

ORIGINAL ARTICLE 

Year : 2022  Volume
: 11
 Issue : 3  Page : 143147 

Prediction of height using hand and foot parameters in Maharashtrian population
Supriya Sarjerao Methepatil^{1}, Vrushali Ashok Dethe^{2}
^{1} Associate Professor, Department of Anatomy, Symbiosis Medical College for Women, Symbiosis International (Deemed University), Pune, Maharashtra, India ^{2} Assistant Professor, Department of Anatomy, R. C. S. M. Govt. Medical College, Kolhapur, Maharashtra, India
Date of Submission  24Feb2022 
Date of Decision  02May2022 
Date of Acceptance  30May2022 
Date of Web Publication  14Jul2022 
Correspondence Address: Supriya Sarjerao Methepatil Flat No 103, C Wing, Windows CoOperative Housing Society Phase 13, Opposite Sus Hospital, Sus, Pune  411 021, Maharashtra India
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/NJCA.NJCA_52_22
Background: Height, a reflection of skeletal growth, is important in calculating various parameters such as lung volumes, glomerular filtration, and metabolic rate. In forensic medicine, where skeletal remains are recovered, calculating the stature is necessary. Estimation of height is important in design of prosthetic body parts and reconstruction surgeries. The objective of this study was to estimate the height of a person from hand and foot parameters. Methodology: In a crosssectional study, height, hand length (HL), hand breadth (HB), foot length (FL), and foot breadth (FB) were measured in a sample of 200 students (116 females and 84 males) aged between 18 and 24 years. Multiple regression model was used to derive equations for calculation of height from these parameters. Results: Three different multiple regression equations were derived for total cases and for male population and female population separately. The equation for estimation of height in males is 78.808 + 3.592 FL − 1.094 FB + 0.508 HL + 0.003 HB, in females is 69.042 + 3.889 FL + 0.729 FB + 0.621 HL − 1.077 HB and in general is 53.649 + 3.884 FL − 0.285 FB + 0.509 HL + 1.084 HB. The multiple correlation coefficient^{®} is 0.844 for total population and 0.673 for male population and 0.733 for female population (P = 0.001 indicating high significance). Conclusion: Multiple regression equations using FL, FB, HL, and HB can predict height.
Keywords: Foot breadth, foot length, hand breadth, hand length, hand stature, stature
How to cite this article: Methepatil SS, Dethe VA. Prediction of height using hand and foot parameters in Maharashtrian population. Natl J Clin Anat 2022;11:1437 
How to cite this URL: Methepatil SS, Dethe VA. Prediction of height using hand and foot parameters in Maharashtrian population. Natl J Clin Anat [serial online] 2022 [cited 2022 Oct 6];11:1437. Available from: http://www.njca.info/text.asp?2022/11/3/143/353720 
Introduction   
Anthropometry gives us the scientific basis for estimating various measurements in living and dead people.^{[1]} Growth of a person iss well assessed by measuring one's height.^{[2]} Height measurement is important for calculating various parameters such as lung volumes, glomerular filtration, and metabolic rate. Lowerlimb deformities, spine problems (scoliosis), and united, malunited lowerlimb fractures may influence the measured height of the person.^{[3]} Height estimation of deceased persons when only body parts are available is a challenging yet an important step in identification of the person.^{[4]}
In the mathematical method, measurement of one of the body parts estimates the height. The mathematical method can be used in forensic science for personal identification, especially where separate body parts are found.^{[5]} If remains of skeleton are found then as a forensic medicine expert, height of that individual is predicted by using the regression equations.^{[1]} Height can be calculated from long bones, hand or foot, or any other segments of the body.^{[6]}
Using multiple regression model, an unknown parameter can be calculated from multiple known parameters having correlation with each other.^{[7]} The present study is formulated to calculate height from hand length (HL), hand breadth (HB), foot length (FL), and foot breadth (FB). This study is important for identification of an individual, especially during accidents and mass calamities like earthquakes. The findings of this study can also help in reconstruction surgeries and prosthesis development. The objective of the study was to formulate the multiple regression equations to calculate the height using HL, HB, FL, and FB.
