Live to 100 With the 80-20 Rule



It’s true: you’re invincible. You’re barely 25 years old, you’re in great (ok, good) shape, you eat well (most of the time), and you’ve only ever gotten a clean bill of health at doctor’s office.

You know you’re supposed to be worried about stuff like retirement and healthcare or <shudder> life insurance, but in the back of your head, you know that with modern medicine improving so fast, you’ll easily live to be 100.

And, if you should happen to catch the sniffles or jam a finger, you’re confident a quick visit to your primary care physician will fix your problems and let you get back to being productive/socializing/rewatching the second season of Mad Men.

Hey – I get it, I’m 26 and that’s how I feel too.

But, part of you wonders: now that you’ve finally got your school debt under control, your career on a decent track, and a social life that’s moving in the right direction, is there anything you could be doing to protect your health?

The short answer is yes, and it’s way easier and much more effective than you’d probably guess. To find out how, read on.

The Rule of Thumb

If you’re interested in productivity, you’ve probably heard of the Pareto principle, a simple rule of thumb that goes something like this:

“80 percent of outcomes are from 20 percent of causes”

It applies to all aspects of life: “80 percent of work gets done by 20% of employees” or “80 percent of my sales come from 20% of my clients” or “I can get this presentation to an 80% with only 20% the effort”

So does it apply to health and living longer too? Check out this chart:

causes of death

…yup, it applies.

The numbers from the Centers for Disease Control (CDC) show that just 15 causes of death are responsible for 80% of all deaths in the United States.

Let’s take a closer look. After eliminating non-medical causes of death (accidents, homicides, and suicides) we see that 4 categories of diseases are responsible for 66% of deaths out of the top 15 causes.

Here’s the chart:

causes of death 2

Um… So What?

[explanation, quick summary]

A note on methodology. We specifically looked at causes of death before the age of 75 because statistics about geriatric medicine are expected to change considerably in the coming decades as people live longer than ever before.1,2

Here’s what changes: while heart disease beats out cancer as an overall cause of death, cancer actually edges it out in the under-75 population. We used 2010 data. Since specifics change from year to year, we ballparked statistics where appropriate.

1 Health Services Research
2 Wien Klin Wochenschr

Lung Disease and Lung Cancer

You could sum this entire section with two simple words: don’t smoke.

Here’s why:

Oh, and add to these the fact that smoking directly increases risk for many cancers besides lung cancer, and you can conclude that smoking directly causes between 20% and 30% of medically-related deaths under the age of 75. And that’s a conservative estimate.

Of course, it’s not always that simple. It’s very easy to become dependent on or addicted to nicotine. Plus, secondhand smoke has the same negative effects as smoking, if not to quite the same severity.

If you do smoke or are regularly exposed to secondhand smoke, know that you are at significantly higher risk for the most deadly diseases. Which means it’s even more important for you to keep a close eye on blood pressure, weight, and lung capacity in order to catch health problems at an early stage.

*Most non-smoking-related cases of chronic respiratory diseases and lung cancer are influenced by genetics. If you have a family history of lung problems or asthmas, you may be at increased risk, especially if you are routinely exposed to smoke.


One in two men and women will be diagnosed with cancer in their lifetimes.

That statistic bears repeating. One in two men and women will be diagnosed with cancer in their lifetimes. It sounds unbelievable but that statistic comes straight from the National Cancer Institute.

It’s hard to make broad recommendations for avoiding cancer in general because there are so many types and each has its own risk factors. That’s why anytime a product or new study says something like “reduces the risk of cancer” or “helps to prevent cancer” you should be skeptical.

So we’ve done the legwork for you.

We looked both at incidence (how often a cancer occurs)3 and mortality rate (how often a patient with a cancer dies from it)4 to identify the cancers you should be most watchful for.*

*Lung cancer, the most deadly type of cancer, is discussed in the previous section.
3 CDC’s data on cancer incidence
4 CDC’s data on cancer mortality. [pdf]

Colon Cancer
(Or why you should be skeptical of all nutrition advice)

Colon cancer is responsible for the majority of cancer deaths after lung cancer. It’s deadly to both men and women.

