Biological Age: Difference between revisions

Revision as of 03:22, 30 January 2024

Biological Age (BA) is a concept used to assess an individual's aging status, offering a more nuanced understanding than Chronological Age (CA). CA refers simply to the amount of time that has elapsed since a person's birth, while BA provides a measure of aging based on various physiological, biochemical, and molecular factors. This distinction is crucial because individuals of the same CA can exhibit significantly different aging processes and health statuses.

Key Aspects of Biological Age

Biomarkers: BA is typically determined by analyzing a range of biomarkers. These can include genetic markers, epigenetic alterations, cellular senescence, telomere length, metabolic markers, and more. The specific biomarkers chosen depend on the method of estimation and the focus of the study.
Health and Functionality: BA reflects the functional state of an individual's organs and systems. A lower BA compared to CA might indicate better health and lower risk for age-related diseases, whereas a higher BA suggests accelerated aging and potentially increased health risks.
Variability: Unlike CA, which is uniform and progresses at a constant rate (one year per year), BA can vary significantly between individuals. Factors such as lifestyle, genetics, environment, and disease can influence the rate at which a person's biological systems age.

Importance in Research and Medicine

Research Tool: In scientific research, BA is valuable for understanding the aging process, identifying aging biomarkers, and evaluating the effectiveness of anti-aging interventions.
Clinical Applications: In a clinical setting, BA can be used to assess an individual's overall health status, predict the risk of age-related diseases, and personalize healthcare and treatment plans.

Estimation Methods

Biological age estimation has emerged as a significant tool in gerontology, aiming to provide a more accurate measure of aging than chronological age.

The accurate estimation of BA has significant implications for clinical practice, including predicting disease onset and prognosis, improving the quality of life for the elderly, and promoting successful aging. Each method offers unique insights, and a comprehensive understanding of these methods can lead to better clinical decision-making and more effective interventions for aging-related conditions.

Various methods have been developed to estimate BA, each with its unique approach and criteria:^[1]

Multiple Linear Regression (MLR) is a statistical technique that estimates BA by relating several independent variables (biomarkers) to a dependent variable (CA). In this method, CA is used as a criterion for selecting biomarkers and is treated as an independent index.
Principal Component Analysis (PCA) is another statistical technique used in BA estimation. PCA reduces the dimensionality of the data by transforming multiple biomarkers into a set of linearly uncorrelated variables, known as principal components.
Hochschild’s Method differs from MLR and PCA by making CA an independent variable. It aims to estimate BA by adjusting CA based on specific biomarkers.
Klemera and Doubal’s Method (KDM) shares a similar concept with Hochschild’s method but uses a more complex statistical approach. It treats CA as an independent variable and incorporates multiple biomarkers to estimate BA.

Role of Chronological Age

The role of Chronological Age (CA) in the estimation of Biological Age (BA) varies significantly between different methods. In MLR and PCA, CA is used as a criterion for selecting biomarkers. This means that the biomarkers are chosen based on how well they correlate with CA. In this context, CA is not an independent variable in the statistical model but rather a reference point against which the predictive power of biomarkers is measured.

In contrast, Hochschild’s method and KDM treat CA as an independent variable. This means CA is directly incorporated into the model as one of the variables that predict BA.

Comparison

Comparisons among MLR, PCA, Hochschild’s method, and KDM

Method	Proposed	Core concept	Advantage	Disadvantage
MLR	1965^[2]	Aging biomarkers are determined by the correlation with CA using MLR model	MLR is the preliminary method and is easy to operate	The standards of aging biomarkers lead to the paradox of CA MLR also distorts the BA at the regression edge and ignores discontinuity in the aging rate^[3]^[4]^[5]
PCA	1985, Nakamura^[6]	PCA uses fewer uncorrelated variables to explain the main variance	Biomarkers are uncorrelated variables^[7] PCA avoids the influence of regression edge in MLR^[4]	PCA cannot avoid the paradox of CA and some statistical deficiencies of MLR^[7]
Hochschild’s method	1989, Hochschild^[8]	Hochschild’s method aims to select aging biomarkers according to their effects on life expectancy^[8]	Hochschild’s method solves the paradox of CA Hochschild’s method avoids statistical problems of MLR	Hochschild’s method is nonstandard and relatively complicated Hochschild’s method is not based on the definition of BA A large number of subjects are required when this approach is adopted for another system^[9]
KDM	2006, Klemera and Doubal^[7]	KDM is based on minimizing the distance between m regression lines and m biomarker points in an m-dimensional space of all biomarkers^[7]	KDM performed better than CA^[10] KDM is precise when compared with other methods^[10]^[9]^[11] KDM solves the paradox of CA^[10]^[9]	The calculation of KDM is complicated^[9]