Materials and Methods   
Study setting and design
In a crosssectional study, height, HL, HB, FL, and FB were measured in a sample of 200 students (116 females and 84 males) aged between 18 and 24 years. The study was conducted at a government medical college, Miraj, Maharashtra, after getting ethics committee approval for the study (letter number GMCM/H/4958/62/12 dated 08.04.2013).
Eligibility criteria
All students willing to take part in the study aged between 18 and 24 years were included. Those with any significant disease, orthopedic limb deformity, and any other disorders which could directly affect the height measurement of an individual are excluded.
After informing about the study and obtaining written consent, height, HL, HB, FL, and FB were measured. Stadiometer, measuring scale, paper, and pencil were used to obtain the measurements. All the parameters were measured under the same conditions using the same measuring instruments in a wellilluminated room. Before taking measurements, it was checked that nails were trimmed. Using standard instruments, both side measurements were taken as follows.
The height was measured using the anthropometer as the distance from the vertex to the floor.^{[8]} Height was measured only after removing the caps or any head accessories or any footwear.
Foot measurements
The participant was standing with equal weightbearing on both the feet, slightly apart from each other. The outline was marked as it is done for shoeprint.^{[9]} FL was distance from the most distal point either of the first toe or of the second toe whichever is the longest (A) to the most prominent point of the heel (B) [Figure 1]. FB was measured from the prominence of the head of the first metacarpal (C) to the prominence of the head of the fifth metacarpal (D) [Figure 1].^{[10],[11]}  Figure 1: Tracing of the foot on the paper. (A) The most distal point either of the first toe or of the second toe whichever is the longest, (B) the most prominent point of the heel, (C) the prominence of the head of the first metacarpal, (D) the prominence of the head of the fifth metacarpal. AB is the foot length and CD is the foot breadth
Click here to view 
Hand measurements
The marking of the outline of hand started from radial styloid process encircling the fingers and then to the ulnar styloid process with thumb adducted and flexed touching the radial border of the palm.^{[10]} It was made sure that the hand was in direct line with the forearm. Interstyloid line connecting the radial and ulnar styloid processes was drawn. HL was measured from the distalmost point of the middle finger (P) to the midpoint of the interstyloid line (Q) [Figure 2]. HB was measured from the prominence of the head of the second metacarpal (R) to the prominence of the head of the fifth metacarpal (S). In cases where the thumb cannot be adducted and flexed, the measurements will be recorded, as shown in [Figure 3], where R coincides with point from where the thumb diverges from the radial border of the palm.^{[12],[13]}  Figure 2: Tracing of the hand with thumb adducted and flexed touching the radial border of the palm. P – the distalmost point of the middle finger, Q – the midpoint of the interstyloid line, R – the prominence of the head of the second metacarpal (this point coincides with point from where the thumb diverges from the radial border of the palm), S – the prominence of the head of the fifth metacarpal
Click here to view 
 Figure 3: Tracing of the hand with thumb in normal anatomical position. P – the distalmost point of the middle finger, Q – the midpoint of the interstyloid line, R – the prominence of the head of the second metacarpal [this point coincides with point from where the thumb diverges from the radial border of the palm, as shown in Figure 2], S – the prominence of the head of the fifth metacarpal
Click here to view 
Sample size
The sample size for the present crosssectional study was decided considering 99.99% confidence interval, population size (N): 1000, hypothesized % frequency of outcome factor in the population (p): 20%±10, confidence limits as % of 100(d): 10%, and design effect (DEFF): 1%. The following equation was used – sample size n = [DEFF × Np(1p)]/[(d^{2}/Z^{2}1α/2 × (N1) +p × (1p)] which came out to be 196.