Here’s the weird thing about colon cancer: way more people in developed contries contract colon cancer than do people in less-developed countries. And that’s true regardless of race and socioeconomic factors.

What does that mean? It means that well-off people are making lifestyle choices that increase their risk of colon cancer.

What lifestyle choices? Researchers usually single out nutrition as the culprit.

But, here’s why nutrition studies are so hard to believe:

In the early 90′s a group of European nations decided to band together and study the effects of nutrition and lifestyle on the development of cancer.

They developed several key takeaways that you’ve probably heard before: to avoid colon cancer,

  • eat more dairy, fiber, vegetables, and fish, and
  • eat less processed or red meat and drink less alcohol.

While these recommendations might sound reasonable, they don’t tell the full story.

It turns out that the European study, (called the EPIC cohort study), mostly covers ground that had already been addressed by epidemiologists in the past.

  • A 1993 epidemiologic review article on colon cancer concluded that only increased vegetable consumption and leading a non-sedentary lifestyle were supported by the majority of studies.
  • More recently, [UpToDate] concludes that obesity is linked with colon cancer, but that fruit and vegetable consumption have little impact, if any. Fish does seem to reduce risk, but only slightly. No other conclusions can be made because the data are just too contradictory.

Why can’t researchers agree on basic nutrition advice? The answer lies somewhere between research methods, politics, and the adaptability of the human body.

We’ll discuss nutrition and nutrition research much more in future articles, but for now, remember to treat any nutrition advice with a healthy dose of skepticism.

[people with 1st-degree relatives who had colon cancer are at increased risk: uptodate.]

Breast Cancer

Breast cancer is the most-often occurring cancer in women and is responsible for the most deaths in women after lung cancer.

The most important action you can take to prevent breast cancer is to get routine mammograms every 1-2 years after the age of 40.


In the past few years, there has been some debate around mammograms. The main issue is that the test generates a high rate of false positives (the test says you have breast cancer when you actually don’t), which can result in unnecessary treatment and hardship.5,6

The current consensus seems to be that the benefits of getting the mammograms outweigh the downsides; just be aware that your doctor may ask you to come in for a second screening and that a second screening most likely does not mean you have breast cancer–the second screening is used to eliminate those annoying false positives.

Some women have an elevated risk of breast cancer due to genetics.

You’re probably aware of Angelina Jolie’s decision to undergo an elective double mastectomy. The actress made that call because she tested positive for BRCA1, a gene that made her lifetime risk for breast cancer about 87%. Statistically, the presence of either the BRCA1 or BRCA2 genes increase lifetime risk for breast cancer to 65% and 45%, respectively.7

Of course, genetic testing can be expensive and not all women should get tested. The most at-risk women are those of Ashkenazi Jewish descent with relatives who have been diagnosed with breast or ovarian cancer.8

If you are concerned about your risks, the Gail Model pools common genetic factors together with other known contributors to estimate an individual’s risk of getting breast cancer. The best way to use the tool is to try it at home and bring your results in next time you see your doctor to discuss what model says about you.

5 Mammogram Fact Sheet from NIH’s National Cancer Institute
6 Free PubMed Article: Mammography screening. Benefits, Harms, and Informed Choice.
7 Stanford Cancer Institute

Prostate Cancer

The prostate is a small gland that is part of the male reproductive system. Prostate cancer is both the most prevalent cancer in men after lung cancer and the number one killer of men after lung cancer.

