ARE the LONGEVITY BENEFITS of LIFE AHEAD VALID?
Abstract: Life Ahead forecasts benefits from Life Style factors that may seem surprisingly high based on past experience with statistical methods and risk factors. A review suggests that these computed benefits probably are quite conservative. A problem, however, is that computed benefits assume lifetime use of habit and risk changes, and full lifetime maintenance of habits may not be achieved by many of those in our population.
The benefits of 15 or more added years of life often assigned by Life Ahead for better health habits seem intuitively high, and are much higher than those estimated from conventional health study statistics. Life Ahead also suggests that risk of Heart Disease often can be reduced ten fold or more and that Cancer risk often can be reduced 4 fold or more. A common question thus becomes “Is this true? Can such benefits really be achieved? Much attention has been given these questions in the design of the program. And it may overestimate some benefits. Yet benefits more often will be underestimated.
Life Ahead is an extensive MULTI-FACTORIAL model. It includes in engineering format how effects of up to a hundred different factors that determine risk rates of a dozen different causes of death can combine mathematically to determine a probable average length of life of a population. The engineering format provides a major technological advance vs. the usual statistical type methods now used in the Health Research field that are incapable of the detailed physical representation needed. The Life Ahead coronary disease risk is a product of 29 different risk factors. Individual health studies usually measure a maximum of one or two different effects on one disease at certain restricted ages. Such narrowly defined statistical risks of individual factors do not value their effect on length of life. And no individual health study is large enough to identify directly more than a very few risk factors on more than one or two diseases. When multiple factors ARE considered in conventional research studies, they nearly always show the expected combination of results. And at least six different considerations could cause present Life Ahead forecasts of Well-Days of life to be conservative.
First, Life Ahead identifies risks on most factors from a careful compilation of and analysis of all useful research studies and measurements found published. All risk factors are amount or ‘dose’ related. A higher amount of a factor as vitamin, exercise, etc is associated with a larger effect on disease. The trend of benefit vs. amount usually is clear and suggests that values higher than the highest measured amount of a factor such as a vitamin or cardiofitness will produce still larger benefits. Life Ahead limits a maximum benefit of every included factor to that actually measured by confirmed study results. In other words, every formula has a limiting benefit set that does not exceed a usual maximum value actually confirmed by multiple research. It remains likely that higher benefits from a number of factors can be obtained than those now accepted by the program. Most meta analyses develop only an average statistic of risk from multiple studies published. Life Ahead moves beyond the meta concept by identifying the dose relationship of the risk to amount of agent used, and the likely biochemical mechanism via which this risk is produced.
A serious problem in the health research today is the unbalanced attention give to one or two studies of a subject recently published compared with the far greater study evidence from prior research. No convenient records are being kept of study results and their error margins on each subject and past research is becoming near completely forgotten. The claim usually is that "My recent study is more rigorous." Health research inevitably will disagree because the reported error margin of the risk ratios obtained from it too often average 50% or more. The idea that one study with such an error margin is "better" or more "rigorous" than another that has the same 50% error margin is absurd. They each are crude approximations. And the best we really can do from this error prone research is to get a best objective result from as many studies that we can.
The Antioxidant Model recognizes further that only some maximum amount of antioxidants can be chemically beneficial. It thus imposes a limit of antioxidant benefit possible from any amount or combination four key antioxidants. This limits benefits to values substantially below those computed for the summed effects of various individual antioxidants.
Second, a problem with all health models is that of duplication of benefits. Benefits of exercise accrue from a summation of its effects on Total and HDL cholesterol and blood pressure in addition to Cardiofitness. Life Ahead attempts to avoid such duplications. For example, if measurements of cholesterol are included, the fraction of risk due to the effect of exercise on cholesterol is subtracted out from the total effect computed for exercise and cardiofitness. Similarly increased body weight increases blood pressures. Thus if blood pressure measurements are provided, the portion of risk caused by the effect of weight on blood pressure is subtracted from that attributable to weight. Another example: Alcohol increases HDL and this comprises part but not all of the benefit of alcohol in reducing risk of heart disease. The model recognizes and removes this duplication if a measurement of HDL is provided. .
Third, Life Ahead recognizes the reality of collapsing risks, and that risks never can reach zero. Research health risks usually are statistically measured individually and on populations of about age 50. Consider a reduction of cholesterol that reduces a risk 50%, an increase in exercise that reduces the same risk 50%, and use of antioxidants that reduce the same risk 50%. A person can perform each action. But it is obvious that risk cannot be reduced by more than 100%. The Life Cycle computation that manages overall risks of each individual disease and their total risk at each year of life solves this problem. The rate type computations of Life Ahead usually show that two 50% reductions will reduce overall risk by 75%, and three 50% reductions would achieve an 87.5% lowering of overall risk. This has important implications. A person that obtains the first 50% from being fit now gains only another 25% benefit for lowering cholesterol and just 12.5% more from antioxidants. Each factor then contributes about a 29% effective reduction. Combinations of any number of multiple risks can approach but never achieve a risk reduction of 100%. The Life Ahead allocation of benefits from improved actions prorates all risks from individual factors to conform to a developed total. If only a single beneficial factor is changed, its benefit will be larger and possibly much larger than this prorated risk. If a single beneficial factor is subtracted from an overall risk of multi-factors, its debit to the total will be smaller than its prorated value.