Data were analyzed using appropriate statistical tests by the use of SPSS version 20 (IBM, United StatesEnglish). We used multiple regression analysis model to estimate height from these parameters. Multiple regression equation is in the following format: Y = β0 + β1 (FL) + β2 (FB) + β3 (HL) + β4 (HB), where y = dependent variable (height in the present study), β0 = the yintercept = the value of y when all other parameters = “0,” and β1, β2, β3, and β4 = regression coefficient for FL, FB, HL, and HB, respectively (FL = foot length, FB = foot breadth, HL = hand length, and HB = hand breadth).^{[14]}
Results   
After measuring the height of an individual, the HL, HB, FL, and FB of the person were also recorded on both the sides. These parameters were analyzed for their distribution in the population. The difference in the measurements of the two sides was insignificant, so we considered the average of the right and left sides of each parameter for further study. The average parameters of the study population are height = 163.1 cm, FL = 24.5 cm, FB = 9.5 cm, HL = 17.7 cm, and HB = 7.5 cm. The average parameters for the male population are height = 169.8 cm, FL = 25.8 cm, FB = 10.1 cm, HL = 18.6 cm, and HB = 8.1 cm. The average parameters for the female population are height = 158.2 cm, FL = 23.5 cm, FB = 9.1 cm, HL = 17.1 cm, and HB = 7.2 cm.
We used multiple regression analysis model to estimate height from these parameters.
The equations derived were as follows:
 Estimated height total cases = 53.649 + 3.884 FL − 0.285 FB + 0.509 HL + 1.084 HB
 Estimated height in males = 78.808 + 3.592 FL − 1.094 FB + 0.508 HL + 0.003 HB
 Estimated height in females = 69.042 + 3.889 FL + 0.729 FB + 0.621 HL − 1.077 HB.
The dependent variable is height and the explanatory variables are HL, HB, FL, and FB. From [Table 1], [Table 2], [Table 3]e, we observe that the multiple correlation coefficient® is highly significant 0.844 for all cases considering both sexes together and 0.673 for both males and 0.733 for female cases considered separately (P = 0.001).  Table 1: Multiple regression equation derived for calculating height from various parameters considering total population
Click here to view 
 Table 2: Multiple regression equation derived for calculating height from various parameters considering male population
Click here to view 
 Table 3: Multiple regression equation derived for calculating height from various parameters considering female population
Click here to view 
Square of the correlation coefficient (R^{2}) determines the strength of association among the parameters considered.^{[6]} For total cases, R^{2} is 0.713 which indicates that 71.3% of the prediction of height is attributed to the parameters considered. For male cases, the value of R^{2} is 0.453 which means 45.3% height prediction is attributed to the parameters. In females, R^{2} is 0.538 indicating that 53.8% prediction of height is attributed to the parameters. The positive values of correlation coefficients indicate that height has a strong positive correlation with these parameters, and if one increases, the other value is bound to be a larger value. If the parameters under consideration are known, height can be calculated using the above formulas with good accuracy for the given population. For example, if measurements obtained in a case are FL = 24.2 cm, FB = 9.5 cm, HL = 17.7 cm, and HB = 8.8 cm, then using the equation, estimated height total cases = 53.649 + 3.884 FL − 0.285 FB + 0.509 HL + 1.084 HB,
Estimated height = 53.649 + (3.884 × 24.2) − (0.285 × 9.5) + (0.509 × 17.7) + (1.084 × 8.8) = 163.5 cm.
The actual height of that individual was 167 cm. This indicates that the given equations predict the height of the individual in a better way. If sex of the individual is known, we can apply the genderspecific equations for a better outcome.
