| Habit | What was found | Evidence | Source |
|---|
| Follow a Mediterranean diet | US health workers with the highest Mediterranean diet scores had an 18% lower death rate than those with the lowest scores. | observed | Shan 2023, JAMA Intern Med |
| Follow a Mediterranean diet (US women, 25 years) | Women with the highest scores had a 23% lower death rate; this shrank to 11% lower after other lifestyle habits were taken into account. | observed | Ahmad 2024, JAMA Netw Open |
| Follow a Mediterranean diet with extra olive oil or nuts | Spanish trial of 7,447 people at high heart risk: about 30% fewer heart attacks, strokes and heart deaths (0.69 with olive oil, 0.72 with nuts) than low-fat advice. | trial | Estruch 2018, N Engl J Med |
| Join a Mediterranean diet programme if you have raised heart risk | Across 40 trials in people at raised heart risk, Mediterranean programmes cut deaths (odds ratio 0.72), about 17 fewer deaths per 1,000 over 5 years. | trial | Karam 2023, BMJ |
| Follow a Mediterranean diet (dementia risk) | UK study found 23% lower dementia risk with high adherence, but a Swedish study of 1,943 dementia cases found no link (0.93, not significant). | mixed | Shannon 2023, BMC Med |
| Follow the DASH diet | Each 5-point rise in DASH score went with a 5% lower death rate; feeding trials show it lowers blood pressure by 7 to 11 points with low salt. | observed | Soltani 2020, Nutr J |
| Follow the MIND diet (brain health) | Cohorts link a higher MIND score to 17% lower dementia risk, but the 3-year US trial of 604 people found no benefit for thinking skills. | mixed | Barnes 2023, N Engl J Med |
| Follow the MIND diet (length of life) | Across 23 observational studies, higher MIND adherence went with a 14% lower death rate. | observed | Ahmadirad 2026, Nutr Metab (Lond) |
| Eat a vegetarian diet | US Adventist vegetarians had a 12% lower death rate than non-vegetarians (0.88); UK vegetarians showed no difference from meat eaters. | mixed | Orlich 2013, JAMA Intern Med |
| Eat a vegetarian diet (UK evidence) | In 60,310 UK adults, vegetarians had the same death rate as regular meat eaters (1.02, no difference). | mixed | Appleby 2016, Am J Clin Nutr |
| Eat a vegetarian diet (heart disease and stroke) | UK vegetarians had 22% less heart disease but 20% more stroke than meat eaters over 18 years. | mixed | Tong 2019, BMJ |
| Eat a vegan diet | US Adventist vegans had a 15% lower death rate than non-vegetarians, but the result was not statistically clear (range 0.73 to 1.01). | mixed | Orlich 2013, JAMA Intern Med |
| Eat a vegan diet (bone health) | UK vegans had 2.3 times the hip fracture risk of meat eaters, about 15 more cases per 1,000 people over 10 years. | observed | Tong 2020, BMC Med |
| Eat mostly healthy plant foods | US health workers with the highest healthy plant-based scores had a 14% lower death rate than those with the lowest. | observed | Shan 2023, JAMA Intern Med |
| Eat a plant-based diet built on refined grains and sweets | People whose diet shifted most toward unhealthy plant foods over 12 years had a 12% higher death rate than those whose diet stayed stable. | observed | Baden 2019, Circulation |
| Follow a low-carb diet | Getting under 40% of calories from carbs went with a 20% higher death rate in pooled cohorts; a global study instead linked high carb intake to more deaths. | mixed | Seidelmann 2018, Lancet Public Health |
| Follow a low-carb diet based on meat and animal fat | Replacing carbs with animal fat and protein went with an 18% higher death rate. | observed | Seidelmann 2018, Lancet Public Health |
| Follow a low-carb diet based on plant fats and proteins | Replacing carbs with plant fat and protein went with an 18% lower death rate. | observed | Seidelmann 2018, Lancet Public Health |
| Eat under 30% of calories from carbs | From age 50, expected remaining life was about 29 years at under 30% carbs versus about 33 years at 50 to 55% carbs. | mixed | Seidelmann 2018, Lancet Public Health |
| Follow a keto diet | No study has followed strict keto dieters for deaths. A 12-week trial found blood sugar no better than a Mediterranean diet, with LDL cholesterol up 10%. | unsupported | Gardner 2022, Am J Clin Nutr |
| Follow a low-carb high-fat diet (heart risk) | UK low-carb high-fat eaters had about twice the rate of major heart events (9.8% vs 4.3%); a small US scan study found no extra artery plaque. | mixed | Iatan 2024, JACC Adv |
| Follow the Atkins diet | In a 12-month trial women lost 4.7 kg on Atkins versus 1.6 to 2.6 kg on other diets. No study has measured deaths. | unsupported | Gardner 2007, JAMA |
| Follow a paleo diet | In 12 cohorts, high paleo-style scores went with a 10% lower death rate; short trials show smaller waist and lower triglycerides. | observed | Bahrami 2026, Crit Rev Food Sci Nutr |
| Follow a low-fat diet | In the largest diet trial (48,835 US women, 8 years), a low-fat diet did not reduce heart disease (0.97) or stroke (1.02). | unsupported | Howard 2006, JAMA |
| Follow a low-fat diet (breast cancer deaths in women) | At about 20 years the same trial found 21% fewer deaths from breast cancer (132 vs 251 deaths), with no effect on other cancer deaths. | trial | Chlebowski 2020, J Clin Oncol |
| Do intermittent fasting or time-restricted eating | In a 12-month trial, eating only between 8am and 4pm plus calorie limits gave no significant extra weight loss over calorie limits alone. | unsupported | Liu 2022, N Engl J Med |
| Eat within a window of under 8 hours a day | A US survey study linked an eating window under 8 hours to 2.35 times the heart death rate, but found no link with deaths from all causes. | mixed | Chen 2025, Diabetes Metab Syndr |
| Do alternate-day fasting | In a 12-month trial, alternate-day fasting gave the same weight loss as daily calorie cutting (6.0% vs 5.3%), with more dropouts and higher LDL. | unsupported | Trepanowski 2017, JAMA Intern Med |
