Kewal Krishan, PhD, Sr. Assistant Professor, Department of Anthropology, Panjab University, Sector-14, Chandigarh-160 014, India. Tel: 91-172-2534230 (Office) 91-172-2687372 (Residence) 91-9876048205 (Mobile); E-mail: firstname.lastname@example.org
The examination of skeletal remains is a challenge to the medical examiner's/coroner's office and the forensic anthropologist conducting the investigation. One of the objectives of the medico-legal investigation is to estimate stature or height from various skeletal remains and body parts brought for examination. Various skeletal remains and body parts bear a positive and linear correlation with stature and have been successfully used for stature estimation. This concept is utilized in estimation of stature in forensic anthropology casework in mass disasters and other forensic examinations. Scientists have long been involved in standardizing the anthropological data with respect to various populations of the world. This review deals with some essential methodological issues that need to be addressed in research related to estimation of stature in forensic examinations. These issues have direct relevance in the identification of commingled or unknown remains and therefore it is essential that forensic nurses are familiar with the theories and techniques used in forensic anthropology.
Positive identification of the deceased is one of the main aims in any forensic medicolegal investigation. Positive identification of the deceased becomes extremely difficult when a collection of bones or dismembered remains are brought for examination. In these situations, the primary aim of the investigation is to determine the age, sex, race, and physical features (height, weight, health), document trauma, and determine the cause of death of the victim (Figure 1). In addition, an attempt is made to determine how long the bones have been exposed to the environment. All investigations begin by attempting to positively identify the remains by employing methods such as DNA fingerprinting, dental comparison, superimposition technique and X-rays.
Estimation of stature is an important parameter of forensic anthropology casework. Being one of the “big fours” (identifying age, sex, stature, and ancestry or race) of forensic anthropology, stature helps a forensic scientist in narrowing down the pool of possible victim matches in the investigation process involving unknown commingled skeletal remains. Numerous studies pertaining to the estimation of stature from various body parts have been cited in the forensic literature (Zeybek, Ergur, & Demiroglu, 2008; Giroux & Wescott, 2008; Krishan, 2008; Cordeiro, Muñoz-Barús, Wasterlain, Cunha, & Vieira, 2009; Menezes, Kanchan, Kumar, Rao, Lobo, Uysal, Krishan, Kalthur, Nagesh, & Shettigar, 2009). Researchers have established a relationship between stature and various body parts like head and face, upper and lower limb bones, vertebral column, hands, and feet and have concluded that stature can be estimated successfully from various body parts and human bones. This review deals with some essential methodological issues that need to be addressed in research related to estimation of stature in forensic examinations.
Technique for measurement and measurement error
Standardized instruments and techniques are used in anthropometric studies pertaining to forensic anthropology casework (Ubelaker, 2008). These techniques are used to create forensic databases and standards pertaining to stature in different population groups. The techniques and instruments used are well described in standard anthropometric textbooks and literature (Vallois, 1965; Hall, Froster-Iskenius, & Allason, 2007). Ideally a stadiometer, a stable and accurate measuring devise with a moveable head board is used for the estimation of stature. Alternatively, tois anthropometrique of Topinard or anthropometer of Martin in four parts (Vallois, 1965), or any other standard anthropometer such as the Harpenden anthropometer (Holtain Limited) can be used for the measurement of stature.
Technique for measuring stature
A standard anthropometric technique of measurement employed among the living is the Vallois (1965) technique. This technique is used in the following manner: Stature (or height vertex) is the vertical distance between the point vertex (the highest point on the head when head is held in the Frankfurt horizontal plane) and the floor. The subject stands erect on a horizontal or resistant plane surface; the upper limbs are pendent, the palms of the hands turned inward and the fingers pointing downward, and heels touching in the military position of attention. It seems preferable that the subject should be placed against a vertical plane, buttocks, and shoulders touching it slightly, but not the occiput. The head is well balanced on the spine, a position more easily attained when the dermograph draws on the left cheek a line from the tragion to the lower point of the left orbital border (Frankfurt horizontal plane). The height of the vertex above the ground is to be measured in this position.
While collecting data, the anthropometer and callipers should be regularly cleaned after each use, for they get dirty, especially after measuring subjects with active perspiration. Body landmarks should be determined beforehand with a dermograph. While taking measurements with callipers, the ends of the branches must not be allowed to depress the skin but only to touch it gently.