Discussion   
Several factors are known to influence the skeletal development. Anthropometry is the technique of various body measurements.^{[1]} The age and sex of an individual influence these measurements.^{[15]} In cases of disasters such as earthquakes, floods, accidents, wars, and bomb blasts, establishing the identity of the individual is very important from the remains.^{[4]} Calculating height is an important aspect of an individual.^{[6]} Anthropometry of foot is also very important considering the designing of a comfortable footwear.^{[16]} Various racial and geographical factors do influence the stature and the skeletal development.^{[17]}
Various researchers have studied various body parts for calculating the stature so far. Long bones of limbs are most widely studied until now for estimation of stature.^{[18]} Singh and Sohal^{[19]} studied the length of the clavicle and found that its correlation with height is significant. Athawale^{[20]} helped us calculate height from the forearm bones. Patel et al.^{[21]} have given an equation for calculating height from the tibia. Saxena et al.^{[22]} derived a linear equation to find height from the head length. Wankhede et al.^{[23]} considered total facial height and nasal height to calculate height. Many authors have found a positive correlation of either of these parameters with height and have derived linear equations to derive height from a single parameter.^{[1],[11],[12],[18],[24],[25],[26],[27],[28],[29]} Chikhalkar et al.^{[17]} derived height from HB. Kavyashree et al.^{[29]} studied FL to calculate height by finding an equation. Munglang et al.^{[28]} estimated height from HL and FL.
In the present study, we calculated stature from HL, HB, FL, and FB. The multiple correlation coefficient® is 0.844 for all cases considering total population and 0.673 for male population and 0.733 for female population separately. These values are larger than the Pearson's correlation coefficients derived considering a single parameter.^{[12],[15],[17],[25]} This fact indicates that multiple regression analysis estimates height more accurately as compared to the linear regression equations. These equations are helpful to calculate height in cases of forensic medicine, or in cases of deformities where height cannot be directly measured. One must note that these equations are specific to the population they have been derived from as many factors such as genetic or racial or environmental factors may influence the parameters considered.
Ozaslan et al.^{[25]} studied hand and foot dimensions with stature. They found that the FL estimated height better than the other parameters under consideration. HB showed a weak correlation with height. Multiple regression equations predicted height better than linear equations. The equations they derived are 746.16 + 2.31 HL − 2.13 HB + 2.85 FL − 0.08 FB for males and 509.44 + 1.52 HL + 2.59 HB + 2.98 FL − 0.04 FB for females.
Khanapurkar and Radke^{[12]} estimated stature using FL, HL, and head length. They found that contribution of head length in estimating height was not significant when it was considered with HL and FL. The equation derived by them was height = 59.451 + 2.552 FL + 2.295 HL.
The limitations of the present study are that these equations are specific to that population as there are various genetic and racial as well as environmental factors affecting the skeletal development of a person. More accurate results will be obtained if it is applied to the same population. These equations will provide more accurate results if applied for the same age group. The equations from this study are applicable only when both the hand and foot measurements are available. Their application is limited if only hand or only foot measurements are available.
Conclusion   
Multiple regression equations using FL, FB, HL, and HB can predict height. The equation for estimation of height in males is 78.808 + 3.592 FL − 1.094 FB + 0.508 HL + 0.003 HB, in females is 69.042 + 3.889 FL + 0.729 FB + 0.621 HL − 1.077 HB, and in general is 53.649 + 3.884 FL − 0.285 FB + 0.509 HL + 1.084 HB.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  Patel SM, Doshi V, Ruparelia S, Zalawadia A, Parekh D, Rathod SP, et al. Anthropological study of the foot and it's relationship between different parameters and stature in an adult population of different areas of Gujarat. NJIRM 2011;2:6770. 
2.  Patel SM, Shah GV, Patel SV. Estimation of height from measurements of foot length in Gujarat region. J Anat Soc India 2007;56:257. 
3.  Zverev YP. Relationship between arm span and stature in Malawian adults. Ann Hum Biol 2003;30:73943. 
4.  Ismaila SO. Anthropometric data of hand, foot and ear of university students in Nigeria. Leonardo J Sci 2009;15:1520. 
5.  Fully G. Une nouvelle méthode de détermination de la taille. Ann Med Legale 1956;35:26673. 
6.  Ozaslan A, Işcan MY, Ozaslan I, Tuğcu H, Koç S. Estimation of stature from body parts. Forensic Sci Int 2003;132:405. 