| Cut calories long term while at a healthy weight | A 2-year US trial in healthy adults achieved about 12% fewer calories and 7.5 kg weight loss, with lower LDL, blood pressure and inflammation. | unsupported | Kraus 2019, Lancet Diabetes Endocrinol |
| Follow a carnivore diet | The main evidence is a social-media survey of 2,029 followers; among those reporting cholesterol, typical LDL was 172 mg/dL. | unsupported | Lennerz 2021, Curr Dev Nutr |
| Follow a raw food diet | A German survey found long-term raw food eaters lost 10 to 12 kg; about 30% of women under 45 had missed periods. | unsupported | Koebnick 1999, Ann Nutr Metab |
| Avoid gluten without having coeliac disease | In 110,000 US adults over 26 years, eating more gluten was not linked to heart disease (0.95, not significant). | unsupported | Lebwohl 2017, BMJ |
| Follow an alkaline diet | A review screened 8,278 papers and found only one usable study and no trials; it concluded promotion for cancer is not justified. | unsupported | Fenton 2016, BMJ Open |
| Follow the Whole30 diet | A search of the medical literature found no published studies with Whole30 in the title. | unsupported | Europe PMC title search, 2026 |
| Follow a Nordic diet | Across 13 cohorts of 930,153 people, the highest adherence went with a 22% lower death rate. | observed | Jalilpiran 2022, Crit Rev Food Sci Nutr |
| Follow a traditional Japanese-style diet | In a large Japanese cohort, the best adherence to national food guidelines went with a 15% lower death rate. | observed | Kurotani 2016, BMJ |
| Follow a Blue Zones or Okinawan diet | No study has tested a Blue Zones diet. A preprint found reports of extreme old age fell 69 to 82% once US states brought in birth certificates. | unsupported | Newman 2019, bioRxiv preprint |
| Eat a high-quality diet overall | US health workers with the highest healthy-eating scores (AHEI) had a 20% lower death rate than the lowest over up to 36 years. | observed | Shan 2023, JAMA Intern Med |
| Follow the planetary health (EAT-Lancet) diet | In 206,404 US health workers, the highest scores went with a 23% lower death rate than the lowest. | observed | Bui 2024, Am J Clin Nutr |
| Follow the Portfolio (cholesterol-lowering) diet | US health workers with the highest Portfolio scores had 14% less heart disease and stroke than the lowest. | observed | Glenn 2023, Circulation |
| Switch from a typical Western diet to an ideal diet for life | A model estimates 10.7 extra years for US women and 13.0 for men if started at age 20, and 8.0 and 8.8 years if started at 60. | observed | Fadnes 2022, PLoS Med |
| Join WW (Weight Watchers) | Trials show at least 2.6% more weight loss at 12 months than control or education. No study has measured deaths. | unsupported | Gudzune 2015, Ann Intern Med |
| Use Noom | No randomised trial results were found; the evidence is a company survey of 840 users who had already lost weight. | unsupported | May 2023, Obes Sci Pract |
| Join a diabetes prevention lifestyle programme | The US trial cut new diabetes by 58% over 3 years and 27% over 15 years, but deaths over 21 years were no different (1.02). | unsupported | Lee 2021, Diabetes Care |
| Join an intensive weight-loss programme if you have type 2 diabetes | A US trial of 5,145 people found no fewer heart events (0.95) and no clear fall in deaths at 16.7 years (0.91, not significant). | unsupported | Look AHEAD Research Group 2022, Diabetes Care |
| Drink 2 to 5 cups of coffee a day | In 498,134 UK adults, people drinking 2 to 5 cups a day had a 12% lower death rate than non-drinkers over 10 years. | observed | Loftfield 2018, JAMA Intern Med |
| Drink 1 cup of coffee a day | In the same UK study, 1 cup a day went with an 8% lower death rate than none. | observed | Loftfield 2018, JAMA Intern Med |
| Drink 8 or more cups of coffee a day | Even at 8 or more cups a day the death rate was 14% lower than in non-drinkers; no upper harm level was seen. | observed | Loftfield 2018, JAMA Intern Med |
| Drink 2 to 3 cups of coffee a day (US multiethnic study) | Across five US ethnic groups, 2 to 3 cups a day went with an 18% lower death rate than none; decaf looked the same. | observed | Park 2017, Ann Intern Med |
| Drink your coffee in the morning only | In 40,725 US adults, morning-only coffee drinkers had a 16% lower death rate than non-drinkers; all-day drinkers had no clear benefit. | mixed | Wang 2025, Eur Heart J |
| Add a little sugar to your coffee | In UK adults, 1.5 to 3.5 lightly sweetened cups a day went with a 28 to 31% lower death rate, similar to unsweetened coffee. | observed | Liu 2022, Ann Intern Med |
| Drink more coffee because of your genes | In 112,509 Danes, genes that raise coffee intake had no effect on death rates (1.01 per gene variant). | unsupported | Nordestgaard 2016, Int J Epidemiol |
| Drink 2 or more cups of tea a day | In 498,043 UK adults, 2 to 3 cups of tea a day went with a 13% lower death rate than none; more cups added nothing. | observed | Inoue-Choi 2022, Ann Intern Med |
| Drink 5 or more cups of green tea a day | In 313,381 Japanese adults, 5 or more cups a day went with a 10% lower death rate in men and 18% lower in women than under 1 cup. | observed | Abe 2019, Eur J Epidemiol |
| Have a glass of wine or one drink a day | After correcting for study flaws in 107 cohorts, light drinkers had no significant advantage over lifelong non-drinkers (0.93, not significant). | unsupported | Zhao 2023, JAMA Netw Open |
| Drink alcohol at any level (genetic evidence) | A UK genetic study found death risk rising in a straight line with alcohol (1.27 per unit step), with no protective dip at low intake. | observed | Kassaw 2024, Int J Epidemiol |