Used Biomarkers

Comparisons of aging biomarkers among MLR, PCA, Hochschild’s method, and KDM

Organ system	PCA	MLR	Hochschild’s	KDM
Cardiovascular system	Pulse pressure{{pmid\|19940465}}{{pmid\|24659482}}{{pmid\|29188884}}
	Systolic blood pressure{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|12672981}}{{pmid\|18840798}}{{pmid\|8803500}}{{pmid\|28110151}}	Systolic blood pressure{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|5841151}}{{pmid\|950448}}{{pmid\|28110151}}		Systolic blood pressure{{pmid\|28110151}}
	Heart rate{{pmid\|2737197}}{{pmid\|2282902}}
	Intima-media thickness{{pmid\|19940465}}{{pmid\|24522464}}{{pmid\|24659482}}
	Maximum internal diameter of carotid artery{{pmid\|24522464}}{{pmid\|24659482}}
	End diastolic velocity{{pmid\|19940465}}
	Mitral valve annulus ventricular septum of the peak velocity of early filling{{pmid\|24522464}}
	Mitral valve annulus lateral wall of peak velocity of early filling{{pmid\|19940465}}
	Mitral annulus peak E anterior wall{{pmid\|24659482}}
	Ratio of peak velocity of early filling to atrial filling{{pmid\|19940465}}

Respiratory system	VO2 max{{pmid\|22433233}}{{pmid\|18597867}}
	Forced expiratory volume in 1 second{{pmid\|22433233}}{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|12672981}}{{pmid\|18840798}}	Forced expiratory volume in 1 second{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|28110151}}	Forced expiratory volume in 1 second{{pmid\|2583248}}	Forced expiratory volume in 1 second{{pmid\|28110151}}
	Forced vital capacity{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|8803500}}	Forced vital capacity{{pmid\|20005245}}	Forced vital capacity{{pmid\|2583248}}
	Maximal mid expiratory flow rate 75/25{{pmid\|24522464}}	Vital capacity{{pmid\|5841151}}
Nervous system	Trail making test{{pmid\|24522464}}{{pmid\|29188884}}
	Digital symbol test{{pmid\|20005245}}{{pmid\|24659482}}	Digital symbol test{{pmid\|20005245}}
	Memory test linking names with faces{{pmid\|20005245}}	Memory test linking names with faces{{pmid\|20005245}}
	Memory test: which picture is at what place{{pmid\|20005245}}	Memory test: which picture is at what place{{pmid\|20005245}}
	Speed test: pointing icons from 1 to 15 sequentially, mixed in random positions{{pmid\|20005245}}
	Visual reaction time{{pmid\|20005245}}{{pmid\|22433233}}		Visual reaction time{{pmid\|2583248}}
			Sequence of lamps{{pmid\|2583248}}
			Alternate button tapping time with/without decision{{pmid\|2583248}}
			Movement time with/without decision{{pmid\|2583248}}
Renal system	Blood urea nitrogen{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|29380856}}{{pmid\|18840798}}{{pmid\|8803500}}	Blood urea nitrogen{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|28110151}}		Blood urea nitrogen{{pmid\|28110151}}
		Serum creatinine{{pmid\|28110151}}		Serum creatinine{{pmid\|28110151}}
	Cystatin C{{pmid\|19940465}}{{pmid\|24522464}}{{pmid\|24659482}}{{pmid\|29188884}}
Liver	Serum albumin{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|29380856}}{{pmid\|18840798}}	Serum albumin{{pmid\|950448}}
	Glutamic oxaloacetic transaminase{{pmid\|2737197}}{{pmid\|2282902}}	Glutamic oxaloacetic transaminase{{pmid\|28110151}}		Glutamic oxaloacetic transaminase{{pmid\|28110151}}
	Glutamic pyruvic transaminase{{pmid\|8803500}}
	Ratio of albumin to globulin{{pmid\|12672981}}
	Lactate dehydrogenase{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|8803500}}
		Serum globulin{{pmid\|950448}}
		Alkaline phosphatase{{pmid\|950448}}
Hematologic system	Erythrocyte sedimentation rate{{pmid\|18597867}}	Erythrocyte sedimentation rate{{pmid\|950448}}
	Mean corpuscular hemoglobin{{pmid\|12672981}}
	Red blood cell count{{pmid\|18840798}}
	Hematocrit{{pmid\|17921421}}
	Hemoglobin concentration{{pmid\|2737197}}{{pmid\|2282902}}
	Fibrinogen{{pmid\|19940465}}
		Ferratin{{pmid\|28110151}}		Ferratin{{pmid\|28110151}}
Metabolism	Glycosylated hemoglobin{{pmid\|18597867}}
	Glucose{{pmid\|12672981}}{{pmid\|8803500}}	Glucose{{pmid\|950448}}
	Low-density cholesterol{{pmid\|18597867}}
	Atherogenic index{{pmid\|2737197}}{{pmid\|2282902}}
	Triglyceride{{pmid\|2282902}}	Triglycerides{{pmid\|6967883}}
	Total cholesterol{{pmid\|8803500}}	Total cholesterol{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|5841151}}{{pmid\|28110151}}		Total cholesterol{{pmid\|28110151}}
Muscle and fat	Grip strength{{pmid\|20005245}}{{pmid\|22433233}}	Grip strength{{pmid\|20005245}}{{pmid\|5841151}}
	Soft lean mass{{pmid\|22433233}}
	Waist circumference{{pmid\|22433233}}{{pmid\|18597867}}	Waist circumference{{pmid\|28110151}}		Waist circumference{{pmid\|28110151}}
	Percent body fat{{pmid\|18597867}}
Sensory system	Hearing threshold{{pmid\|18597867}}
	Highest audible pitch{{pmid\|20005245}}	Highest audible pitch{{pmid\|20005245}}	Highest audible pitch{{pmid\|2583248}}
		Light extinction test{{pmid\|5841151}}
		Visual acuity{{pmid\|5841151}}
		Auditory function{{pmid\|5841151}}
		Vibrotactile sensitivity{{pmid\|5841151}}	Vibrotactile sensitivity{{pmid\|2583248}}
		Auditory reaction time{{pmid\|20005245}}	Auditory reaction time{{pmid\|2583248}}
		Focal range test using a Landolt ring{{pmid\|20005245}}	Visual accommodation{{pmid\|2583248}}
Genetic index	Telomere restriction fragment{{pmid\|24659482}}

