The Era of Risk Factors Should End; the Era of Biologic Age Should Begin

Abstract

Introduction: Risk factors, a 75-year-old concept, are instrumental in the management of the general population. Newer biomarkers can explain residual risk and protection from risk. The population needs a new platform to make more comprehensible the importance of managing risk. Biologic age, the number of years left to live, is the platform that will receive the attention of patients. Method: Risk factor odds ratios are used to approximate the years lost to the modifiable risk, calculating a biologic age. Newer biomarkers confirm the predication and can be used to explain the pleomorphic properties of medications and unrealized risk. The biomarkers represent the following biologic processes: repair, inflammation, immune function, hematologic, clotting factors, metabolic-nutritional, organ maintenance, anthropomorphic, environmental, endothelial function, sleep, co-morbidities, frailty, and electromagnetic. Risk factors and biomarkers are ranked in the order of significance in reducing biologic age. Results: A six-step method of patient management using biologic age and biomarkers is presented. Conclusions: Knowledge of risk factors and therapies to improve risk has increased over the last 75 years. Biologic age is more appropriate in explaining the significance of this knowledge and may improve patient compliance to lifestyle changes and medication compliance. Appropriate counseling with utilization of biomarkers of biologic processes, such as high sensitivity-CRP, circulating stem cells, number of co-morbidities, frailty, electrocardiogram, and pulse wave velocity will improve compliance and personalize care. The 6-minute walk should be incorporated into the vital signs due to prognostic significance.

1. Introduction

The era of risk factors began over 75 years ago in Framingham, Massachusetts. The study was funded to determine the cause of vascular disease, myocardial infarction, strokes, peripheral vascular disease, and other afflictions. By recording health data on the 5209 volunteers (roughly half of the population) risk factors, mortality, and morbidity for vascular disease was recorded. The study included risk factors: age, sex, glucose intolerance, total serum cholesterol, cigarette smoking, educational level, systolic and diastolic blood pressure, body mass index, physical activity index, pulse pressure, and electrocardiographic left ventricular hypertrophy [1]. Other risk factors were added as discovered over the decades.

Of the thousands of Framingham studies, the study by Terry illustrates the usefulness of risk factors in counseling patients. Their objective was to examine whether midlife cardiovascular risk factors predict survival and survival free of major comorbidities to the age of 85 [2].

Table 1. Multivariate-adjusted odds ratio (OR) and confidence intervals (CIs) for risk factors related to survival and survival free of major morbidity to age 85 and older.

	Survival to Age 85 (n = 903)			Survival to Age 85 Free of Major Comorbidity * (n = 542)
Risk Factor	OR (95%CI) p-Value
Female	2.00	(1.66–2.41)	<0.001	2.08	(1.66–2.61)	<0.001
Systolic blood pressure (per 20 mmHg)	0.57	(0.50–0.64)	<0.001	---	---	---
Diastolic blood pressure (per 10 mmHg)	---	---	---	0.64	(0.57–0.72	<0.001
Serum cholesterol (per 40 mg/dL)	0.89	(0.79–0.96)	<0.005	0.82	(0.76–0.92)	0.001
Glucose intolerance (present versus absent)	0.3	(0.14–0.64)	<0.002	0.13	(.03–0.54)	0.005
Smoking history (present versus absent)	0.47	(0.39–0.57)	<0.001	0.51	(0.41–0.63)	<0.001
Education (one category increase)	1.25	(1.12–1.39)	<0.001	1.20	(1.06–1.35)	0.004

Note: Risk factors considered in stepwise models were sex, systolic blood pressure, diastolic blood pressure, pulse pressure, anti-hypertensive medication usage, total serum cholesterol, body mass index, glucose intolerance, electrocardiographic left ventricular hypertrophy, smoking, education, and physical activity index. * Comorbidities were myocardial infarction, coronary insufficiency, congestive heart failure, stroke, cancer, and dementia. Adapted with permission—Terry DF, Pencina MJ, Vasan RS, Murabito JM, Wolf PA, Hayes MK, Levy D, D’Agostino RB, Benjamin EJ. Cardiovascular risk factors predictive for survival and morbidity-free survival in the oldest-old Framingham Heart Study participants. J Am Geriatr Soc. 2005 Nov;53(11):1944-50. doi: 10.1111/j.1532-5415.2005.00465.x. PMID: 16274376 [2].

is from their paper with permission, demonstrating the power of risk factors. The age of death of the population studied is represented in Figure 1, again with permission.