| Drink about 7 to 14 drinks a week | Among 599,912 drinkers, 100 to 200 g of alcohol a week cost about 6 months of life at age 40 versus under 100 g. | observed | Wood 2018, Lancet |
| Drink about 14 to 25 drinks a week | Among 599,912 drinkers, 200 to 350 g of alcohol a week cost 1 to 2 years of life at age 40 versus under 100 g. | observed | Wood 2018, Lancet |
| Drink more than 25 drinks a week | Among 599,912 drinkers, over 350 g of alcohol a week cost 4 to 5 years of life at age 40 versus under 100 g. | observed | Wood 2018, Lancet |
| Drink 65 g or more of alcohol a day (about 5 drinks) | Across 107 cohorts, drinking 65 g or more a day went with a 35% higher death rate than lifelong non-drinking; 45 to 64 g a day, 19% higher. | observed | Zhao 2023, JAMA Netw Open |
| Choose wine instead of beer or spirits | Among UK drinkers of the same amount, spirits drinkers had a 25% higher and beer drinkers an 18% higher death rate than red wine drinkers. | mixed | Jani 2021, BMC Med |
| Drink alcohol only with meals | In UK drinkers, drinking without food went with a 10% higher death rate than drinking with food. | mixed | Jani 2021, BMC Med |
| Have one drink a day (cancer risk) | Lifetime risk of an alcohol-related cancer for women is about 16.5% at under 1 drink a week and 19.0% at 1 a day; for men 10.0% and 11.4%. | observed | US Surgeon General's Advisory 2025 |
| Drink alcohol (dementia risk) | In 559,559 US and UK adults, genetic analysis showed dementia risk rising steadily with drinking, with no protective light-drinking level. | observed | Topiwala 2026, BMJ Evid Based Med |
| Eat a serving of processed meat every day | In US health workers, each daily serving of bacon, sausage or deli meat went with a 20% higher death rate. | observed | Pan 2012, Arch Intern Med |
| Eat a serving of red meat every day | US data link each daily serving of unprocessed red meat to a 13% higher death rate, but a guideline panel judged the evidence low-certainty. | mixed | Pan 2012, Arch Intern Med |
| Keep eating red and processed meat as you do now | A 14-member panel reviewing the same studies advised adults to continue current intake, a weak recommendation on low-certainty evidence. | mixed | Johnston 2019, Ann Intern Med |
| Cut back on red meat | Raising red meat by half a serving a day went with a 10% higher death rate, but cutting by the same amount was not linked to a lower rate. | mixed | Zheng 2019, BMJ |
| Eat 50 g of processed meat a day (bowel cancer) | Each 50 g a day of processed meat raises bowel cancer risk by 18%; the WHO cancer agency classes processed meat as a cause of cancer. | observed | Chan 2011, PLoS One |
| Eat processed red meat most days (dementia risk) | In 133,771 US adults, a quarter serving or more a day of processed red meat went with a 13% higher dementia risk. | observed | Li 2025, Neurology |
| Eat a handful of nuts every day | US health workers eating nuts 7 or more times a week had a 20% lower death rate than those who never ate them. | observed | Bao 2013, N Engl J Med |
| Use more than half a tablespoon of olive oil a day | US health workers using over half a tablespoon a day had a 19% lower death rate than those using none or little. | observed | Guasch-Ferre 2022, J Am Coll Cardiol |
| Use olive oil daily (dementia deaths) | In 92,383 US adults, 7 g or more of olive oil a day went with a 28% lower rate of death from dementia. | observed | Tessier 2024, JAMA Netw Open |
| Eat a lot of butter | In 221,054 US adults, the highest butter intake went with a 15% higher death rate than the lowest. | observed | Zhang 2025, JAMA Intern Med |
| Swap butter for plant oils | Replacing 10 g a day of butter with plant oils went with a 17% lower death rate. | observed | Zhang 2025, JAMA Intern Med |
| Eat 3 servings of whole grains a day | Across 45 studies, each 90 g a day of whole grains (about 3 servings) went with a 17% lower death rate. | observed | Aune 2016, BMJ |
| Eat 25 to 29 g of fibre a day | Across 185 studies, the highest fibre eaters had 15 to 30% lower death and heart disease rates than the lowest. | observed | Reynolds 2019, Lancet |
| Eat 5 servings of fruit and vegetables a day | Eating 5 servings a day rather than 2 went with a 13% lower death rate; more than 5 added nothing. Juice and potatoes did not count. | observed | Wang 2021, Circulation |
| Eat beans or lentils every day | Across 32 cohorts, each 50 g a day of legumes went with a 6% lower death rate. | observed | Zargarzadeh 2023, Adv Nutr |
| Eat a serving of fish every day | Each daily serving of fish went with a 7% lower death rate in pooled studies. | observed | Schwingshackl 2017, Am J Clin Nutr |
| Eat fried food every day | In 106,966 older US women, a serving or more of fried food a day went with an 8% higher death rate than none. | observed | Sun 2019, BMJ |
| Eat fried chicken every week | In the same US women, a serving or more of fried chicken a week went with a 13% higher death rate. | observed | Sun 2019, BMJ |
| Drink 2 or more sugary drinks a day | US health workers drinking 2 or more sugary drinks a day had a 21% higher death rate than those drinking under 1 a month. | observed | Malik 2019, Circulation |
| Drink 2 or more diet sodas a day | Two or more diet drinks a day went with a 4% higher death rate in the US study (women only) and 26% higher in a European study. | mixed | Malik 2019, Circulation |
| Get a quarter or more of your calories from added sugar | In US adults, 25% or more of calories from added sugar went with 2.75 times the heart death rate versus under 10%. | observed | Yang 2014, JAMA Intern Med |
| Eat a lot of ultra-processed food | Pooled studies show a 21% higher death rate, but the longest US study found only 4% higher, driven by processed meat and sweet drinks. | mixed | Fang 2024, BMJ |