References

↑ ^{Jump up to: 1.0} ^1.1 Jia L et al.: Common methods of biological age estimation. Clin Interv Aging 2017. (PMID 28546743) [PubMed] [DOI] [Full text] Abstract
At present, no single indicator could be used as a golden index to estimate aging process. The biological age (BA), which combines several important biomarkers with mathematical modeling, has been proposed for >50 years as an aging estimation method to replace chronological age (CA). The common methods used for BA estimation include the multiple linear regression (MLR), the principal component analysis (PCA), the Hochschild's method, and the Klemera and Doubal's method (KDM). The fundamental differences in these four methods are the roles of CA and the selection criteria of aging biomarkers. In MLR and PCA, CA is treated as the selection criterion and an independent index. The Hochschild's method and KDM share a similar concept, making CA an independent variable. Previous studies have either simply constructed the BA model by one or compared the four methods together. However, reviews have yet to illustrate and compare the four methods systematically. Since the BA model is a potential estimation of aging for clinical use, such as predicting onset and prognosis of diseases, improving the elderly's living qualities, and realizing successful aging, here we summarize previous BA studies, illustrate the basic statistical steps, and thoroughly discuss the comparisons among the four common BA estimation methods.
↑ Hollingsworth JW et al.: Correlations between tests of aging in Hiroshima subjects--an attempt to define "physiologic age". Yale J Biol Med 1965. (PMID 5841151) [PubMed] [Full text]
↑ Dubina TL et al.: Biological age and its estimation. II. Assessment of biological age of albino rats by multiple regression analysis. Exp Gerontol 1983. (PMID 6873212) [PubMed] [DOI] Abstract
Multiple regression model of biological age (BA) theoretically gives agreement with the main concept of BA. When assessment of BA is based on the model, the age being in regression center, the method provides satisfactory results, whereas BA estimates of individuals in extreme age groups are erroneous. Investigation of male and female Wistar rats of age 5-29 months showed the BA estimates calculated from 4-10 physiological indices in young (5-7 mo) animals are overestimated, and in old (24-28 mo) animals are underestimated. Coincidence of average BA in one-age group of animals with its chronological age served as a criterion for the correspondence of the estimate to "real" BA. The paper also examines the following questions: the necessary and sufficient number of physiological indices; the sample size from the intact animal population to establish normal aging standard; the relationship between BA and animal weight.
↑ ^{Jump up to: 4.0} ^4.1 Nakamura E et al.: Assessment of biological age by principal component analysis. Mech Ageing Dev 1988. (PMID 3226152) [PubMed] [DOI] Abstract
A method of assessing biological age by the application of principal component analysis is reported. Healthy individuals (462) randomly selected from about 6000 men who had taken a 2-day health examination were studied. Out of the 30 physiological variables examined in routine check-ups, 11 variables were selected as suitable for the assessment of biological age based on the results of factor analysis and the physiological meaning of each test. This variable set was then submitted to principal component analysis, and the 1st principal component obtained from this analysis was used as an equation for assessing one's biological age. However, the biological age calculated from this equation is expressed as a score, so the estimated score was transformed to years (biological age) using the T-score idea. The biological age estimated by this method is practically useful and theoretically valid in contrast with the multiple regression model, because this approach eliminates and overcomes the following 2 big problems of the multiple regression model: (1) the distortion of the individual biological age at the regression edges; and (2) a theoretical contradiction in that a perfect model will merely be predicting the subject's chronological age, not his biological age.
↑ Webster IW & Logie AR: A relationship between functional age and health status in female subjects. J Gerontol 1976. (PMID 950448) [PubMed] [DOI] Abstract
A multiple regression equation was used to predict age from seven clinical variables in 1080 apparently well female subjects. A multiple correlation coefficient of R = 0.77 was achieved by five of the variables: timed forced expiratory volume, systolic blood pressure, plasma urea nitrogen, cholesterol, and alkaline phosphatase. On the basis of selection by medical questionnaire responses and other objective criteria, 9% of the subjects were nonsmokers and healthier than the rest. These selected subjects showed a significant reduction in preducted age. Within this group, subjective perception of health was associated with differences in predicted age: poor health with an increase and good health with a decrease in functional age. This study of functional age was based on the healthiest segment of the population in order to minimize the effect of overt pathological processes on the aging rate. An association has been demonstrated between health impairment and predicted age as a measure of the aging rate.
↑ Nakamura E et al.: Biological age versus physical fitness age. Eur J Appl Physiol Occup Physiol 1989. (PMID 2737197) [PubMed] [DOI] Abstract
A population of healthy middle-aged (n = 69) and elderly men (n = 12), who participated in a health promotion program, was studied to determine whether really physically fit individuals are in good biological condition, and also whether improvement of physical fitness in the middle-aged and the elderly reduces their "rate of aging". Biological and physical fitness ages of the individuals studied were estimated from the data for 18 physiological function tests and 5 physical fitness tests, respectively, by a principal component model. The correlation coefficient between the estimated biological and physical fitness ages was 0.72 (p less than 0.01). Detailed analyses of the relationship between the estimated biological and physical fitness ages revealed that those who manifested a higher ("older") physical fitness age did not necessarily have a higher biological age, but those who manifested a lower ("younger") physical fitness age were also found to have a lower biological age. These results suggested that there were considerable individual variations in the relationship between biological condition and physical fitness among individuals with an old physical fitness age, but those who were in a state of high physical fitness maintained a relatively good biological condition. The data regarding the elderly men who had maintained a regular exercise program indicated that their estimated biological ages were considerably younger than the expected values. This might suggest that in older individuals regular physical activity may provide physiological improvements which in turn might reduce "the rate of aging".
↑ ^{Jump up to: 7.0} ^7.1 ^7.2 ^7.3 Klemera P & Doubal S: A new approach to the concept and computation of biological age. Mech Ageing Dev 2006. (PMID 16318865) [PubMed] [DOI] Abstract