The risk factors for prostate cancer are pretty straightforward. Prostate cancer rarely occurs in men under 40 and becomes significantly more likely in men older than 60. A family history of prostate cancer increases risk. Diet makes a difference, but exactly how is not clear.9

And that’s about all you can do for prostate cancer–besides, you know, get your prostate checked. There is a marker call prostate-specific antigen (PSA) that’s elevated in patients with prostate cancer. But, studies have shown that getting your PSA levels checked has no effect on actually reducing mortality.10,11

Charlene says:
Most PSAs increase with age (and can also increase with prostate/bladder infection I think), so by screening everybody, you get a lot of false positives and a lot of unnecessary prostate biopsies that may do more harm (complications, infections) than good. Some family medicine docs still do PSAs, so I think it’s still pretty controversial even between individual primary care providers. I think testing should definitely be done for people with high risk and I would be ok with testing a patient feels strongly about getting the test. Plus, most insurance companies will still pay for the test.

9 National Library of Medicine, Prostate Cancer – Causes, Incidence, and Risk Factors
10 New England Journal of Medicine, Mortality results from a randomized prostate-cancer screening trial
11 National Cancer Institute, Prostate-Specific Antigen (PSA) Test

Lymphoma and Leukemia
-and Other Cancers of the Blood

[How dangerous?] [Why are we talking about it?]

In lymphoma and related cancers, the white blood cells–the very cells normally responsible for fighting infection and disease–become cancerous.

That’s why this family of diseases is often referred to as “hematological neoplasms,”–that’s doctor-speak for “cancers of the blood.”

Since blood travels everywhere in the body, the tumors associated with hematological neoplasms can appear in many places at once, such as in the lymph nodes, bone marrow, or spleen.

And since these tumors aren’t localized, they’re harder to treat, which means mortality rates are high.

The two main types of hematological neoplasms are leukemia and lymphoma:

Leukemia affects the bone marrow and disrupts production of white blood cells. Children usually develop acute leukemia, while adults are equally susceptible to both chronic and acute forms of leukemia. Only 56% of patients will survive leukemia after 5 years.12

Lymphoma affects lymphocyte production (lymphocytes are a type of white blood cell) . Both types of lymphocytes–bursa-derived (B-cells) and thymus (T-cells)–are affected.

There are two main types of lymphoma. Hodgkin’s lymphoma is characterized by the orderly spread of the disease from lymph node to lymph node, whereas non-Hodgkin’s lymphoma is characterized by the non-systematic spread of the disease. Non-Hodgkin’s lymphoma is more deadly with a five-year survival rate of 69%, compared to Hodgkin’s lymphoma’s survival rate of 85%.13

Unlike other cancers, hematological neoplasms do not appear to be caused by genetics.14 And that means that environmental exposure to chemicals appears to be the primary cause of lymphomas and leukemias.

Here are the specific substances you should avoid:

  • Exposure to diesel fumes
  • Exposure to ionizing radiation
  • Exposure to hair dyes
  • Exposure to pesticides, specifically:
    • Dichlorodiphenyltrichlorethane (DDT)
    • cis-nonachlor
    • Oxychlordane

References: 15, 16

And by exposure, we mean that being in contact with these chemicals over the course of 20 years can increase your risk of developing cancer by about 50%.

[The research on hematological neoplasms is still developing and the subject of some debate, but here’s the bottom line: although this class of cancers is rare, they are very deadly and should be avoided if possible.]

12 Surveillance Epidemiology and End Results (SEER), Leukemia Incidence and Mortality
13 SEER, Lymphoma Incidence and Mortality
14 Current Opinion in Hematology 1999
15 Environmental Health 2008, Occupational exposures and non-Hodgkin’s lymphoma: Canadian case-control study
16 Environmental Health Perspectives 2012, A Prospective Study of Organochlorines in Adipose Tissue and Risk of Non-Hodgkin Lymphoma

Dishonorable Mentions
Pancreatic, Ovarian/Uterus, Urinary, and Skin Cancers

Although rarely deadly, skin cancer is the most common form of cancer in the U.S. In fact, one in five Americans will develop skin cancer in their lifetime.22 Basal cell carcinoma is the most common form, followed by squamous cell carcinoma. 90% of nonmelanoma skin cancers are associated with UV exposure from the sun.23

Melanomas are more rare but also more serious: 1 in 50 men and women will be diagnosed with melanoma during their lifetimes and the overall 5-year mortality rate is 10%.24 Melanoma is also associated with UV exposure. In a recent study, researchers at Yale found that people who tanned indoors were 69% more likely to develop melanoma. The findings were so conclusive that a law in Connecticut banned indoor tanning only months after the findings were reported.