Besides, calculations of measurement error in terms of inter and intra observer bias is an imperative ingredient of anthropometry. A minimal error can affect the reliability and precision of the outcome of the study and lead to erroneous conclusions. The precision, reliability, and reproducibility of the measurement are always considered essential in the field of forensic anthropology (Kanchan & Krishan, 2011). The extent to which the measurement error can influence the measurement of bones as well as estimation of stature should be taken into consideration in forensic casework.
Regression formulae in stature estimation
Regression formulae are used in forensic cases involving partial or complete skeletal remains or body parts for the estimation of stature. A regression formula is a statistically developed formula used to establish the relationship of a dependent variable and one or more independent variables. Thus, linear regression is an approach of modeling the relationship between a scalar variable y and one or more variables denoted x. In linear regression, models of the unknown parameters are estimated from the data using linear functions. Such models are called “linear models.” Most commonly, linear regression refers to a model in which the conditional mean of y given the value of x is an affine function of x. Less commonly, linear regression could refer to a model in which the median, or some other quantile of the conditional distribution of y given x is expressed as a linear function of x (Gupta, 2011).
In an estimation of stature by regression analysis, stature remains the dependent variable and the long bone or a body part as independent variable. A regression equation thus derived reflects the relationship between the body part and stature. While calculating the regression formula, the values of constants “a” and “b” are generated mathematically or with the help of computer software; where “a” is the regression coefficient for the dependent variable (stature) and “b” is the regression coefficient for the independent variable (part of the body or a bone from which stature is to be estimated). Thus the regression equation derived is: stature (y) = a + bx, where, “x”= the measurement of a bone or body part. Rollet was the first to explore the correlation between long bone length and stature with a sample of 100 mixed-sex cadavers in 1889 (Cited in: Pearson, 1899). He provided tables from which one could look up stature from long bone length, or vice versa. Pearson, using Rollet's data, produced the first regression equations for estimating stature from long bone lengths (Brandt, 2009). Since then, the regression analysis has successfully been used in standardizing the stature data in populations and consequently in estimating stature in forensic and medico-legal examinations. Trotter and Gleser (1952, 1958) did pioneering work in this regard that is still considered as a groundbreaking study in the field.
Stature estimation and FORDISC 3.0
Another method of stature estimation is FORDISC 3.0. FORDISC (Jantz and Ousley, 1993–2005) is a computer program developed to allow for quick and simple analysis of skeletal measurements through the application of Howell's data set and modern forensic data. It allows easy classification of unknown adults with regards to sex and ancestry, based on known samples (Ubelaker, 1998). Measurements of a variety of bone lengths can be used, and one or multiple bones can be applied to derive a linear regression equation or stature estimate. Stature is estimated by plugging long bone lengths into FORDISC 3.0. The formulae provided by FORDISC based upon forensic data bank and Trotter and Gleser regression equations are the most commonly used in forensic anthropology today. However, FORDISC is still restricted in its usability to those populations from which the data has been utilized to create the program; therefore, it needs much further inputs to ensure that its formulae and its data are applicable to other populations worldwide. In some cases, long bones are unlikely to give results especially where they have been destroyed or damaged too badly for measurements. In such cases, one would have to fall back on basic measurements of smaller bones and other body parts for stature estimation/identification. Thus, in such cases, FORDISC would not be of much use. This would be especially true in the case of commingled remains where different body parts would need to be matched together.
Along with measurement error, factors such as ethnicity, age, asymmetry and diurnal variation are known factors that can affect stature estimation of an individual (Figure 2).
Effect of race/ethnicity/population on stature estimation
The regression formulae derived for stature estimation are population specific (Gill, 2009). Different formulae thus, need to be derived for different population groups, owing to inherent population differences in various dimensions that are attributed to genetic and environmental factors (Kanchan, Menezes, & Kotian, 2008; Krishan & Sharma, 2007). For example, Eskimos of Labrador and Greenland (also known as the Inuit) are shorter and those of Southern Alaska a little taller (the average being 168 cm). In some Sudanese tribes, the males and females can have an average height of 190 cm (6 feet 4 inches) and 180 cm (6 feet) respectively (Hastings & Selbie, 2003), thus much taller when compared to other populations worldwide. As such, scientists are further advised to use more recent formulae devised for a particular population because of secular trends in the populations.
