There is a paucity of data regarding stature estimation using the upper limb bone length in North Indian population. The present study was conducted around the National capital region to develop a regression analysis-based formula to calculate the stature from the upper limb measurements.
The study enrolled 1000 healthy participants after written informed consent (500 male and 500 female) of age group between 18-30 years between January 2025 to December 2025. The participants were undergraduate students of SGT University, Gurugram (Haryana). Length of the long bones and height of the participants were recorded percutaneously using a stadiometer. A linear regression model was used to estimate the extent of reliable prediction of the stature from the measurements of the bone length for all three bones.
The mean height of male was 173.5cm and the mean height for female was 163.4cm. In both males and females, the lengths for humerus, radius and ulna were positively correlated with the stature on both sides and they were statistically significant. The standardized beta for humerus, radius and ulna were 0.343, 0.089 and 0.401 respectively. Using the regression model, stature equation was derived as 38.564 + 1.386 x (Length of Humerus) + 0.559 x (Length of Radius) + 2.710 x (Length of Ulna).
The value of standardized beta shows that the length of ulna serves as the most reliable and consistent parameter for estimation of stature among the North Indian population.
Forensic medicine delves upon stature estimation based on the skeletal remains, as an important tool for deriving the identification of an individual. Such part or complete bones additionally serve to identify the gender of the victim or casualty with the applications going beyond a crime scene to include mass casualty and natural events where identification becomes essential for personal reasons [1]. Researchers have tried to use a variety of methods for the purpose of estimating the stature from the long bones but there have been challenges in a one size fit all strategy since the estimates vary from region to region and between races [2].
The stature estimation is mainly based on two standard techniques: (1) the anatomical method where a complete skeleton is analyzed to ascertain the stature, or (2) the mathematical model based on the regression analysis where a single long bone (complete) is sufficient to predict the stature. This is especially useful when a complete skeleton is not available [3]. Another morphological criteria for sorting the bones for the purpose of associating the bone to an individual involves analysis of the bones' compatibility on the basis of their colour, texture, size and robustness [4].
The search of the available literature showed that a few studies have been conducted from regions around North India such as Faridabad [5], Himachal Pradesh [6,7], Allahabad [8] and Amritsar [9]. There is a paucity of data regarding stature estimation using the long bones of the upper limb among North Indian population from the Delhi capital region. Moreover, there is a substantial movement of the young workforce around the Indian capital region and periodic population-based studies could serve as a more effective tool to estimate the stature owing to possible changing population composition with time. In view of the above, the present study was conducted at a tertiary care center from North India around the National capital region to develop a new regression analysis-based formula to calculate the stature from the upper limb measurements.
The present study was a prospective, cross-sectional study involving 1000 participants (500 male and 500 female) of age group between 18-30 years at a tertiary care center from North India between January 2025 to December 2025. The participants were students of the campus belonging to Haryana (n=627), Delhi (n=214), Uttar Pradesh (n=128), Punjab (n=13) and Rajasthan (n=18). Only apparently healthy individuals were included in the study after informed consent and participants with documented musculoskeletal, congenital, nutritional or acquired deformity or amputated upper limb were excluded from the study. The study was approved by the institutional ethics committee vide letter No. IEC/FMHS/PhD/2024-5.
Anthropometric measurements were taken from the above-mentioned normal healthy volunteers by a single observer from the department of Anatomy from SGT University, Gurugram (Haryana). The study participants were asked to stand straight with their bare feet in approximation and maintaining the head in a Frankfort horizontal plane. A stadiometer (Model name: Prestige, item code: HM 006A, Hardik Medi-Tech, Sonipat, Haryana) was employed for the measurement of the height of the study participant from crown to heel. The lengths of the long bones, viz. humerus, radius and ulna were measured by the standard measuring tape (Generic fiberglass measuring tape gauged for accuracy with the stadiometer reading to ensure precision among the two models) from the specific anatomical landmarks. The measurement of humerus was taken with extended elbow while the measurement of radius and ulna was taken with flexed elbow. The length of the three bones was measured using the following landmarks: (1) Length of Humerus: It was measured percutaneously between greater tubercle of humerus and the lateral epicondyle of humerus (Fig 1). (2) Length of Radius: Head of radius to maximum prominent point of styloid process of the radius (Fig 2). (3) Length of Ulna: Olecranon process to maximum prominent point of styloid process of ulna (Fig 3).