The study further demonstrated for men with no risk factors the probability of reaching 85 was 37%, whereas if four or more risk factors were present the chance fell to 2%. In females the chances were much better, 65% and 14%, respectively. Clearly, risk factors can predict the probability of mortality. In order of importance, glucose intolerance is particularly deadly followed by smoking, hypertension, and total cholesterol. The odds ratios are, respectively, 0.3, 0.47, 0.57, and 0.89. A favorable odds ratio of 2.0 demonstrate the female advantage. Rare patients still escape the prediction with 2% of males and 14% of females reaching 85, despite having four risk factors; therefore, other protective factors must be present. These rare patients are a hazard to the general population, demonstrating risk factors failed to predict dire events. The general population believes they too may escape the predicted morbidity and mortality. Risk factors have lost their significance, warranting a new approach.

Over the past 75 years, patients have been identified and inundated with messages of lifestyle modifications and medications. If the risk factors are musical notes, patients have tone deafness. Risk factors must be replaced by something that is more tangible and understood by all. The concept is described by a previously traditional history and exam documentation. ”The patient appears (older, younger, or stated) age”. Chronological age correlates with your birthday. Biologic age correlates with how long you can live.

Based on the Arias expectation of life [3], a 50-year-old male has an expected lifetime of 30 years. If he has no risk factors, his biological age is identical to his true age. If he has all four independent risk factors, his expected life expectancy is reduced to 30 years times the odds ratio of the risk factors.

Table 2. Expectation of life, by race, Hispanic origin, age, and sex: United States, 2012 adapted from Arias E, Heron M, Xu J. United States Life Tables, 2012. Natl Vital Stat Rep. 2016 Nov;65(8):1-65. PMID: 27906644. [3].

	All Races and Origins			White			Black			Hispanic			Non-Hispanic White
Age	Total	Male	Female	Total	Male	Female	Total	Male	Female	Total	Male	Female	Total	Male	Female
0	78.8	76.4	81.2	79.1	76.7	81.4	75.5	72.3	78.4	81.9	79.3	84.3	78.9	76.5	81.2
1	78.3	75.9	80.6	78.5	76.1	80.7	75.3	72.2	78.2	81.3	78.7	83.7	78.3	75.9	80.6
5	74.4	72.0	76.7	74.5	72.2	76.8	71.5	68.5	74.3	77.4	74.8	79.8	74.3	72.0	76.6
10	69.4	67.0	71.7	69.6	67.3	71.8	66.5	63.4	69.3	72.5	69.8	74.8	69.4	67.1	71.6
15	64.5	62.1	66.8	64.6	62.3	66.9	61.6	58.4	64.4	67.5	64.8	69.6	64.4	62.1	66.7
20	59.6	57.3	61.9	59.8	57.5	62.0	56.8	53.7	59.5	62.6	60.0	64.9	59.6	57.3	61.8
25	54.9	52.6	57.0	55.0	52.8	57.1	52.1	49.2	54.6	57.8	55.3	60.0	54.8	52.6	56.9
30	50.1	48.0	52.1	50.2	48.2	52.3	47.4	44.6	49.8	53.0	50.5	55.1	50.1	48.0	52.1
35	45.4	43.3	47.3	45.5	43.5	47.4	42.8	40.1	45.1	48.2	45.8	50.3	45.3	43.3	47.3
40	40.7	38.7	42.6	40.8	38.8	42.7	38.2	35.6	40.4	43.4	41.1	45.4	40.7	38.7	42.5
45	36.1	34.1	37.9	36.2	34.3	38.0	33.7	31.2	35.9	38.7	36.4	40.6	36.1	34.2	37.8
50	31.6	29.7	33.3	31.7	29.9	33.4	29.4	27.0	31.5	34.1	31.9	35.9	31.6	29.8	33.3
55	27.3	25.6	28.9	27.4	25.7	28.9	25.4	23.0	27.3	29.7	27.6	31.4	27.3	25.6	28.8
60	23.2	21.7	24.6	23.3	21.7	24.6	21.6	19.5	23.3	25.4	23.5	26.9	23.2	21.7	24.5
65	19.3	17.9	20.5	19.3	18.0	20.4	18.1	16.2	19.5	21.4	19.6	22.6	19.2	17.9	20.4
70	15.6	14.4	16.5	15.6	14.4	16.5	14.8	13.2	15.9	17.5	15.9	18.5	15.5	14.4	16.5
75	12.2	11.2	12.9	12.1	11.1	12.9	11.8	10.4	12.6	13.8	12.5	14.6	12.1	11.1	12.9
80	9.1	8.3	9.7	9.1	8.3	9.7	9.1	8.0	9.7	10.5	9.4	11.1	9.1	8.2	9.6
85	6.6	5.9	6.9	6.5	5.9	6.9	6.8	6.0	7.2	7.7	6.8	8.1	6.5	5.9	6.9
90	4.6	4.1	4.8	4.5	4.0	4.8	5.1	4.5	5.2	5.5	4.7	5.7	4.5	4.0	4.8
95	3.2	2.8	3.3	3.1	2.8	3.2	3.7	3.4	3.8	3.8	3.3	3.9	3.1	2.8	3.2
100	2.3	2.0	2.3	2.2	2.0	2.3	2.8	2.6	2.8	2.7	2.4	2.7	2.2	2.0	2.3