The lack of exact definition of the concept of biological age (BA) is a typical feature of works concerning BA. That is why comparison of results of various published methods makes little sense and eventual proof of their optimality is impossible. Based on natural and simple presumptions, an attempt to express mathematically the supposed relation between chronological age (CA) and BA has proven to be unexpectedly fruitful. In the present paper, an optimum method of estimation of BA, which is easily applicable even in nonlinear cases, is derived. Moreover, the method allows evaluating the precision of the estimates and also offers tools for validation of presumptions of the method. A special feature of the method is that CA should be used as a standard biomarker, leading to essential improving the precision of BA-estimate and illuminating relativity of the known "paradox of biomarkers". All theoretical results of the method were fully approved by means of a special simulation program. Further, the theory and the results of the simulation have proven that many published results of BA-estimates using multiple linear regression (MLR) are very probably disserviceable because CA is typically more precise estimate of BA than estimates computed by MLR. This unpleasant conclusion also concerns methods, which use MLR as the final step after transformation of the battery of biomarkers by factor analysis or by principal component analysis.
↑ ^{Jump up to: 8.0} ^8.1 Hochschild R: Improving the precision of biological age determinations. Part 1: A new approach to calculating biological age. Exp Gerontol 1989. (PMID 2684676) [PubMed] [DOI] Abstract
In calculating biological age, almost all prior studies used multiple regression of chronological age on scores of biomarkers of aging. Multiple regression is invalid for this purpose for three, and in some circumstances four, reasons. These are: a) weighting of the contribution of each biomarker's scores according to strength of association with chronological age; b) regression of calculated ages to sample mean age and the inadequacy of proposed corrections; c) frequent occurrence of regression coefficients whose sign equates poorer adult performance on a test to younger biological ages; and d) multicollinearity when lung function scores and height are on the same side of the regression equation. An alternative method for calculating biological age is outlined. Regression to sample mean age and its solution are illustrated on data for highest audible pitch, one of 12 biomarkers measured in a study of 2462 office workers. Prior published studies employing multiple regression to calculate biological age appear to have been in error.
↑ ^{Jump up to: 9.0} ^9.1 ^9.2 ^9.3 Cho IH et al.: An empirical comparative study on biological age estimation algorithms with an application of Work Ability Index (WAI). Mech Ageing Dev 2010. (PMID 20005245) [PubMed] [DOI] Abstract
In this study, we described the characteristics of five different biological age (BA) estimation algorithms, including (i) multiple linear regression, (ii) principal component analysis, and somewhat unique methods developed by (iii) Hochschild, (iv) Klemera and Doubal, and (v) a variant of Klemera and Doubal's method. The objective of this study is to find the most appropriate method of BA estimation by examining the association between Work Ability Index (WAI) and the differences of each algorithm's estimates from chronological age (CA). The WAI was found to be a measure that reflects an individual's current health status rather than the deterioration caused by a serious dependency with the age. Experiments were conducted on 200 Korean male participants using a BA estimation system developed principally under the concept of non-invasive, simple to operate and human function-based. Using the empirical data, BA estimation as well as various analyses including correlation analysis and discriminant function analysis was performed. As a result, it had been confirmed by the empirical data that Klemera and Doubal's method with uncorrelated variables from principal component analysis produces relatively reliable and acceptable BA estimates.
↑ ^{Jump up to: 10.0} ^10.1 ^10.2 Levine ME: Modeling the rate of senescence: can estimated biological age predict mortality more accurately than chronological age?. J Gerontol A Biol Sci Med Sci 2013. (PMID 23213031) [PubMed] [DOI] [Full text] Abstract
Biological age (BA) is useful for examining differences in aging rates. Nevertheless, little consensus exists regarding optimal methods for calculating BA. The aim of this study is to compare the predictive ability of five BA algorithms. The sample included 9,389 persons, aged 30-75 years, from National Health and Nutrition Examination Survey III. During the 18-year follow-up, 1,843 deaths were counted. Each BA algorithm was compared with chronological age on the basis of predictive sensitivity and strength of association with mortality. Results found that the Klemera and Doubal method was the most reliable predictor of mortality and performed significantly better than chronological age. Furthermore, when included with chronological age in a model, Klemera and Doubal method had more robust predictive ability and caused chronological age to no longer be significantly associated with mortality. Given the potential of BA to highlight heterogeneity, the Klemera and Doubal method algorithm may be useful for studying a number of questions regarding the biology of aging.
↑ Jee H & Park J: Selection of an optimal set of biomarkers and comparative analyses of biological age estimation models in Korean females. Arch Gerontol Geriatr 2017. (PMID 28110151) [PubMed] [DOI] Abstract
To date, an optimal working model which predicts biological age (BA) with a set of working biomarkers has not been devised to represent the Korean female population. Accuracy of prediction and applicability are required of an optimal set of commonly assessed biomarkers to provide information on the health status. The goal of this study was to identify a set of biomarkers that represent the aging process and to develop and compare different BA prediction models to elucidate the most fitting and applicable model for providing information on health status in the Korean female population. Using a series of selection processes, eight clinically assessable variables were selected by analyzing relations between 31 clinical variables and chronologic age in 912 normal, healthy individuals among 3642 female participants with ages ranging from 30 to 80 years. The multiple linear regression (MLR), principal component analysis (PCA), and the Klemera-Doubal (KDM) statistical methods were applied to obtain three different sets of BA prediction models. These three models were assessed by calculating and performing the coefficient determinations (r2), regression slopes, effect sizes, pairwise t-tests, and Bland-Altman plots. The BA models were further compared for the applicability by calculating the BAs of clinical risk groups. MLR showed the narrowing effects at the either ends of the age spectrum with greatest effect sizes. PCA showed the greatest degree of dispersion and deviation from the regression center. These MLR and PCA trends were also exhibited by clinically risk groups. In conclusion, the KDM BA prediction model based on the selected biomarkers was found to provide the most reliable and stable results for the practical assessment of BA.