| Eat an egg a day | US pooled data link each half egg a day to an 8% higher death rate, but a review of 33 cohorts found no link (1.02). | mixed | Zhong 2019, JAMA |
| Eat chocolate regularly | In US women, 1 to 3 servings a week went with a 7% lower death rate than none, with no pattern of more being better. | mixed | Sun 2023, J Acad Nutr Diet |
| Take a cocoa extract supplement | A US trial found no significant fall in heart events (0.90) or deaths from all causes (0.89); heart deaths were 27% lower. | mixed | Sesso 2022, Am J Clin Nutr |
| Eat hot chili peppers | US adults who ate any hot red chili pepper had a 13% lower death rate than those who ate none. | mixed | Chopan 2017, PLoS One |
| Always add salt to food at the table | In 501,379 UK adults, always adding salt went with a 28% higher early death rate and 1.5 (women) to 2.3 (men) fewer years of life at 50. | observed | Ma 2022, Eur Heart J |
| Use a potassium-enriched salt substitute | A trial of 20,995 older adults in rural China found 12% fewer deaths and 14% fewer strokes with salt substitute than regular salt. | trial | Neal 2021, N Engl J Med |
| Eat a high-salt diet | In 10,709 adults with measured urine sodium, each extra 1,000 mg a day went with 18% more heart attacks and strokes. | observed | Ma 2022, N Engl J Med |
| Eat yogurt regularly | US women eating yogurt had a 9 to 15% lower death rate than non-eaters, with no dose pattern; men showed no difference. | mixed | Schmid 2020, Am J Clin Nutr |
| Skip breakfast | US adults who never ate breakfast had 87% more heart deaths, but the rise in deaths from all causes (19%) was not significant. | mixed | Rong 2019, J Am Coll Cardiol |
| Eat only one meal a day | US adults eating one meal a day had a 30% higher death rate than those eating three. | mixed | Sun 2023, J Acad Nutr Diet |
| Stay well hydrated | High-normal blood sodium, used as a stand-in for low fluid intake, went with a 21% higher early death rate in 15,752 US adults. | mixed | Dmitrieva 2023, eBioMedicine |
| Take a daily multivitamin | In 390,124 healthy US adults, daily multivitamin users had no lower death rate (1.04); a trial also found no significant effect (0.93). | unsupported | Loftfield 2024, JAMA Netw Open |
| Take a vitamin D supplement | Across 52 trials and 75,454 people, vitamin D did not change deaths from all causes (0.98); cancer deaths were 16% lower. | unsupported | Zhang 2019, BMJ |
| Take a fish oil supplement | In a US trial of 25,871 adults, fish oil did not change deaths (1.02); heart attacks were 28% lower. | unsupported | Manson 2019, N Engl J Med |
| Take a magnesium supplement | No death data exist for magnesium pills. Each extra 100 mg a day from food went with a 10% lower death rate. | unsupported | Fang 2016, BMC Med |
| Take creatine | A literature search found no study of creatine and deaths; a review of 15 studies found no kidney damage. | unsupported | de Souza e Silva 2019, J Ren Nutr |
| Walk about 5 miles a week (about 15 minutes a day) | In 416,175 Taiwanese adults, about 92 minutes a week of moderate exercise went with a 14% lower death rate and 3 more years of life than none. | observed | Wen 2011, Lancet |
| Walk for exercise at around the recommended weekly amount | In 272,550 older US adults, walking for exercise at about the guideline amount went with a 9% lower death rate than no walking. | observed | Watts 2022, JAMA Netw Open |
| Walk about 2 to 3 hours a week | Across 14 studies, a standard weekly walking dose went with an 11% lower death rate, with the biggest gain at the lowest amounts. | observed | Kelly 2014, Int J Behav Nutr Phys Act |
| Walk more than 2 miles a day | Among 707 retired non-smoking men, 12-year death rates were 23.8% for those walking over 2 miles a day and 40.5% for under 1 mile. | observed | Hakim 1998, N Engl J Med |
| Walk fast for 15 minutes a day | In about 85,000 mostly low-income US adults, 15 minutes a day of fast walking went with a 19% lower death rate. | observed | Liu 2025, Am J Prev Med |
| Walk at a brisk pace | In 50,225 British walkers, a brisk or fast usual pace went with a 24% lower death rate than a slow pace; average pace, 20% lower. | observed | Stamatakis 2018, Br J Sports Med |
| Be a brisk rather than slow walker (life expectancy) | In UK adults, brisk walkers had a life expectancy of about 85 to 88 years; slow walkers with low weight, 65 (men) to 72 (women). | observed | Zaccardi 2019, Mayo Clin Proc |
| Take 7,000 steps a day | Across 14 device studies, 7,000 steps a day went with a 47% lower death rate than 2,000 steps a day. | observed | Ding 2025, Lancet Public Health |
| Take 8,000 steps a day | In 4,840 US adults, 8,000 steps a day went with a 51% lower death rate than 4,000; walking speed added nothing extra. | observed | Saint-Maurice 2020, JAMA |
| Add 1,000 steps a day | One review found a 15% lower death rate per extra 1,000 steps; a UK device study found about 8% lower per extra 2,000 steps. | mixed | Banach 2023, Eur J Prev Cardiol |
| Take about 4,400 steps a day (older women) | In 16,741 US women aged about 72, roughly 4,400 steps a day went with a 41% lower death rate than 2,700; gains levelled off near 7,500. | observed | Lee 2019, JAMA Intern Med |
| Take about 9,800 steps a day (dementia risk) | In 78,430 UK adults, about 9,800 steps a day went with half the dementia risk; about 3,800 steps, 25% lower. | observed | del Pozo Cruz 2022, JAMA Neurol |
| Do some activity, but less than the guideline | In 661,137 adults, some activity below the recommended minimum went with a 20% lower death rate than none. | observed | Arem 2015, JAMA Intern Med |
| Do 150 minutes of moderate activity a week | US adults meeting the aerobic guideline had a 29% lower death rate than those not meeting it. | observed | Zhao 2020, BMJ |
| Meet both the aerobic and muscle-strengthening guidelines | US adults meeting both guidelines had a 40% lower death rate than those meeting neither. | observed | Zhao 2020, BMJ |