The data collected in the present study was analyzed by using SPSS software ver. 27 (IBM Corp., USA). The baseline data was entered in Microsoft Office Excel for Microsoft 365 suite (Microsoft Corp., USA) and the mean, standard deviation and range of data were recorded for all variables. Student t-test was employed for comparing the mean height, mean length of the humerus, radius and ulna between male and female participants whereas the Pearson correlation coefficient was used for determining the relationship between the height and bone length in the participants. A linear regression model was finally used to estimate the extent of reliable prediction of the stature from the measurements of the bone length for all three bones. p value of <0.05 was considered statistically significant.
A total of 1000 healthy personnel were included in the present study, comprising of 500 males and 500 females. The age of the participants was between 18-30 years. The mean height of male was 173.5cm and the mean height for female was 163.4cm. The present study estimated the relationship between the stature (height) of the participants and the different bones data. It was observed that in female participants, the bone lengths for humerus, radius and ulna were positively correlated with the stature on both sides and they were statistically significant implicating that they have a positive impact in determining the stature of the female. In male participants, there was very high positive correlation of stature and lengths of humerus, radius and ulna for both the sides and also statistically significant leading to the interpretation that these bones have a very high positive role in the stature of the male. The correlation for stature and bony lengths for both the genders are discussed in table 1. Statistically significant (p<0.001) differences were found in the height and the upper limb bone length between males and females.
| Variable | Females (r) | Males (r) | P value |
|---|---|---|---|
| Humerus Right Hand (HumR) | 0.4195* | 0.7114* | <0.01 |
| Radius Right Hand (RadR) | 0.2517* | 0.7904* | <0.01 |
| Ulna Right Hand (UlnR) | 0.3043* | 0.8261* | <0.01 |
| Humerus Left Hand (HumL) | 0.3507* | 0.7972* | <0.01 |
| Radius Left Hand (RadL) | 0.2526* | 0.7567* | <0.01 |
| Ulna Left Hand (UlnL) | 0.2812* | 0.7286* | <0.01 |
Table 2 shows the values of R-square which shows the variation of the dependent variable explained by the independent variables. The value of R-square is 61.6% which was excellent and shows that 61.6% of the variation in stature is explained by the three bones together a substantial proportion of the variance, but about 38.4% of the variation was still unexplained by these bones, likely due to other factors (genetics, other body parts, etc.).
| Model | R | R2 | Adjusted R2 | Std. error of the estimate |
|---|---|---|---|---|
| 1 | 0.785 | 0.616 | 0.615 | 6.53843 |
Table 3 below shows multiple linear regression based on the standardized coefficients: Stature = 38.564 + 1.386 x (Length of Humerus) + 0.559 x (Length of Radius) + 2.710 x (Length of Ulna). This was the equation for overall linear regression. The coefficients change in stature (in cm) for a one-unit increase in the predictor, holding other variables constant: Humerus (B = 1.386): For every 1 cm increase in humerus length, stature increases by 1.386cm, holding radius and ulna constant; Radius (B = 0.559): For every 1 cm increase in radius length, stature increases by 0.559cm, holding humerus and ulna constant; Ulna (B = 2.710): For every 1 cm increase in ulna length, stature increases by 2.710cm, holding humerus and radius constant.
| Model | Unstandardized Coefficients (B) | Std. error | Standardized coefficients (Beta) | t | Sig. |
|---|---|---|---|---|---|
| (Constant) | 38.564 | 3.664 | - | 10.526 | <0.001 |
| Length of Humerus Right forearm | 1.386 | 0.139 | 0.343 | 9.977 | <0.001 |
| Length of Radius Right forearm | 0.559 | 0.303 | 0.089 | 1.847 | 0.065 |
| Length of Ulna Right forearm | 2.710 | 0.310 | 0.401 | 8.744 | <0.001 |
Personal identification on the basis of stature, age, gender and ethnicity is employed by forensic experts as cost effective tools [10] that aids DNA based identification which is a more cost intensive exercise. Identification is especially important in events of mass casualties. Among different methods of identification, estimation of an individual's stature on the basis of measurement of specific parts of the body has been employed by forensic anthropologists [11]. Studies have been conducted on stature estimation based on arm or long bone measurements of upper limbs in individuals of varying ethnicity, regions or age groups [12].