[pmid28546743-1] {Jump up to: 1.0} ^1.1 Jia L et al.: Common methods of biological age estimation. Clin Interv Aging 2017. (PMID 28546743) [PubMed] [DOI] [Full text] Abstract
At present, no single indicator could be used as a golden index to estimate aging process. The biological age (BA), which combines several important biomarkers with mathematical modeling, has been proposed for >50 years as an aging estimation method to replace chronological age (CA). The common methods used for BA estimation include the multiple linear regression (MLR), the principal component analysis (PCA), the Hochschild's method, and the Klemera and Doubal's method (KDM). The fundamental differences in these four methods are the roles of CA and the selection criteria of aging biomarkers. In MLR and PCA, CA is treated as the selection criterion and an independent index. The Hochschild's method and KDM share a similar concept, making CA an independent variable. Previous studies have either simply constructed the BA model by one or compared the four methods together. However, reviews have yet to illustrate and compare the four methods systematically. Since the BA model is a potential estimation of aging for clinical use, such as predicting onset and prognosis of diseases, improving the elderly's living qualities, and realizing successful aging, here we summarize previous BA studies, illustrate the basic statistical steps, and thoroughly discuss the comparisons among the four common BA estimation methods.

[pmid5841151-2] Hollingsworth JW et al.: Correlations between tests of aging in Hiroshima subjects--an attempt to define "physiologic age". Yale J Biol Med 1965. (PMID 5841151) [PubMed] [Full text]

[pmid6873212-3] Dubina TL et al.: Biological age and its estimation. II. Assessment of biological age of albino rats by multiple regression analysis. Exp Gerontol 1983. (PMID 6873212) [PubMed] [DOI] Abstract
Multiple regression model of biological age (BA) theoretically gives agreement with the main concept of BA. When assessment of BA is based on the model, the age being in regression center, the method provides satisfactory results, whereas BA estimates of individuals in extreme age groups are erroneous. Investigation of male and female Wistar rats of age 5-29 months showed the BA estimates calculated from 4-10 physiological indices in young (5-7 mo) animals are overestimated, and in old (24-28 mo) animals are underestimated. Coincidence of average BA in one-age group of animals with its chronological age served as a criterion for the correspondence of the estimate to "real" BA. The paper also examines the following questions: the necessary and sufficient number of physiological indices; the sample size from the intact animal population to establish normal aging standard; the relationship between BA and animal weight.

[pmid3226152-4] {Jump up to: 4.0} ^4.1 Nakamura E et al.: Assessment of biological age by principal component analysis. Mech Ageing Dev 1988. (PMID 3226152) [PubMed] [DOI] Abstract
A method of assessing biological age by the application of principal component analysis is reported. Healthy individuals (462) randomly selected from about 6000 men who had taken a 2-day health examination were studied. Out of the 30 physiological variables examined in routine check-ups, 11 variables were selected as suitable for the assessment of biological age based on the results of factor analysis and the physiological meaning of each test. This variable set was then submitted to principal component analysis, and the 1st principal component obtained from this analysis was used as an equation for assessing one's biological age. However, the biological age calculated from this equation is expressed as a score, so the estimated score was transformed to years (biological age) using the T-score idea. The biological age estimated by this method is practically useful and theoretically valid in contrast with the multiple regression model, because this approach eliminates and overcomes the following 2 big problems of the multiple regression model: (1) the distortion of the individual biological age at the regression edges; and (2) a theoretical contradiction in that a perfect model will merely be predicting the subject's chronological age, not his biological age.