It does not matter which order these benefits are taken, simple mathematics dictates that benefits must collapse as the number of benefits multiply. Stated another way, a person that stays cardiofit will achieve less benefit from diet, and can have a bit more flexibility in what he or she eats for a reasonable health program. This shows that the risk reduction numbers from population health research studies that usually are measured as single factors on populations at about age 50 are not widely applicable. The real effect of a health action can vary widely depending on how this action combines with other actions. And up to 20 different factors can contribute to an overall health risk of one disease. The benefit of some health actions as for example reducing serum cholesterol and smoking also can vary substantially with age. (See the Health Library).
Any health model that identifies benefits in terms of a set number of added years of life for a given change in factor cannot be valid. Any attempt to include multiples of many such factors will produce absurd results because the results of health factors simply cannot be added. A number of added years for some health action can change more than ten fold depending on number of factors involved, user age, gender, etc. At least one such developer found that after claiming specific "years of life" benefits for dozens of individual factors the total years of such benefits had to divided by factors ranging from 5 to 10 to keep the results from being absurd. Although not useful quantitatively or scientifically, such simplistic health models still can be useful for advising people of their areas of health risk and for stimulating them to take action.
Fourth, the Life Cycle and Life Table method of Life Ahead recognizes that reductions in risk of any disease inevitably will increase risks and rates of other diseases. For example, if cardiovascular disease death is reduced, Life Ahead computes at each following year that a larger number of survivors will be exposed to and will suffer cancer and other diseases. This obvious factor usually is disregarded in statistical comparisons and models. As years of life into the future expand, this interaction of diseases with each other becomes very large.
Fifth, the present Life Ahead although including the 15 most important causes of US deaths and 4 additional health diminishing problems, still does not consider the 25% of deaths that occur from a large number other causes. And because little useful research on lifestyle habits exists for most of these smaller causes, life habits do not reduce the computed risk from this important segment. . This limits maximum average population computed life expectancy for men and women of age 50 to about ages 92 and 95 even for near complete elimination of all cardiovascular disease most cancer and maximum of other factors. Almost certainly many of these other not-considered diseases are affected by life habits. If risk of these other causes of death could be reduced 50% by life habits, Life Ahead could show an additional potential 6 years of Well-Days for good habits and risks. Good health habits then might extend potential Well-Years of life at age 50 to an average of age 100.
Sixth, deaths from CVD and cancer are recognized in Life Ahead as duration related. Life Ahead effects of diet on both CVD and cancer and for effects of exercise on cancer develop gradually as increase in rate of disease modifiers, and not as the 'instantaneous' changes in risk assumed from conventional population study risk measurement. Life Ahead recognizes that it can take 7 -10 years before continued changes in habits produce substantial accumulated changes in risk from antioxidants. Further, a limit of 20 years is set for an accumulation of maximum duration related risks. The data on diabetes confirms at least 30 years of factor duration can continue to change risk, and it is likely that such benefits continue over full lifetime. This method thus produces more conservative estimates of factor benefits than the conventional assumption that risk responds immediately and in full to a change in risk factor.
Another question of this author - and I am sure that others will raise is: Life Ahead shows that heart disease risks are reduced more than twenty times and to near the vanishing level for good habits. Can this really be true? A review of the computations shows that risk is reduced via good habit Cardiofitness by about 70%. Yet nine actually published research risk ratios showed risk reductions of 75-88% for best Cardiofitness. Life Ahead computes that optimum dietary antioxidant reduces risk another 50%. Yet each of the 4 antioxidants used individually accomplished conservative risk reductions of 25-30% and a combination of these risks would compute a reduction of over risk by 75%. Adequate Omega-3 fats can contribute another at least a 50% reduction - the research available shows a higher reduction than this from the Mediterranean diet..
Now consider further reductions in risk for use of aspirin and some alcohol that should total another 50%. Bring in improved folic acid, magnesium sodium, and potassium in diet that in total easily can contribute another 50% reduction in risk. Even recognizing the concept of collapsing benefits discussed elsewhere, the remaining risk of heart disease from a careful selection of good habits for an adequate duration of time must become very small. Most of these factors produce benefits by differing mechanisms. Even if there is some duplication of benefits involved among these factors, the remaining risk of the disease for good habits would continue from obvious mathematics still to be very small. This does not mean that some who do everything right will not suffer. Even one in twenty or one in fifty of those suffering heart disease now represents a very large number of individuals.
Some Possible Optimism: As opposed to these conservative considerations now included in the program, some assumptions could be optimistic. Perhaps most important is the fact the computations assume a full lifetime devotion to each included benefit. A substantial part of computed benefits may accrue for good habits during the last decade of life. Many people are unlikely to maintain habits for this duration. As example, some who are older may not be able to exercise because a disabled leg or arthritis.
There also remains some probable duplication of factors in the present Model. For example, the effects of baldness on heart disease for men and risks of time of sleeping could duplicate a risk included as a family risk of heart disease. Risk reductions from some minerals for for example calcium for example could be due to antioxidant behavior and thus might be offset by good values of other antioxidants. Some other factors in Life Ahead for which mechanisms are inadequately defined could involve duplication.
On balance, it seems likely that forecasts by Life Ahead of potential health benefits should be reasonable and quite conservative. A major deficiency in this and every other health risk estimate today is that of genetics. Life Ahead does approach this problem by a detailed model for Family History risks. The program also identifies likely genetic risks for Total and HDL cholesterol and for Cardiofitness that have not been recognized in the past. Thus the many factors in and format of the present model should narrow the remaining level of unknown genetic risks. Yet, genetics still comprises the #1 risk for cancer. Hopefully, use of DNA some time in the future will provide a much more accurate identification of the genetic risk that remains as the #1 probable unknown now in Life Ahead and in all other estimates of a probable length of individual's life.