| Do muscle-strengthening exercise | US adults meeting only the strength guideline had an 11% lower death rate; the best results come at just 30 to 60 minutes a week. | observed | Zhao 2020, BMJ |
| Go from inactive to active | UK adults who went from inactive to meeting the guideline over 5 years had a 24% lower death rate afterwards. | observed | Mok 2019, BMJ |
| Be active in your free time (years of life) | In 654,827 adults, up to 75 minutes a week of brisk walking went with 1.8 more years of life after 40; 450 minutes or more, 4.5 years. | observed | Moore 2012, PLoS Med |
| Be as active as the most active quarter of Americans | A model using device data estimates Americans over 40 would live 5.3 years longer if all were as active as the top quarter. | observed | Veerman 2025, Br J Sports Med |
| Start a structured exercise programme (trial evidence) | A review of trials in about 50,000 older or chronically ill people found exercise did not reduce deaths or new heart disease. | mixed | Ballin 2021, J Intern Med |
| Do supervised exercise twice a week in your 70s | A 5-year Norwegian trial of 1,567 adults found no difference in deaths between supervised exercise and following standard advice. | mixed | Stensvold 2020, BMJ |
| Join a walking-based programme if you are over 70 and unsteady | A US trial of 1,635 adults aged 70 to 89 found 18% less major mobility disability over 2.6 years. | trial | Pahor 2014, JAMA |
| Be more active because of your genes | Genetic studies found no causal link between a tendency to be active or to sit and death rates. | mixed | Han 2026, Geriatr Nurs |
| Climb 6 to 10 flights of stairs a day | In 280,423 UK adults, 6 to 10 flights a day at home went with a 9% lower death rate, but with no fall in heart deaths. | mixed | Sanchez-Lastra 2021, J Cachexia Sarcopenia Muscle |
| Get a few short bursts of hard effort in daily life | UK non-exercisers with about 3 one-minute bursts a day had a 38 to 40% lower death rate; a US repeat faded once ill people were excluded. | mixed | Stamatakis 2022, Nat Med |
| Do about 15 minutes of vigorous activity a week | In 71,893 UK adults, about 15 minutes a week of vigorous activity went with an 18% lower death rate. | observed | Ahmadi 2022, Eur Heart J |
| Do all your exercise on one or two days a week | In 350,978 US adults, weekend-only exercisers did no worse than people who spread the same amount across the week. | observed | dos Santos 2022, JAMA Intern Med |
| Take a short walk after meals | One-day trials show light walking breaks lower blood sugar and insulin after meals. No study has measured deaths. | unsupported | Buffey 2022, Sports Med |
| Sit for most of the day | Across 47 studies, the most sedentary people had a 24% higher death rate than the least, after allowing for exercise. | observed | Biswas 2015, Ann Intern Med |
| Sit more than 8 hours a day | Each hour of sitting beyond 8 a day went with a 4% higher death rate; below 8 hours the rise was about 1% an hour. | observed | Patterson 2018, Eur J Epidemiol |
| Sit more than 8 hours a day and do little activity | In a million adults, sitting over 8 hours with little activity went with a 59% higher death rate; very active sitters had no excess. | observed | Ekelund 2016, Lancet |
| Watch 2 or more hours of TV a day | Each 2 hours a day of TV went with a 13% higher death rate; risk rose above 3 hours a day. | observed | Grontved 2011, JAMA |
| Spend more of the day standing | In 16,586 Canadian adults, those standing most had a 33% lower death rate than those standing least, only among the inactive. | observed | Katzmarzyk 2014, Med Sci Sports Exerc |
| Cycle to work | In 263,450 UK commuters, cycling to work went with a 41% lower death rate than driving or public transport. | observed | Celis-Morales 2017, BMJ |
| Walk to work | In the same UK commuters, walking to work went with fewer heart deaths but no significant difference in deaths from all causes. | mixed | Celis-Morales 2017, BMJ |
| Go running | Across 14 studies, any running went with a 27% lower death rate than none, with no extra gain from running more. | observed | Pedisic 2020, Br J Sports Med |
| Go running (US at equal effort) | In older US adults, running at the guideline amount went with a 15% lower death rate than not running. | observed | Watts 2022, JAMA Netw Open |
| Play tennis or other racquet sports | In older US adults, racquet sports at the guideline amount went with a 16% lower death rate than not playing. | observed | Watts 2022, JAMA Netw Open |
| Play tennis (years of life) | A Danish study of 8,577 adults reported 9.7 more years for tennis players, 6.2 for badminton and 3.2 for jogging versus sedentary people. | mixed | Schnohr 2018, Mayo Clin Proc |
| Cycle for exercise | In older US adults, cycling at the guideline amount went with a 3% lower death rate; pooled studies of the keenest cyclists show 23% lower. | observed | Watts 2022, JAMA Netw Open |
| Go swimming | In older US adults, swimming at the guideline amount went with a 5% lower death rate; a British study found 28% lower for any swimming. | observed | Watts 2022, JAMA Netw Open |
| Play golf | In older US adults, golf at the guideline amount went with a 7% lower death rate; Swedish golf club members had 40% lower than the public. | observed | Watts 2022, JAMA Netw Open |
| Go dancing | In 48,390 British adults, moderate-intensity dancing went with a 46% lower rate of heart death. | observed | Merom 2016, Am J Prev Med |
| Do gardening | In 13,812 US adults, any gardening went with a 24% lower death rate than none. | observed | Liang 2024, Nutr Metab Cardiovasc Dis |
| Do regular housework | In Korean adults, housework alone went with a 28% lower death rate in men; the result in women (16% lower) was not significant. | mixed | Park 2020, PLoS One |