[pmid950448-5] Webster IW & Logie AR: A relationship between functional age and health status in female subjects. J Gerontol 1976. (PMID 950448) [PubMed] [DOI] Abstract
A multiple regression equation was used to predict age from seven clinical variables in 1080 apparently well female subjects. A multiple correlation coefficient of R = 0.77 was achieved by five of the variables: timed forced expiratory volume, systolic blood pressure, plasma urea nitrogen, cholesterol, and alkaline phosphatase. On the basis of selection by medical questionnaire responses and other objective criteria, 9% of the subjects were nonsmokers and healthier than the rest. These selected subjects showed a significant reduction in preducted age. Within this group, subjective perception of health was associated with differences in predicted age: poor health with an increase and good health with a decrease in functional age. This study of functional age was based on the healthiest segment of the population in order to minimize the effect of overt pathological processes on the aging rate. An association has been demonstrated between health impairment and predicted age as a measure of the aging rate.

[pmid2737197-6] Nakamura E et al.: Biological age versus physical fitness age. Eur J Appl Physiol Occup Physiol 1989. (PMID 2737197) [PubMed] [DOI] Abstract
A population of healthy middle-aged (n = 69) and elderly men (n = 12), who participated in a health promotion program, was studied to determine whether really physically fit individuals are in good biological condition, and also whether improvement of physical fitness in the middle-aged and the elderly reduces their "rate of aging". Biological and physical fitness ages of the individuals studied were estimated from the data for 18 physiological function tests and 5 physical fitness tests, respectively, by a principal component model. The correlation coefficient between the estimated biological and physical fitness ages was 0.72 (p less than 0.01). Detailed analyses of the relationship between the estimated biological and physical fitness ages revealed that those who manifested a higher ("older") physical fitness age did not necessarily have a higher biological age, but those who manifested a lower ("younger") physical fitness age were also found to have a lower biological age. These results suggested that there were considerable individual variations in the relationship between biological condition and physical fitness among individuals with an old physical fitness age, but those who were in a state of high physical fitness maintained a relatively good biological condition. The data regarding the elderly men who had maintained a regular exercise program indicated that their estimated biological ages were considerably younger than the expected values. This might suggest that in older individuals regular physical activity may provide physiological improvements which in turn might reduce "the rate of aging".

[pmid16318865-7] {Jump up to: 7.0} ^7.1 ^7.2 ^7.3 Klemera P & Doubal S: A new approach to the concept and computation of biological age. Mech Ageing Dev 2006. (PMID 16318865) [PubMed] [DOI] Abstract
The lack of exact definition of the concept of biological age (BA) is a typical feature of works concerning BA. That is why comparison of results of various published methods makes little sense and eventual proof of their optimality is impossible. Based on natural and simple presumptions, an attempt to express mathematically the supposed relation between chronological age (CA) and BA has proven to be unexpectedly fruitful. In the present paper, an optimum method of estimation of BA, which is easily applicable even in nonlinear cases, is derived. Moreover, the method allows evaluating the precision of the estimates and also offers tools for validation of presumptions of the method. A special feature of the method is that CA should be used as a standard biomarker, leading to essential improving the precision of BA-estimate and illuminating relativity of the known "paradox of biomarkers". All theoretical results of the method were fully approved by means of a special simulation program. Further, the theory and the results of the simulation have proven that many published results of BA-estimates using multiple linear regression (MLR) are very probably disserviceable because CA is typically more precise estimate of BA than estimates computed by MLR. This unpleasant conclusion also concerns methods, which use MLR as the final step after transformation of the battery of biomarkers by factor analysis or by principal component analysis.

[pmid2684676-8] {Jump up to: 8.0} ^8.1 Hochschild R: Improving the precision of biological age determinations. Part 1: A new approach to calculating biological age. Exp Gerontol 1989. (PMID 2684676) [PubMed] [DOI] Abstract
In calculating biological age, almost all prior studies used multiple regression of chronological age on scores of biomarkers of aging. Multiple regression is invalid for this purpose for three, and in some circumstances four, reasons. These are: a) weighting of the contribution of each biomarker's scores according to strength of association with chronological age; b) regression of calculated ages to sample mean age and the inadequacy of proposed corrections; c) frequent occurrence of regression coefficients whose sign equates poorer adult performance on a test to younger biological ages; and d) multicollinearity when lung function scores and height are on the same side of the regression equation. An alternative method for calculating biological age is outlined. Regression to sample mean age and its solution are illustrated on data for highest audible pitch, one of 12 biomarkers measured in a study of 2462 office workers. Prior published studies employing multiple regression to calculate biological age appear to have been in error.