BUT, although the above study stands as a pretty significant condemnation of intentional tanning, don’t think that scientists are saying you have to hide from the sun. The truth is, most Americans don’t get enough sunlight. To avoid skin cancers, simply use sunscreen if you anticipate being outside for an extended period of time.

Pancreatic cancer: though the lifetime risk for men and women is only 1 in 67, the five-year survival rate for pancreatic cancer is a dismal 6%.17 Smoking accounts for 20% of pancreatic cancer cases. Also, people with the BRCA1 mutation (see breast cancer, above) carry about a 116% increased risk of getting the disease.18

There’s also a protein-based screen for pancreatic cancer developed by a 15-year-old that looks promising, but we don’t know when it will be available to the general public.

Ovarian cancer is only the 10th most common cancer among women, but it’s the 5th most deadly. Genetics play the most significant role in predicting ovarian cancer. BRCA1 increases lifetime risk to 39% and BRCA2 increases it to 11%.19 Women with a family history of the disease may want to consider getting tested for these genes.

Uterine cancer affects 1 in 37 women and has an overall 5-year survival rate of 81.5%.20 Like ovarian cancer, there is no known screen; the disease is largely genetic. Oral contraceptives reduce the risk for both ovarian and uterine cancer.21

Bladder cancer is a less common cancer, but it can be dangerous, with an overall 5-year survival rate of 77.9%.25 About 50% of all bladder cancers are attributed to smoking.26 The rest of cases are usually attributed to genetics. If you think you are at high risk (you’re a smoker or have a family history), there are some simple screens you can do. A urinalysis is the simplest method, but you’ll have to ask specifically for a bladder cancer screen, as that test does not come standard. More specific tests exist, such as UroVysion, bladder tumor-associated antigen, immunocyt, and NMP22 bladder check.27 Once again, these tests are only recommended if you know you have a family history of the disease or you’re a smoker.

17 SEER, Cancer of the Pancreas.
18 Journal of the National Cancer Institute, Cancer Incidence in BRCA1 Mutation Carriers.
19 Stanford Cancer Institute, Hereditary Breast Ovarian Cancer Syndrome (BRCA1 / BRCA2).
20 SEER, Cancer of the Corpus and Uterus.
21 CDC, Uterine Cancer Prevention.
23 Link
24 (SEER)
25 (Source: SEER)
26 (Source: NCI)
27 (ACS)

Conclusions about Cancer

Despite huge strides in medical research and technology, cancer remains a poorly-understood disease. Or rather, a poorly understood family of very different diseases.

Here are the most important takeaways:

  • Smoking greatly increases the risk of multiple types of cancer, not just lung cancer
  • A family history of a particular type of cancer greatly increases your risk of getting it. Depending on the particular type of cancer, genetic testing can give you a better picture of your risk.
  • It’s still questionable whether men should have their prostates checked after age 40, but women should definitely get mammograms after age 40.
  • Diet and exercise affect cancer, but causation remains unclear.

Heart Disease

Heart disease is most often known for being the most deadly disease in the developed world. The reason it falls behind cancers in our analysis is because we looked only at causes of death in persons aged 75 years or younger.

Either way you look at it, the numbers are alarming, especially when you consider that the two best ways to prevent heart disease are through the proper management of diet and lifestyle.