life expectancy is the source of 30 years selected for the above patient. Life expectancy with risk factors is calculated as:

Life expectancy = (30 years) (0.3) (0.47) (0.57) (0.89) = 2.15 years.

At a chronological age of 50, his biological age is 74.3 with an estimated 2.15 years to live.

A female utilizing the risk, reducing odds ratio of 2 for being female with the same age and risk factors, would have life expectancy of 4.3 years. At age 50 her biologic age is 78.7 years with 4.3 years to live. Patients understand birthdays and years left to live. Patients also understand gambling and probabilities such as odds ratio. Odds do not predict a winner, only what is more likely.

This patient may be more inclined to modify their lifestyle and take risk-reducing medication based on this information. Biologic age should replace risk factors in counseling patients how to manage their health status. Physicians are uncomfortable with assigning a death prediction. This is the likely reason the methodology of Terry was not adopted. Physicians are not making a death prediction only the odds of death unless lifestyle is not changed. Fate is still in the realms of the cosmos.

2. Biomarkers, Personalize Care, and Longevity

Biomarkers, discovered in the last 75 years, further refine risk factors, residual risk, and predict mortality [4]. Biomarkers can differentiate members of a population reaching centurion status [5]. In biologic age models, biomarkers can refine risk factors accounting for known protective and harmful factors. The testing should correlate with age, known and evolving risk factors, inflammation, repair mechanisms, lifestyle, genetic pre-dispositions, and co-morbidities.

Longitudinal data providing a predictive value of biomarkers are sparse since many of these markers are recently available and initially used in a disease model. Currently, these markers can best be used in advisory role to confirm biologic age prediction based on risk factors. As data are collected over time, their predictive value will surpass a risk factor model with the markers providing a personalized prediction incorporating genetic predisposition and genome activation/deactivation. Some markers have a variability being influenced by other environmental influences and may need to be repeated to achieve a true biomarker profile. Other markers are integrated over a lifetime and thus more predictive of years left to live. Electrocardiograms and pulse wave velocity are examples of integrated biomarkers. The electrocardiogram has already predicted biologic age with the help of artificial intelligence [6]. The Figure 2 ranks risk factors and biomarkers in order of their importance in reducing biological age. The order of significance suggests which risk factor modification will gain biologic age. The biomarkers representing biological processes can be modified by medications, thus ranking the importance of the medications. The markers are chosen for prediction of mortality.

