib­u­tors to Mor­tal­ity.

  41. Sleep du­ra­tion and all-cause mor­tal­ity: a crit­i­cal re­view of mea­sure­ment and as­so­ci­a­tions.

  42. Sleep du­ra­tion and mor­tal­ity: a sys­tem­atic re­view and meta-anal­y­sis.

  43. How Much Sleep Is Enough? Na­tional Lung, Blood, and Heart In­sti­tute.

  44. How many hours of sleep are enough for good health? Mayo Clinic.

  45. Assess Your Sleep Needs. Har­vard Med­i­cal School.

  46. A Life-Span Devel­op­men­tal Per­spec­tive on So­cial Sta­tus and Health.

  47. Suicide. Mer­riam-Web­ster.

  48. Can testos­terone ther­apy pro­mote youth and vi­tal­ity? Mayo Clinic.

  49. Breast Self-Exam. Su­san G. Komen.

  50. Screen­ing Guidelines. The Me­mo­rial Sloan Ket­ter­ing Cancer Cen­ter.

  51. Breast Cancer Screen­ing Overview. The Na­tional Cancer In­sti­tute.

  52. Tes­tic­u­lar self-exam. The Amer­i­can Cancer So­ciety.

  53. Life Span Ex­ten­sion Re­search and Public De­bate: So­cietal Con­sid­er­a­tions.

  54. SENS Re­search Foun­da­tion: About.

  55. Science for Life Ex­ten­sion Homepage.

  56. Google’s pro­ject to ‘cure death,’ Cal­ico, an­nounces $1.5 billion re­search cen­ter. The Verge.

  57. A New, Ev­i­dence-based Es­ti­mate of Pa­tient Harms As­so­ci­ated with Hospi­tal Care.

  58. When Sur­geons Cut the Wrong Body Part. The New York Times.

  59. Cold facts about cry­on­ics. The Guardian.

  60. The cry­on­ics or­ga­ni­za­tion founded by the “Father of Cry­on­ics,” Robert C.W. Et­tinger. Cry­on­ics In­sti­tute.

  61. Es­cape Ve­loc­ity: Why the Prospect of Ex­treme Hu­man Life Ex­ten­sion Mat­ters Now.

  62. In­ter­na­tional Jour­nal of Ma­chine Con­scious­ness In­tro­duc­tion.

  63. The Philos­o­phy of ‘Her.’ The New York Times.

  64. How to Sur­vive the End of the Uni­verse. Dis­cover Magaz­ine.

  65. A Space-Time Crys­tal to Out­live the Uni­verse. Uni­verse To­day.

  66. Con­junc­tion Fal­lacy. Less Wrong.

  67. Cog­ni­tive Bi­ases Po­ten­tially Affect­ing Judg­ment of Global Risks.

  68. Ge­netic in­fluence on hu­man lifes­pan and longevity.

  69. First Drug Shown to Ex­tend Life Span in Mam­mals. MIT Tech­nol­ogy Re­view.

  70. Siroli­mus (Oral Route). Mayo Clinic.

  71. Micro­morts. Un­der­stand­ing Uncer­tainty.

  72. Falls. WHO.

  73. Smoke alarm out­reach ma­te­ri­als. US Fire Ad­minis­tra­tion.

  74. What causes chok­ing? 17 pos­si­ble con­di­tions. Health­line.

  75. Chok­ing. Bet­ter Health Chan­nel.

  76. Aspira­tion pneu­mo­nia. HealthCen­tral.

  77. First aid—Re­cov­ery po­si­tion. NHS Choices.

  78. Elec­tric Shock. HowStuffWorks.

  79. Hy­pother­mia pre­ven­tion. Mayo Clinic.

  80. Ex­treme Heat: A Preven­tion Guide to Pro­mote Your Per­sonal Health and Safety. CDC.

  81. Un­der­stand­ing the Light­ning Threat: Min­i­miz­ing Your Risk. Na­tional weather ser­vice.

  82. The Case Against QuikClot. The sur­vival mom.

  83. Does the Per­cep­tion that Stress Affects Health Mat­ter? The As­so­ci­a­tion with Health and Mor­tal­ity.

  84. Cancer Preven­tion. WHO.

  85. In­fec­tions That Can Lead to Cancer. Amer­i­can Cancer So­ciety.

  86. Pol­lu­tion. Amer­i­can Cancer So­ciety.

  87. Oc­cu­pa­tions or Oc­cu­pa­tional Groups As­so­ci­ated with Car­cino­gen Ex­po­sures. Cana­dian Cen­tre for Oc­cu­pa­tional Health and Safety.

  88. Radon. Amer­i­can Cancer So­ciety.

  89. Med­i­cal ra­di­a­tion. Amer­i­can Cancer So­ciety.

  90. Ul­travi­o­let (UV) Ra­di­a­tion. Amer­i­can Cancer So­ciety.

  91. An Un­healthy Glow. Amer­i­can Cancer So­ciety.

  92. Sun ex­po­sure and vi­tamin D suffi­ciency.

  93. Cell Phones and Cancer Risk. Na­tional Cancer In­sti­tute.

  94. Nutri­tion for Every­one. CDC.

  95. How Can I Tell If My Body is Miss­ing Key Nutri­ents?

  96. De­caf­feina­tion, Green Tea and Benefits. Teas etc.

  97. Red and Pro­cessed Meat Con­sump­tion and Risk of In­ci­dent Coronary Heart Disease, Stroke, and Di­a­betes Mel­li­tus.

  98. Lifestyle in­ter­ven­tions to in­crease longevity.

  99. Chem­i­cals in Meat Cooked at High Tem­per­a­tures and Cancer Risk. Na­tional Cancer In­sti­tute.

  100. Are You Liv­ing in a Si­mu­la­tion?

  101. How re­li­able are sci­en­tific stud­ies?

  102. Ge­nomics: What You Should Know. Forbes.

  103. Or­ganic foods: Are they safer? More nu­tri­tious? Mayo Clinic.

  104. Health screen­ing—men—ages 18 to 39. MedlinePlus.

  105. Why do I need med­i­cal check­ups. Ban­ner Health.

  106. Reg­u­lar Check-Ups are Im­por­tant. CDC.

  107. Gen­eral health checks in adults for re­duc­ing mor­bidity and mor­tal­ity for dis­ease (Re­view).

  108. Let’s (Not) Get Phys­i­cals.

  109. Effec­tive­ness of gen­eral prac­tice-based health checks: a sys­tem­atic re­view and meta-anal­y­sis.

  110. Sup­ple­ments: Nutri­tion in a Pill? Mayo Clinic.

  111. Nutri­tional Effects of Food Pro­cess­ing. SelfNutri­tionData.

  112. What Is the Healthiest Drink? SFGate.

  113. Lead­ing Causes of Death. CDC.

  114. Bias De­tec­tion in Meta-anal­y­sis. Statis­ti­cal Help.

  115. The sum­mary of Sodium In­take in Pop­u­la­tions: Assess­ment of Ev­i­dence. In­sti­tute of Medicine.

  116. Com­pared With Usual Sodium In­take, Low and Ex­ces­sive -Sodium Diets Are As­so­ci­ated With In­creased Mor­tal­ity: A Meta-anal­y­sis.

  117. The Cochrane Re­view of Sodium and Health.

  118. Is there any link be­tween cel­l­phones and can­cer? Mayo Clinic.

  119. A glass of red wine a day keeps the doc­tor away. Yale-New Haven Hospi­tal.

  120. Com­ment on Lifestyle In­ter­ven­tions to In­crease Longevity. Less Wrong.