Most people know heart disease as heart attack – the malfunction and subsequent stopping of the heart that can cause sudden death. In the medical world, heart attacks are separated into categories based on what portions of the heart are failing and how that failure was caused. The term heart disease also covers blood clots that can cause death at the brain or lungs, as well as abnormal heart rhythms (ventricular tachycardia and ventricular fibrillation).

Regardless of the particular type of heart disease in question, the risk factors are largely the same. You can start taking a look at your risk factors for using this online Reynolds Risk Score calculator. Most of the numbers you’ll need you can find on a standard lab test. (For the hs-CRP box, put “2” for middle of the average if you don’t have it available. Bump it up to 3 or higher to see how your risk changes. And keep in mind that the calculator only assesses data for people older than 45.)

For a simpler calculator, check out the 10-year CVD risk calculator based on the [Framingham study].

While these calculators can give you a rough idea of where you stand, remember that they are based on epidemiological data and each person’s situation is different. Bring your results in to the doctor’s office if you have concerns about what your numbers say.

Looking at the calculators, it’s pretty clear that there are established markers for assessing risk for heart disease. The major culprits are blood pressure, cholesterol, triglycerides, and body mass index (BMI).

Blood pressure (BP)

BP is an indicator of heart health and should be monitored a little more closely with increased age. The American Heart Association has a good website for blood pressure monitoring at home. Levels below 120/90 are healthy.[JNC # 7 or # 8] It is important to monitor blood pressure over time. An increase in blood pressure over time could be indicative of development of a chronic disease.

Blood pressure is a good indicator across many types of people—that is, athletes with a BP of 140/100 carry the same cardiovascular disease risks as non-athletes.

Blood pressure should not be confused with pulse. Pulse is one measure that is always assessed at the doctor’s office, but is not a strong indicator of health, unless it’s combined with body temperature, respiratory rate, and blood pressure.


Cholesterol is a fat-like substance that is a precursor to many hormones and vitamin D. It is in the bloodstream, but cannot travel freely because it is not water soluble. Thus, it travels in “packages” in the bloodstream. These molecular packages are called high-density lipoproteins (HDL) and low-density lipoproteins (LDL). Perhaps counter-intuitively, higher HDL levels are healthier and lower LDL levels are better. LDL levels should be below 100 mg/dL, HDL levels should be above 60 mg/dL, and total cholesterol levels should be below 200 mg/dL (American Heart Association). If you have certain diseases, such as diabetes, these recommended levels become more stringent.

There’s some confusion about which of these indicators should be used to assess health and indicate possible disease. HDL levels alone? LDL levels? Total cholesterol? A ratio of these? Well, a holistic consideration is warranted when measuring all three of these. A recent study suggests that the best indicator for overall health and risk for cardiovascular disease is the ratio of non-HDL:HDL ratio (Source: Eur J Prev Cardio). I recommend using this measure as opposed to total cholesterol and LDL levels [why?]. But note that this contradicts American Heart Association’s official recommendation of using absolute levels of total cholesterol and HDL.


A triglyceride is comprised of a glyceride molecule attached to three fatty acid chains, and can be found circulating freely in the blood stream. Since these molecules can aggregate in blood vessels, the goal is to keep these levels as low as possible, below 150 mg/dL. These levels can be kept down with a healthy diet and by eating good ratios of fats (olive oil, for example).

[recommended cholesterol and triglycerides also depend on other risk factors, such as diabetes]

Body Mass Index (BMI)

BMI is the most commonly used measure for assessing body fat distribution and is calculated like this:

BMI = weight (kg) / height2 (m2).

For Americans:

BMI = (4.89) weight (lbs) / height2 (feet2)

BMI = (704.55) weight (lbs) / height2 (inches2)

underweight <18.5
healthy weight 18.5-24.9
overweight 25-29.9
obese >30

There are two very good reasons why doctors use BMI to estimate body fat: it’s easy and it’s free.

Unfortunately, BMI is also pretty terrible metric when used incorrectly. And for a metric that’s considered the gold standard in obesity research it’s used incorrectly an awful lot.