The twelve biological processes below and roughly 30 biomarkers listed in

Table 3. List of biologic processes and biomarkers. Italics investigated in Table 4.

Biological Processes	Biomarkers of Biological Processes
Repair	Number of Progenitor Cells	Type of Progenitor Cells
Inflammation and Immune Function	Hs-CRP  Interleukin-6 Tumor Necrosis Factor-Alpha	CBC—red cell diameter width Lymphocyte to neutrophil ratio
Hematologic and Clotting Factors	Clonal Hematopoiesis Polycythemia	Clotting factors D-dimer
Metabolic Nutritional	HgbA1c glucose	Albumin Albuminuria	Lipids Lipoprotein a	CBC-minimum corpuscular volume red cell count
Organ Maintenance	Creatinine Cystatin C	Urinary albumin-to-creatinine ratio	Liver enzymes	BNP
Anthropomorphic	Body Mass Index	Waist circumference
Environmental Factors	Educational Status	Social Economic	Nutritional intake	Minutes of daily exercise
Endothelial function	Coronary Artery Calcium	Pulse Wave Velocities	Endothelial progenitor cells
Sleep	Quantity	Rapid Eye Movement sleep
Number of Co-Morbidities	CAD, PVD, CVD CRI, COPD, Liver Disease	Sleep Apnea
Frailty	Six-Minute Walk	Hand Grip Strength
Electromagnetic	ECG

Hs-CRP—high sensitivity C-reactive protein, CBC—complete blood count, Hgb—hemoglobin, BNP—brain natriuretic peptide, CAD—coronary artery disease, PVD—peripheral vascular disease, CVD—cerebral vascular disease, CRI—chronic renal insufficiency, COPD—chronic obstructive coronary disease, ECG—electrocardiogram.

pales to the potential number of biomarkers which include 2 million proteins, 300 cell types of 50 trillion cells, 45/46 chromosomes, billions of base pairs, 20,000 genes, signaling hormones, vitamins, minerals, microbiomes, and countless electrical impulses and imaging tests. Table 3 is a promising list, covering most biological processes of repair and protection from the environment. Nutrition, sleep, and exercise, the pillars of health, are also represented. The greater the number of co-morbidities increases biologic age [7]. Frailty is also included due to its dire prediction but is not necessarily age dependent and can occur at any age. In females, frailty is more predictive than co-morbidities of advanced biologic age [7,8]. Only a handful of potential markers in Table 3 (Italics) were selected for

Table 4. Biomarkers and all-cause mortality.

Biomarker	Change with Age	Mortality Prediction	ODDs Ratio of All-Cause Mortality	Correlation with Risk Factors	Integrated over Years Lived	Ref.
Number of Co-Morbidities	Increases	yes	0.89—men 1.0—women	yes	yes	[7]
Frailty	Not age dependent	yes	0.84—men 0.88—women	yes		[7,8]
Six-Minute Walk distance	Decreases	yes	>414 to <290 1.0 to 0.37			[9]
Hs-CRP ≥ 2.0	Increases	yes	0.7	yes		[10]
Number of Progenitor cells	Decreases	yes	0.61	yes		[11,12,13,14,15,16]
Waist Circumference	Increases	yes	0.60—men 0.64—women	yes		[17,18]
Coronary Artery Calcium 400	Increases	yes	0.9	no	yes	[19]
Short Sleep < 7 Long Sleep	Increases < 65 years	Yes U shaped	0.9			[20,21]
Pulse Wave Velocity	Increases	Yes	0.16	Yes	Yes	[22]
ECG Findings RAE LAE LVH	No	Yes Yes Yes	0.67 0.63 0.53	yes	yes	[23]

Hs-CRP—high sensitivity C-reactive protein, ECG—electrocardiogram, RAE—right atrial enlargement, LAE—left atrial enlargement, LVH—left ventricular hypertrophy.

.