| Practise tai chi | In 61,477 Chinese men, tai chi went with a 20% lower death rate, about the same as walking; a Singapore study found no clear effect. | observed | Wang 2013, Am J Epidemiol |
| Practise yoga | One US survey study found a 27% lower death rate in yoga users; an earlier one found no link after allowing for age. Trials show better balance. | mixed | Sheehan 2026, Am J Prev Med |
| Do stretching exercises | In 26,727 US adults, stretching went with 10% lower odds of death after allowing for all other exercise types. | mixed | Sheehan 2020, Med Sci Sports Exerc |
| Have a resting heart rate over 80 beats a minute | Across 46 studies, a resting heart rate over 80 went with a 45% higher death rate than the lowest group; each 10 beats higher, 9%. | observed | Zhang 2016, CMAJ |
| Have a weak hand grip | In 139,691 adults in 17 countries, each 5 kg lower grip strength went with a 16% higher death rate. | observed | Leong 2015, Lancet |
| Be unable to stand on one leg for 10 seconds | In 1,702 adults aged 51 to 75 at a Brazilian clinic, failing the test went with an 84% higher death rate over 7 years. | observed | Araujo 2022, Br J Sports Med |
| Need hands or knees to sit down on and get up from the floor | In 2,002 adults aged 51 to 80 at a Brazilian clinic, the lowest scorers had 5.4 times the death rate of the highest. | observed | Brito 2014, Eur J Prev Cardiol |
| Walk slowly when measured (age 65 and over) | In 34,485 adults over 65, each 0.1 metre per second faster walking speed went with a 12% lower death rate. | observed | Studenski 2011, JAMA |
| Be able to do more than 40 push-ups | Among 1,104 male firefighters, those doing over 40 push-ups had 96% fewer heart events than those doing under 10. | observed | Yang 2019, JAMA Netw Open |
| Sleep 6 hours a night instead of 7 | Pooled cohorts show the lowest death rate at about 7 hours; each hour less went with a 6% higher rate. | observed | Yin 2017, J Am Heart Assoc |
| Sleep 8 hours or more a night | Each hour of sleep above 7 went with a 13% higher death rate. | observed | Yin 2017, J Am Heart Assoc |
| Sleep 6 hours or less in midlife (dementia risk) | In 7,959 UK adults followed 25 years, 6 hours or less at ages 50 and 60 went with 22% and 37% higher dementia risk than 7 hours. | observed | Sabia 2021, Nat Commun |
| Keep regular sleep and wake times | In 60,977 UK adults wearing wrist devices, more regular sleepers had a 20 to 48% lower death rate than the least regular fifth. | observed | Windred 2024, Sleep |
| Be a night owl | Finnish twins followed 37 years showed a 9% higher death rate in evening types, explained mainly by smoking and drinking. | unsupported | Hublin 2023, Chronobiol Int |
| Work rotating night shifts for 5 years or more | In 74,862 US nurses, 6 or more years of rotating night shifts went with an 11% higher death rate. | observed | Gu 2015, Am J Prev Med |
| Catch up on sleep at weekends | In 43,880 Swedes under 65, short weekday sleep with long weekend sleep carried no higher death rate; short sleep all week, 65% higher. | mixed | Akerstedt 2019, J Sleep Res |
| Have all five good sleep habits | In 172,321 US adults, all five good sleep habits went with a 30% lower death rate and 4.7 (men) or 2.4 (women) more years at age 30. | observed | Li 2024, QJM |
| Have ongoing insomnia | One review of 17 studies found no link with deaths (1.07, not significant); another found 13% higher for trouble falling asleep. | mixed | Lovato 2019, Sleep Med Rev |
| Have untreated severe sleep apnoea | Across 12 cohorts, severe sleep apnoea went with a 92% higher death rate than no apnoea. | observed | Wang 2013, Int J Cardiol |
| Use CPAP for sleep apnoea | Ten trials in 7,266 patients found no fall in deaths (1.13, not significant); observational studies show large benefits in regular users. | mixed | Yu 2017, JAMA |
| Take sleeping pills | One US study found 3.6 to 5.3 times the death rate in users; a better-controlled one found no increase at 6 months (1.00). | mixed | Patorno 2017, BMJ |
| Take melatonin long term | A 2025 conference report on 130,828 insomnia patients linked a year or more of recorded melatonin use to twice the death rate (8% vs 4%). | unsupported | AHA Scientific Sessions 2025, Circulation abstract |
| Nap for an hour or more a day | Across 11 cohorts, naps of 60 minutes or more went with a 27% higher death rate; naps under an hour showed no link. | observed | Yamada 2015, Sleep |
| Keep nights dark and days bright | In 88,905 UK adults with wrist light sensors, the brightest nights went with a 21 to 34% higher death rate; bright days, 17 to 34% lower. | observed | Windred 2024, Proc Natl Acad Sci USA |
| Live with depression | Across 293 studies, depression went with a 52% higher death rate, but an umbrella review judged a causal effect unproven. | observed | Cuijpers 2014, Am J Psychiatry |
| Work 55 or more hours a week | Working 55 or more hours a week went with 33% more strokes and 13% more heart disease; a European study found no link with overall deaths. | mixed | Kivimaki 2015, Lancet |
| Have a high-strain job | In 197,473 European workers, job strain went with 23% more heart disease. | observed | Kivimaki 2012, Lancet |
| Be unemployed | Across 42 studies, unemployment went with a 63% higher death rate. | observed | Roelfs 2011, Soc Sci Med |
| Suffer a sudden large loss of wealth | In 8,714 US adults over 50, a sudden major loss of wealth went with a 50% higher death rate. | observed | Pool 2018, JAMA |
| Feel a lot of stress | US adults reporting a lot of stress and believing it harmed them had a 43% higher death rate; a study of 719,671 UK women found no link. | mixed | Keller 2012, Health Psychol |
| Be happy most of the time | Pooled studies show an 8% lower death rate with higher wellbeing, but in 719,671 UK women unhappiness made no difference (0.98). | mixed | Liu 2016, Lancet |