Effect of age on stature estimation
The effect of age on estimation of stature is well known (Trotter & Gleser, 1951). The adult stature declines significantly with age (Rogers, 1982; Coles, Clements, & Evans, 1994). After the age of 25, for every 25 years, stature is shortened by about 2.5 cm (Vij, 2008). Studies have confirmed this well established fact relating to the substantial decrease in stature after the mature stature is attained. Most of the studies show that the stature loss starts at the age of 40 years (van Leer, van Noord, & Seidell, 1992; Galloway, Stini, Fox, & Stein, 2005), and thereafter there is a relatively rapid decrease in stature. Thus, age based categorizations of the subjects should be done especially when individuals above the age of 30 years are included in a study on stature estimation.
Asymmetry of the human body
Bilateral asymmetry of the human body is also known to affect stature estimation in forensic examinations (Kanchan, Mohan Kumar, Pradeep Kumar, & Yoganarasimha, 2008). Bilateral asymmetry is a general and natural phenomenon in the human body (Krishan & Sidhu, 2008). Many studies have reported asymmetry in the human body dimensions as well as in bones. A study conducted by Krishan, Kanchan, & DiMaggio in 2010 showed that significant bilateral asymmetry exists in limb dimensions and they also showed the effect of this asymmetry on estimation of stature. Due to occurrence of bilateral asymmetry, the forensic professionals are advised to first assign the side to the bone or body part and then apply appropriate formula devised for that side in forensic casework.
Diurnal variations in stature
Diurnal variation in stature has been documented and substantial diurnal variation in stature is known to affect stature data in forensic examinations (Krishan & Vij, 2007). Measuring the height of an individual at one defined time in a day in humans is proposed to avoid variations in stature estimation (Voss & Bailey, 1997; Lampl, 1992; Krishan, Sidhu, Kanchan, Menezes, & Sen, 2009). These studies report that stature is maximum in the morning hours and decreases by 1.5–3.0 cm in the evening. The studies further confirm that a greater proportion of decrease in height occurs in the vertebral column due to reduction of elasticity of intervertebral discs (Krishan & Vij, 2007). This ultimately affects the standards generated and formulae derived for estimation of stature. Future studies should possibly take into consideration the issue of diurnal variation while measuring stature of individuals in standardizing the data.
Implications for Forensic Nurses
The present literature review is intended to emphasize the essential issues in future research related to stature estimation in forensic examinations. The professionals and researchers working in the specialized field of forensic anthropology are advised to further refer to some of the many available works on this issue (Eveleth & Tanner, 1976; Ulijaszek & Masci-Taylor, 1994; Hoppa & FitzGerald, 1999; Hauspie, Cameron, & Molinari, 2004; Warren, Walsh-Haney, & Freas, 2008; Pickering & Bachman, 2009; Katzenberg & Saunders, 2008; Burns, 2007; Schmitt, Cunha, & Pinheiro, 2006; Larsen, 2010; Hall, et al., 2007) that would provide a step-by-step approach to identifying the elements, side and normal range of variation related to specific human populations. This would not only ensure accuracy but standardization of procedures and would help in creating a better class of measurements that could be improved with experience. Forensic nursing inevitably involves dealing with diverse cases of sudden and accidental deaths. In some of the more unfortunate cases, the remains are the only clue to identifying the person(s) involved. Sometimes, the remains are separated and in some cases commingled. Estimation of the stature of a person from the present body parts then becomes an important part of identification as well as understanding the chain of events that led to such incidents. Without being sensitized to such possibilities, it is possible that forensic nurses might overlook or even unwittingly destroy the background for making such accurate measurements. Hence, it is imperative that forensic nurses as adjuncts to the system of such forensic identification should be trained in multiple methods in order to assist in this area of work.
This review article on methodological issues has been the outcome of the authors’ experiences in the field of stature estimation and forensic anthropology of over a decade. It is hoped that the results of these experiences would be fruitful to future researchers. This paper has emerged from the preliminary reviews carried out as a part of the Major Research Project awarded by University Grants Commission, New Delhi, India to Dr Kewal Krishan vide grant F.No. 34-120/2008 (SR) dated January 2, 2009. The principal author is grateful to honorable Padamshree Professor R.C. Sobti, Vice-Chancellor, Panjab University, Chandigarh, India, for encouraging research and its publication in international journals of repute. We also thank the three anonymous reviewers whose suggestions have improved the manuscript immensely.