To create BMI tables, epidemiologists[] took a cross section of Americans in the 19[]’s and associated BMI with measured body fat. Of course, people in the 19[]’s were generally processed-food-eating, car-driving, desk-working [Caucasians]. So if you don’t live sort of life (or don’t want to live that life), BMI might kind of suck for you.

In fact, BMI is different for different races:

Race Overweight BMI
White 25.0
Black 26.3
Chinese 23.1
Thai 22.1
Ethiopian 20.4
Indonesian 21.8
Polynesian 29.5

(Source: Int J Obes Relat Metab Disord).

And even between same races, BMI can be different depending on what country you came from [].

[Here is a calculator for BMI from the NIH.]

How to really measure body fat

Here’s the obvious, but underpublicized truth: BMI is simply a poor estimate for body fat. If you’re any sort of athlete, you probably know what I’m talking about. Every year at the doctor’s office, your doctor inevitably blushes when he tells you that “according to his measurements” you fall in the overweight BMI category – you know, despite your 6-pack and the fact that your blood markers are all perfect.

There are many better ways to measure body fat, and if it’s not practical to do it in the doctor’s office, you can always do it yourself and bring in your results. Personally, we use a combination of the U.S. Navy taping method and the Jackson skin caliper algorithm[]. Simply looking at comparison photos of people at various body fat levels can also provide a quick guess[]. A recent study found that skinfold thickness was 10% more accurate than BMI in assessing body fat. (Source: Medical Hypotheses).

[what’s the conversion for BMI to body fat?]

What do we mean when we say “diet and exercise”

[walking 30 minutes a day? raising heart rate?][google:10,000 steps a day]

*Note that these levels (BP, cholesterol, triglycerides) are associated with a number of a chronic illnesses in addition to cardiovascular disease. Diabetes, for example, is strongly associated with lower HDL levels, high TGs and high blood pressure. So if you are older than 50, have a family history of cardiovascular disease or diabetes, or are overweight, pay very close attention to these measures.


Diabetes may not be a leading cause of death, but it does have significant impacts on quality of life, on everything from energy levels, to being able to eat certain foods, to possibly losing a limb due to poor blood flow.

Once again, regular light exercise and having a healthy body weight can drastically reduce your risk for diabetes. With that being said, causation remains uncertain. [Sources: ADA, CDC, JECH]

But whereas we normally find it difficult to make specific recommendations about diet, with diabetes, we can safely say that excess sugar consumption likely contributes to poor insulin regulation. The worst culprits are processed sweets, such as sodas, fruit juices, and candy. Next comes highly processed grains like pre-cooked white rice, white bread, and corn-derived processed foods. Finally, even some less processed starches, like pasta and potatoes could make an impact, but the buffering effects of fiber might also mitigate the insulin response.

Besides diet and exercise, diabetes is marked by a resistance to insulin and the best marker for insulin, is blood glucose levels. Here is a sense of the scale of glucose levels:

  • Normal: <100 mg/dl
  • Pre-diabetes: 100-126 mg/dl
  • Diabetes: ≥126 mg/dl

(Source: American Diabetes Association [ADA]).

You get your blood glucose levels checked every time you have blood drawn [no, it depends on what labs are ordered]. That’s one reason it’s important to fast before a blood draw: blood glucose levels are affected by food intake [2nd part is true, but the only lab that usually needs to be done fasting is cholesterol]. We would not recommend monitoring blood glucose at home unless you are pre-diabetes or already have diabetes and must monitor glucose on orders from your doctor.

Although high BMI, high blood pressure, and high triglyceride levels all increase your risk for developing diabetes, the best intervention is to develop a plan of action to improve diet and lifestyle, not monitor glucose.

About the Author

Kevin Liu…


Travis Whitfill contributed research to this article. Travis is currently a graduate student at Yale University studying chronic disease epidemiology.


Top image by Tuna Akçay