The present study observed a statistically significant difference between the length of long bones of upper limb and stature among male and female participants. Height of males was significantly taller than females, as observed in previous studies [13]. The mean length of the three upper limb bones, namely humerus, radius and ulna, were significantly larger in males as compared to females, which is in accordance with the study by Patel [14]. Another study conducted by Krishan [15] in the age group of 17-20 years employed linear and multiple regression models to estimate body stature on the basis of measurements of hands and feet among the Rajput community in North India. Kim et al [16] also reported similar findings among Sudanese adults. Supare et al [17] enrolled 400 participants in Maharashtra to examine the relationship between stature and arm span (R = 0.89 in males, 0.90 in females), concluding arm span is an accurate predictor. Hariharan et al [18] measured right ulna percutaneously, observing a positive correlation.
| Author | Age (yrs) | Region | Parameter Used | R2 coefficient of determination | Findings |
|---|---|---|---|---|---|
| Mahakkanukrauh et al (2011) [19] | 19-94 (M), 26-93 (F) | Thailand | Humerus, radius and Ulna for both gender | Hum=0.667, Rad=0.672, Ulna=0.660 (M); Hum=0.552, Rad=0.616, Ulna=0.567 (F) | Male humeral, radial and ulnar length are larger than females. Ulna bone is the most reliable predictor for stature estimation. |
| Mahakizadeh et al (2016) [20] | 18-30 | Iran | Upper limb length upto tip of middle finger | Male = 0.89 | Strong correlation between hand length and height |
| Boonthai et al (2024) [21] | 34-90 | Thailand | Radius and ulna | Radius = 0.509, Ulna = 0.516 | Ulna is more suitable predictor for stature than radius |
| Mehta et al (2015) [22] | 18-30 | India | Ulna in both gender | Rt ulna = 0.569, Lt ulna = 0.49 | Ulna is the most suitable predictor for stature |
| Pandeya et al (2018) [23] | 18-25 | Nepal | Hand length and stature | 0.25 | Hand was more suitable predictor for stature |
| Pandey et al (2017) [24] | 18-25 | Southern India | Ulna in both gender | Male = 0.45, Female = 0.27 | Ulna was the suitable predictor of stature |
| Present study 2026 | 18-30 | North India | Humerus, radius, ulna | 0.616 | Ulna was the suitable predictor of stature followed by humerus and radius |
The present study has a limitation that the sample population was heterogeneous owing to the study population comprising a large group of young workforce that often migrates away from their permanent settlement. The study sample was enrolled among the population based on the present residence status to ascertain the relationship of stature and bone length in this sample that closely resembles the heterogeneous population prevalent in this region. But, no in-depth analysis of the ethnicity/region of the study participants could be incorporated due to constraints.
Among the three long bones of the upper limb, length of ulna serves as the most reliable and consistent parameter for estimation of stature among the North Indian population. The regression equation derived in the present study is applicable to the population around the National Capital region and should serve as a tool for stature estimation for different applications, such as anthropology, anatomical studies, clinical and forensic analysis. The clinical utility of the regression model is its application in estimating stature in adults when direct height measurement is not feasible, such as in bedridden, immobilized, or physically disabled individuals.
The authors offer their sincere gratitude to the participants for their involvement in the research and to Dr. Pardeep Singh, Department of Forensic Medicine, Pt Jawahar Lal Nehru Govt Medical College, Chamba, Himachal Pradesh for their valuable inputs and expert insight.