3. Definitions and Realizations of Table 4

Odds = (probability of survival)/(1 − probability of survival)

Odds ratio = odds in one group/odds in unexposed group

Odds ratio approximated from hazard ratio − hazard ratio/(1 + hazard ratio)

Hazard ratio is the hazard rate of one group’s outcome compared to a hazard rate of a control. It represents an instantaneous rate. The separation of survival or mortality curves at a particular time.

Hazard rate is probability of death over an interval—the number of deaths over the interval by the survivors at the beginning of the interval.

The odds ratio for this study was computed from epidemiological papers listed and from the above definitions. All had appropriate confidence intervals. The populations were not identical, so these findings are an approximation. The table reveals many interesting aspects of biological markers. The lower the odds ratio the more influence it has on mortality, the greater the number of life expectancy years lost. The odds ratios of biologic markers and risk factors suggest the order of significance is pulse wave velocity, diabetes mellitus, less than 290 m in the 6-min walk, smoking, ECG LVH, hypertension, waist circumference, circulating stem cells, Hs-CRP, followed by the rest as noted in the Figure 2.

The electrocardiogram represents integration over the entire lifespan. The changes reflect previous cardiac events and recent and past exposure to hypertension volume excess due to environmental salt, the stiffness of blood vessels from hypertension, inflammation, and nutrition stresses. Pulse wave velocity is also an integration factor over time and incorporates inflammation, nutrition, hypertension, exercise, and most of the risk factors into the stiffness of blood vessels, a potent risk for mortality. Both the electrocardiogram and pulse wave velocity represent structural changes. Similarly, but less predictive of death, are the integrative factors of co-morbidities and coronary artery calcium.

The value of the odds ratio is directing therapy toward the biggest benefits and helps explain the pleomorphic effects of medications. The odds ratio for total cholesterol is 0.9, which is not particularly deadly. How can you explain the significant mortality benefit of HMG CoA reductase inhibitor medications if cholesterol is not that deadly? The pleomorphic effects of statins include lowering Hs-CRP and increasing circulating stem cells with odds ratio, respectively, of 0.7 and 0.61, targeting protective therapies. Colchicine also lowers Hs-CRP, exercise increases circulating stem cells. These therapies need to be added to patients who have residual risk. The biomarker profile of an individual patient will confirm biologic age and predict protective medications.

Diabetes as a risk factor is responsible for most years of life lost, therefore increases biologic age. Sodium–glucose cotransporter 2 inhibitors decrease diabetic risk, preserves kidney function, and lowers the inflammatory marker Hs-CRP. The pleomorphic favorable effect on multiple biomarkers of this medication is demonstrated by the remarkable outcomes in clinical trials. These medications decrease years lost to diabetes. Additional medications listed in

Table 5. Medications or interventions with favorable endothelial improvement.

Medication Intervention	Lowers Hs-CRP	Increases Stem Cells	Anti-Fibrotic	Ref.
Exercise	Yes	Yes		[24,25]
HMG CoA Reductase Inhibitors	Yes	Yes		[26]
Aldosterone Inhibitor	Yes	Yes	Yes	[27,28]
Clopidogrel	Yes	Yes		[29]
ACEI ARB ARNI	yes	Yes		[30,31,32]
Sodium–Glucose Co-transporter 2 Inhibitor	yes	Yes		[33,34,35]

ACEI—angiotensin converting enzyme inhibitor, ARB—angiotensin receptor blocker, ARNI—angiotensin receptor blocker/neprilysin inhibitor.

have favorable responses in vascular repair.

4. Method of Management of Patients

Hypothetical Case Illustration

A 50-year-old Hispanic male presents to the clinic for evaluation. He states he is reluctant to take medication. He has been sedentary but has no complaints other than muscle and joint pain. He is taking no medication but admits to having diabetes, hypertension, actively smoking, and elevated cholesterol.