7.  Mahajan BK. Methods in Biostatistics for Medical Students and Research Workers. 7 ^{th} ed. New Delhi: Jaypee Brothers Medical Publishers Ltd.; 2010. 
8.  Jasuja OP, Singh G. Estimation of stature from hand and phalange length. J Indian Acad Forensic Med 2004;26:1006. 
9.  Pawar PK, Dadhlich A. Study of correlation between human height and foot length in residents of Mumbai. Int J Biol Med Res 2012;3:22325. 
10.  Oommen A, Mainker A, Oommen T. A study of the correlation between hand length and foot length in humans. J Anat Soc India 2005;54:19. 
11.  Rani M, Tyagi AK, Ranga VK, Rani Y, Murari A. Stature estimates from foot dimensions. J Punjab Acad Forensic Med Toxicol 2011;11:2630. 
12.  Khanapurkar S, Radke A. Estimation of stature from the measurement of foot length, hand length and head length in Maharashtra region. Indian J Basic Appl Med Res 2012;1:7785. 
13.  Amirsheybani HR, Crecelius GM, Timothy NH, Pfeiffer M, Saggers GC, Manders EK, et al. The natural history of growth of hand. Part I: Hand area as a percentage of body surface area. Plast Reconstr Surg 2001;107:72633. 
14.  
15.  Krishan K. Anthropometry in forensic medicine and forensic science – 'Forensic anthropometry'. Internet J Forensic Sci 2007;2:117. 
16.  Fessler DM, Haley KJ, Lal RD. Sexual dimorphism in foot length proportionate to stature. Ann Hum Biol 2005;32:4459. 
17.  Chikhalkar BG, Mangaonkar AA, Nanandkar SD, Peddawad RG. Estimation of stature from measurements of long bones, hand and foot dimensions. JIAFM 2009;32:32931. 
18.  Trotter M, Gleser GC. A reevaluation of estimation of stature based on measurements of stature taken during life and of long bones after death. Am J Phys Anthropol 1958;16:79123. 
19.  Singh B, Sohal HS. Estimation of stature from clavicle in Punjabis; a preliminary report. Indian J Med Res 1952;40:6771. 
20.  Athawale MC. Estimation of height from length of forearm bones – A study of one hundred Maharashtrian male adults of ages between 2530 years. Am J Phys Anthropol 1963;21:10512. 
21.  Patel MP, Joshi NB, Dongre AV. Regression equation of height on tibial length. Indian J Med Res 1964;52:5314. 
22.  Saxena SK, Jeyasingh P, Gupta AK, Gupta CD. The estimation of stature from head length. J Anat Soc India 1981;30:789. 
23.  Wankhede KP, Kamdi NY, Parchand MP, Anjankar VP, Bardale RV. Estimation of stature from maxillafacial anthropometry in a central Indian population. J Forensic Dent Sci 2012;4:347. [ PUBMED] [Full text] 
24.  Methepatil SS, Herekar NG. Study of correlation of hand parameters with stature in humans. Indian J Anat 2019;8:1915. 
25.  Ozaslan A, Karadayi B, Kolusayin MO, Kaya A, Afsin H. Predictive role of hand and foot dimensions in stature estimation. Rom J Leg Med 2012;20:416. 
26.  Uhrová P, Beňuš R, Masnicová S. Stature estimation from various foot dimensions among Slovak population. J Forensic Sci 2013;58:44851. 
27.  Lalwani R, Gupta H, Athavale S, Kotgirwar S. Estimation of stature and sex by foot length measurements using linear regression and discriminant function analysis, respectively: A study in central India population. Natl J Clin Anat 2019;8:14. [Full text] 
28.  Munglang M, Bora D, Roy R, Medhi A. Comparison of height, hand length and foot length between the tribal and the non tribal population of north east India. Natl J Clin Anat 2020;9:146. [Full text] 
29.  Kavyashree AN, Bindurani MK, Asha KR, Lakshmiprabhasubhash. Study on correlation of foot length with stature. Natl J Clin Anat 2016;5:1915. [Full text] 
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]