[pmid20005245-9] {Jump up to: 9.0} ^9.1 ^9.2 ^9.3 Cho IH et al.: An empirical comparative study on biological age estimation algorithms with an application of Work Ability Index (WAI). Mech Ageing Dev 2010. (PMID 20005245) [PubMed] [DOI] Abstract
In this study, we described the characteristics of five different biological age (BA) estimation algorithms, including (i) multiple linear regression, (ii) principal component analysis, and somewhat unique methods developed by (iii) Hochschild, (iv) Klemera and Doubal, and (v) a variant of Klemera and Doubal's method. The objective of this study is to find the most appropriate method of BA estimation by examining the association between Work Ability Index (WAI) and the differences of each algorithm's estimates from chronological age (CA). The WAI was found to be a measure that reflects an individual's current health status rather than the deterioration caused by a serious dependency with the age. Experiments were conducted on 200 Korean male participants using a BA estimation system developed principally under the concept of non-invasive, simple to operate and human function-based. Using the empirical data, BA estimation as well as various analyses including correlation analysis and discriminant function analysis was performed. As a result, it had been confirmed by the empirical data that Klemera and Doubal's method with uncorrelated variables from principal component analysis produces relatively reliable and acceptable BA estimates.

[pmid23213031-10] {Jump up to: 10.0} ^10.1 ^10.2 Levine ME: Modeling the rate of senescence: can estimated biological age predict mortality more accurately than chronological age?. J Gerontol A Biol Sci Med Sci 2013. (PMID 23213031) [PubMed] [DOI] [Full text] Abstract
Biological age (BA) is useful for examining differences in aging rates. Nevertheless, little consensus exists regarding optimal methods for calculating BA. The aim of this study is to compare the predictive ability of five BA algorithms. The sample included 9,389 persons, aged 30-75 years, from National Health and Nutrition Examination Survey III. During the 18-year follow-up, 1,843 deaths were counted. Each BA algorithm was compared with chronological age on the basis of predictive sensitivity and strength of association with mortality. Results found that the Klemera and Doubal method was the most reliable predictor of mortality and performed significantly better than chronological age. Furthermore, when included with chronological age in a model, Klemera and Doubal method had more robust predictive ability and caused chronological age to no longer be significantly associated with mortality. Given the potential of BA to highlight heterogeneity, the Klemera and Doubal method algorithm may be useful for studying a number of questions regarding the biology of aging.

[pmid28110151-11] Jee H & Park J: Selection of an optimal set of biomarkers and comparative analyses of biological age estimation models in Korean females. Arch Gerontol Geriatr 2017. (PMID 28110151) [PubMed] [DOI] Abstract
To date, an optimal working model which predicts biological age (BA) with a set of working biomarkers has not been devised to represent the Korean female population. Accuracy of prediction and applicability are required of an optimal set of commonly assessed biomarkers to provide information on the health status. The goal of this study was to identify a set of biomarkers that represent the aging process and to develop and compare different BA prediction models to elucidate the most fitting and applicable model for providing information on health status in the Korean female population. Using a series of selection processes, eight clinically assessable variables were selected by analyzing relations between 31 clinical variables and chronologic age in 912 normal, healthy individuals among 3642 female participants with ages ranging from 30 to 80 years. The multiple linear regression (MLR), principal component analysis (PCA), and the Klemera-Doubal (KDM) statistical methods were applied to obtain three different sets of BA prediction models. These three models were assessed by calculating and performing the coefficient determinations (r2), regression slopes, effect sizes, pairwise t-tests, and Bland-Altman plots. The BA models were further compared for the applicability by calculating the BAs of clinical risk groups. MLR showed the narrowing effects at the either ends of the age spectrum with greatest effect sizes. PCA showed the greatest degree of dispersion and deviation from the regression center. These MLR and PCA trends were also exhibited by clinically risk groups. In conclusion, the KDM BA prediction model based on the selected biomarkers was found to provide the most reliable and stable results for the practical assessment of BA.

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

Organ system	PCA	MLR	Hochschild’s	KDM
Cardiovascular system	Pulse pressure{{pmid\|19940465}}{{pmid\|24659482}}{{pmid\|29188884}}
	Systolic blood pressure{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|12672981}}{{pmid\|18840798}}{{pmid\|8803500}}{{pmid\|28110151}}	Systolic blood pressure{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|5841151}}{{pmid\|950448}}{{pmid\|28110151}}		Systolic blood pressure{{pmid\|28110151}}
	Heart rate{{pmid\|2737197}}{{pmid\|2282902}}
	Intima-media thickness{{pmid\|19940465}}{{pmid\|24522464}}{{pmid\|24659482}}
	Maximum internal diameter of carotid artery{{pmid\|24522464}}{{pmid\|24659482}}
	End diastolic velocity{{pmid\|19940465}}
	Mitral valve annulus ventricular septum of the peak velocity of early filling{{pmid\|24522464}}
	Mitral valve annulus lateral wall of peak velocity of early filling{{pmid\|19940465}}
	Mitral annulus peak E anterior wall{{pmid\|24659482}}
	Ratio of peak velocity of early filling to atrial filling{{pmid\|19940465}}