Physical exam reveals an “older than stated age” individual with increased abdominal girth, large neck consistent with risk of obstructive sleep apnea, ear lobe creases, and yellow tinged fingernails. Vital signs—pulse 82, blood pressure 170/80, weight 225 pounds, 6-minute walk obtained at check in was 225 feet [36]. Physical exam revealed jugular venous distention of 12 cm, without Kussmaul’s sign, positive systolic pulsation of the pulmonary artery, a hypertrophied displaced point of maximal impact, and auscultation revealed an S_4.

Laboratory—Hgb A1c 7.8, creatinine 1.8, increased urinary albumin to creatinine ratio, Hs-CRP of 4, total cholesterol of 270.

He consents to an estimate of his biologic age.

• Determine risk factors age gender and ethnicity

• 50-year old Hispanic male with diabetes, hypertension, smoking, and elevated total cholesterol

2.

Determine expected years of life chronological age with no risk factors, Table 3

• 31.9 expected years, chronological age 50, expected age at death 81.9 with no risks

3.

Determine the risk factor weighted reduction of expected years due to risk factor odd ratios from Table 1

• 31.9 (0.3)(0.57)(0.47)(0.89) = 2.28 years of expected life due to risk factors biologic age is 81.9–2.28 = 79.6 years

The patient is shocked that he may only have 3 years to live and is disbelieving. Biomarkers are reviewed

• Hs-CRP = 4
• Co-morbidity renal insufficiency, sleep apnea, renal insufficiency
• Six minute walk—225 feet
• Elevated albumin to creatinine and creatinine of 1.8

4.

Biomarker panel confirms dire expected life of just over 2 years

5.

The patient is now willing to be compliant with recommended treatment plan

• Exercise, calorie and carbohydrate reduced diet, smoking cessation, Hmg CoA reductase inhibitors, spironolactone, Sodium–Glucose transporter 2 Inhibitor, ARB and Clopidogrel (primary prevention due to high risk), sleep study

6.

Follow-up for compliance and repeat biomarkers for improvement

• Smoking discontinued, exercising 45 min per day, 25 pound weight loss, compliant with medications Hs-CRP -1.8, Six-minute walk increases to 450 feet
• Steady improvements of biomarkers will improve life expectancy

5. Caveat

In the general population biomarkers are limited to lipids, Hgba1c, blood count, comprehensive metabolic profile, urinary albumin. The other useful biomarkers are reserved for diseased populations. For the above method to be applied, the biomarkers listed must be obtained to confirm the health/unhealthy status of the patient. These same markers need to be followed for trends. Circulating stem cells are a powerful predictor of the effect of age and risk factors but are generally reserved for oncology patients. Circulating stem cells have a strong predictive role in the general population, reflecting all the risk factors. This testing should be made available. The value of this test is in those individuals who have events without apparent risk.

The calculation of biologic age was by Terry [2] based on data from Framingham. The justification and validity were established in their paper with appropriate confidence intervals. The beneficial outcome of using probability is the number ranges from 0 to 1 with mortality as a hard endpoint. Probability can be compared, giving weight to various risk factors. Currently, biomarkers are not used in the calculation of the biologic age only to confirm dire predictions or account for protective factors. Biomarkers can also be expressed as probabilities of all-cause mortality. There is not enough data in the well populations to use biomarkers in the model. In the future as prospective data are obtained, a new model will incorporate biomarkers. More importantly medications and new therapies will be tested for how the biomarkers are altered in addition to clinical outcomes. Biomarkers will explain the pleiotropic effect of medication and personalize care.

6. Conclusions

Knowledge of risk factors and therapies to improve risk has increased over the last 75 years. Biologic age is more appropriate in explaining the significance of this knowledge. Odd ratio suggests which risk factors and biomarkers are more deadly. Biomarkers of fundamental processes such as repair and inflammation can further refine biologic age. Utilization of Hs-CRP and circulating stem cells directly reflects the risk factors. Altering risk factors with lifestyle changes and medications can alter processes of repair and inflammation, extending biologic age. In addition, the number of co-morbidities, frailty, ECG findings, and pulse wave velocity can indicate a shorter biologic age. The 6-minute walk with strong predictive value should be incorporated into the vital signs {36].