| Practise meditation | One trial in 201 Black adults with heart disease found 48% fewer deaths, heart attacks and strokes with Transcendental Meditation. | mixed | Schneider 2012, Circ Cardiovasc Qual Outcomes |
| Take cold plunges or cold showers | A review of 11 trials in 3,177 adults found brief stress relief at 12 hours and no effect on immunity or mood. No death data exist. | unsupported | Cain 2025, PLoS One |
| Live in a greener neighbourhood | Across 9 cohorts of 8.3 million people, each step up in greenery near home went with a 4% lower death rate. | observed | Rojas-Rueda 2019, Lancet Planet Health |
| Have acupuncture | The only death study found was in Korean adults with disabilities and heart disease; nothing applies to healthy adults. | unsupported | Jun 2026, BMC Complement Med Ther |
| Own a dog | A pooled analysis found a 24% lower death rate in dog owners, but it was not adjusted for other factors; adjusted studies found none. | mixed | Kramer 2019, Circ Cardiovasc Qual Outcomes |
| Own a dog (US national study) | In 11,394 US adults, living with a dog went with a 19 to 21% higher death rate after adjustment, the opposite of the headline claim. | mixed | Gillum 2010, Int J Environ Res Public Health |
| Own a dog (English study) | In 59,352 English adults followed 11.5 years, living with a dog made no difference to death rates (1.03). | mixed | Ding 2018, Am J Prev Med |
| Own a dog if you live alone | In 3.4 million Swedes, dog owners living alone had a 33% lower death rate; in other homes, 11% lower. | mixed | Mubanga 2017, Sci Rep |
| Own a cat | Pooled adjusted studies show no link between cat ownership and deaths from all causes (1.04). | unsupported | El-Qushayri 2020, J Cardiovasc Med |
| Own a pet (memory and thinking) | In 7,945 older English adults, pet owners living alone showed slower decline in verbal memory; those living with others showed none. | observed | Li 2023, JAMA Netw Open |
| Walk a dog | About 60% of dog owners walk their dog, a median of 160 minutes a week; owners are only slightly more active than non-owners. | observed | Christian 2013, J Phys Act Health |
| Be socially isolated | Across 90 cohorts of 2.2 million people, social isolation went with a 32% higher death rate. | observed | Wang 2023, Nat Hum Behav |
| Feel lonely | Across 90 cohorts, loneliness went with a 14% higher death rate. | observed | Wang 2023, Nat Hum Behav |
| Live alone | In 458,146 UK adults, living alone went with a 25% higher death rate. | observed | Foster 2023, BMC Med |
| See friends or family less than once a month | In 458,146 UK adults, visits only every 3 months went with an 11% higher death rate than daily visits; weekly or monthly made no difference. | observed | Foster 2023, BMC Med |
| Never see friends or family | In 458,146 UK adults, never being visited by friends or family went with a 39% higher death rate than daily visits. | observed | Foster 2023, BMC Med |
| Take part in no weekly group activity | In 458,146 UK adults, having no weekly group activity went with a 13% higher death rate. | observed | Foster 2023, BMC Med |
| Be married | Across 53 comparisons in older adults, married people had a 12% lower death rate than unmarried people. | observed | Manzoli 2007, Soc Sci Med |
| Attend religious services more than once a week | In 74,534 US women, attending more than once a week went with a 33% lower death rate than never attending. | observed | Li 2016, JAMA Intern Med |
| Volunteer regularly | In adults over 55, volunteering went with a 24% lower death rate after adjustment (47% before). | observed | Okun 2013, Psychol Aging |
| Take part in social activities in older age | Across 25 cohorts of older adults, high social participation went with a 22% lower death rate. | observed | Xiao 2026, Geriatr Nurs |
| Have sex at least weekly | In 12,598 US adults aged 20 to 59, sex 52 or more times a year went with a 49% lower death rate than once a year or less. | mixed | Cao 2020, J Sex Med |
| Be optimistic | Across 15 studies of 229,391 people, optimism went with a 14% lower death rate. | observed | Rozanski 2019, JAMA Netw Open |
| Be among the most optimistic (lifespan) | The most optimistic US women lived 14.9% longer than the least in one study and 5.4% longer in another. | observed | Lee 2019, PNAS |
| Have a strong sense of purpose | Across 10 studies of 136,265 people, a strong sense of purpose went with a 17% lower death rate. | observed | Cohen 2016, Psychosom Med |
| Feel grateful often | In 49,275 older US women, the most grateful third had a 9% lower death rate over 3 years. | mixed | Chen 2024, JAMA Psychiatry |
| Hold positive views about getting older | In 660 Ohio adults over 50, those with more positive views of ageing lived 7.5 years longer. | observed | Levy 2002, J Pers Soc Psychol |
| Be less organised and careful than most people | In 76,150 adults, the least conscientious third had a 37% higher death rate. | observed | Jokela 2013, Am J Epidemiol |
| Laugh often | In 17,152 Japanese adults, laughing less than once a month went with a 95% higher death rate than laughing weekly. | mixed | Sakurada 2020, J Epidemiol |
| Read books | In 3,635 US adults over 50, book readers had a 20% lower death rate than non-readers over up to 12 years. | observed | Bavishi 2016, Soc Sci Med |
| Go to museums, concerts or theatre every few months | In 6,710 English adults over 50, going every few months or more went with a 31% lower death rate than never going. | observed | Fancourt 2019, BMJ |
| Play a musical instrument (dementia risk) | Across 3 cohorts, playing an instrument went with a 36% lower dementia risk. | observed | Arafa 2022, BMC Neurol |