Respiratory system	VO2 max{{pmid\|22433233}}{{pmid\|18597867}}
	Forced expiratory volume in 1 second{{pmid\|22433233}}{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|12672981}}{{pmid\|18840798}}	Forced expiratory volume in 1 second{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|28110151}}	Forced expiratory volume in 1 second{{pmid\|2583248}}	Forced expiratory volume in 1 second{{pmid\|28110151}}
	Forced vital capacity{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|8803500}}	Forced vital capacity{{pmid\|20005245}}	Forced vital capacity{{pmid\|2583248}}
	Maximal mid expiratory flow rate 75/25{{pmid\|24522464}}	Vital capacity{{pmid\|5841151}}
Nervous system	Trail making test{{pmid\|24522464}}{{pmid\|29188884}}
	Digital symbol test{{pmid\|20005245}}{{pmid\|24659482}}	Digital symbol test{{pmid\|20005245}}
	Memory test linking names with faces{{pmid\|20005245}}	Memory test linking names with faces{{pmid\|20005245}}
	Memory test: which picture is at what place{{pmid\|20005245}}	Memory test: which picture is at what place{{pmid\|20005245}}
	Speed test: pointing icons from 1 to 15 sequentially, mixed in random positions{{pmid\|20005245}}
	Visual reaction time{{pmid\|20005245}}{{pmid\|22433233}}		Visual reaction time{{pmid\|2583248}}
			Sequence of lamps{{pmid\|2583248}}
			Alternate button tapping time with/without decision{{pmid\|2583248}}
			Movement time with/without decision{{pmid\|2583248}}
Renal system	Blood urea nitrogen{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|29380856}}{{pmid\|18840798}}{{pmid\|8803500}}	Blood urea nitrogen{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|28110151}}		Blood urea nitrogen{{pmid\|28110151}}
		Serum creatinine{{pmid\|28110151}}		Serum creatinine{{pmid\|28110151}}
	Cystatin C{{pmid\|19940465}}{{pmid\|24522464}}{{pmid\|24659482}}{{pmid\|29188884}}
Liver	Serum albumin{{pmid\|17921421}}{{pmid\|18597867}}{{pmid\|29380856}}{{pmid\|18840798}}	Serum albumin{{pmid\|950448}}
	Glutamic oxaloacetic transaminase{{pmid\|2737197}}{{pmid\|2282902}}	Glutamic oxaloacetic transaminase{{pmid\|28110151}}		Glutamic oxaloacetic transaminase{{pmid\|28110151}}
	Glutamic pyruvic transaminase{{pmid\|8803500}}
	Ratio of albumin to globulin{{pmid\|12672981}}
	Lactate dehydrogenase{{pmid\|2737197}}{{pmid\|2282902}}{{pmid\|8803500}}
		Serum globulin{{pmid\|950448}}
		Alkaline phosphatase{{pmid\|950448}}
Hematologic system	Erythrocyte sedimentation rate{{pmid\|18597867}}	Erythrocyte sedimentation rate{{pmid\|950448}}
	Mean corpuscular hemoglobin{{pmid\|12672981}}
	Red blood cell count{{pmid\|18840798}}
	Hematocrit{{pmid\|17921421}}
	Hemoglobin concentration{{pmid\|2737197}}{{pmid\|2282902}}
	Fibrinogen{{pmid\|19940465}}
		Ferratin{{pmid\|28110151}}		Ferratin{{pmid\|28110151}}
Metabolism	Glycosylated hemoglobin{{pmid\|18597867}}
	Glucose{{pmid\|12672981}}{{pmid\|8803500}}	Glucose{{pmid\|950448}}
	Low-density cholesterol{{pmid\|18597867}}
	Atherogenic index{{pmid\|2737197}}{{pmid\|2282902}}
	Triglyceride{{pmid\|2282902}}	Triglycerides{{pmid\|6967883}}
	Total cholesterol{{pmid\|8803500}}	Total cholesterol{{pmid\|23213031}}{{pmid\|6967883}}{{pmid\|5841151}}{{pmid\|28110151}}		Total cholesterol{{pmid\|28110151}}
Muscle and fat	Grip strength{{pmid\|20005245}}{{pmid\|22433233}}	Grip strength{{pmid\|20005245}}{{pmid\|5841151}}
	Soft lean mass{{pmid\|22433233}}
	Waist circumference{{pmid\|22433233}}{{pmid\|18597867}}	Waist circumference{{pmid\|28110151}}		Waist circumference{{pmid\|28110151}}
	Percent body fat{{pmid\|18597867}}
Sensory system	Hearing threshold{{pmid\|18597867}}
	Highest audible pitch{{pmid\|20005245}}	Highest audible pitch{{pmid\|20005245}}	Highest audible pitch{{pmid\|2583248}}
		Light extinction test{{pmid\|5841151}}
		Visual acuity{{pmid\|5841151}}
		Auditory function{{pmid\|5841151}}
		Vibrotactile sensitivity{{pmid\|5841151}}	Vibrotactile sensitivity{{pmid\|2583248}}
		Auditory reaction time{{pmid\|20005245}}	Auditory reaction time{{pmid\|2583248}}
		Focal range test using a Landolt ring{{pmid\|20005245}}	Visual accommodation{{pmid\|2583248}}
Genetic index	Telomere restriction fragment{{pmid\|24659482}}