| Do puzzles, games or other mentally active hobbies (dementia risk) | Across 38 studies of 2.15 million people, mentally stimulating activities went with a 23% lower dementia risk. | observed | Su 2022, Neurology |
| Speak two languages | Bilingual people showed dementia symptoms 3.45 years later in clinic studies, but studies following people over time found no lower risk. | mixed | Li 2026, Aging Neuropsychol Cogn |
| Take adult education classes (dementia risk) | In 499,337 UK adults, taking adult education classes went with an 18% lower dementia risk. | observed | Jin 2025, Geroscience |
| Have a hobby | In 93,263 adults over 65 in 16 countries, having a hobby went with fewer depressive symptoms and higher life satisfaction. | unsupported | Mak 2023, Nat Med |
| Take regular vacations | In 12,338 US men at high heart risk, more frequent yearly vacations went with a 17% lower death rate over 9 years. | mixed | Gump 2000, Psychosom Med |
| Keep working past retirement age | A review of 25 studies found no significant link between retirement timing and deaths once earlier health was counted (1.12). | unsupported | Sewdas 2020, J Epidemiol Community Health |
| Always wear a seat belt | Seat belts cut the risk of fatal injury for front-seat car occupants by 45%; 48% of occupants killed in 2024 were unbelted. | observed | NHTSA Seat Belts page, 2025 |
| Wear a helmet on a motorcycle | Among riders who crash, helmets cut the odds of death by 42% and head injury by 69%. | observed | Liu 2008, Cochrane Database Syst Rev |
| Wear a helmet on a bicycle | Among cyclists who crash, helmets cut the odds of serious head injury by 69% and fatal head injury by 65%. | observed | Olivier 2017, Int J Epidemiol |
| Keep working smoke alarms at home | The death rate in reported home fires is about 60% lower in homes with working smoke alarms than in homes without. | observed | McGree 2024, NFPA Smoke Alarms in US Home Fires |
| Keep a gun in the home | Gun access goes with about 3 times the odds of suicide and twice the odds of being killed; owners had no higher overall death rate. | observed | Anglemyer 2014, Ann Intern Med |
| Store guns locked and unloaded | A study of youth suicides and accidental shootings found locked and unloaded storage went with about 70% lower odds. | observed | Grossman 2005, JAMA |
| Get a flu shot soon after a heart attack | In a trial of 2,571 heart attack patients, a flu shot cut deaths over the next year from 4.9% to 2.9%. | trial | Frobert 2021, Circulation |
| Get a yearly flu shot (older adults in general) | Trials in over 5,000 older adults show flu shots cut flu from 6% to 2.4%, but were too small to show any effect on deaths. | unsupported | Demicheli 2018, Cochrane Database Syst Rev |
| Get the pneumonia vaccine at 65 or over | A trial of 84,496 adults over 65 found 46% fewer cases of vaccine-type pneumonia; deaths were similar in both groups. | unsupported | Bonten 2015, N Engl J Med |
| Get the shingles vaccine (dementia risk) | A natural experiment in Wales found the shingles vaccine cut new dementia diagnoses by about 20% over 7 years. | observed | Eyting 2025, Nature |
| Get vaccinated against COVID-19 | A model estimates vaccination prevented 14.4 million deaths worldwide in its first year, a 63% reduction. | observed | Watson 2022, Lancet Infect Dis |
| Brush your teeth every night | In 5,611 older Californians, never brushing at night went with a 20 to 35% higher death rate than brushing every night. | observed | Paganini-Hill 2011, J Aging Res |
| Floss your teeth | A 2026 population study found any flossing went with a 30% lower death rate; the authors call this a lead, not proof. | observed | Liu 2026, J Dent |
| See a dentist twice a year | In 5,611 older Californians, no dental visit in a year went with a 30 to 50% higher death rate than two or more visits. | observed | Paganini-Hill 2011, J Aging Res |
| Have lost 10 or more teeth | Across 18 studies, each 10 teeth lost went with a 15% higher death rate. | observed | Peng 2019, Biosci Rep |
| Use sunscreen every day | An Australian trial of 1,621 adults found daily sunscreen cut invasive melanoma by 73% (3 vs 11 cases). | trial | Green 2011, J Clin Oncol |
| Avoid the sun | In 29,518 Swedish women, sun avoiders lived 0.6 to 2.1 years less than the most sun-exposed. | mixed | Lindqvist 2016, J Intern Med |
| Use a sauna 3 or more times a week | In 2,575 Finnish men followed 28 years, 3 to 7 sauna sessions a week went with a 14% lower death rate than 2 or fewer. | observed | Kunutsor 2022, Eur J Epidemiol |
| Take a hot bath almost every day | In 30,076 Japanese adults, a daily tub bath went with 28% less heart disease and stroke than bathing twice a week or less. | observed | Ukai 2020, Heart |
| Take prescribed medicines as directed | People who took their pills reliably had 44% lower odds of death, but so did people who reliably took dummy pills. | mixed | Simpson 2006, BMJ |
| See the same doctor over time | In 18 of 22 studies, people with more continuity of care had lower death rates; no single figure could be pooled. | observed | Pereira Gray 2018, BMJ Open |
| Wear hearing aids if you have hearing loss | US adults with hearing loss who wore aids regularly had a 24% lower death rate; the one large trial found no benefit for thinking skills. | mixed | Choi 2024, Lancet Healthy Longev |
| Have cataract surgery when needed (dementia risk) | In 3,038 older adults with cataract, surgery went with a 29% lower dementia risk; glaucoma surgery showed no such link. | observed | Lee 2022, JAMA Intern Med |
| Have uncorrected poor eyesight | Across 17 studies, poor vision went with a 29% higher death rate than good vision. | observed | Ehrlich 2021, Lancet Glob Health |
| Fall-proof your home | Trials in 8,463 older people show home hazard fixes cut falls by 26%, and by 38% in those at high risk; no benefit at low risk. | trial | Clemson 2023, Cochrane Database Syst Rev |