McKern-Stewart method as a technique for analysing age related pubic symphyseal changes: A systematic review and meta-analysis

Abstract

Age estimation is one of the essential criteria in the identification process. The method of age estimation employed depends on the availability of skeletal material brought for forensic examination. McKern and Stewart's method constitutes one of the principal approaches towards pubic symphyseal age estimation. The method entails evaluating morphological changes within the pubic symphysis and subsequently allotting a specific score corresponding to the observed changes. Based on the obtained scores, an age range is then assigned to the remains presenting for examination. The present systematic review was undertaken to ascertain the applicability of the McKern-Stewart method for age estimation. Studies pertaining to the use of the McKern-Stewart method for age estimation in skeletal remains were retrieved by keying in a combination of MeSH terms and other free terms from four databases. The retrieved articles were subjected to a stringent inclusion and exclusion criteria, following which the risk of bias was assessed and the overall quality of evidence was established. Once the final tally of relevant articles was obtained, data specific to the mean age corresponding to each score was extracted. Non-parametric tests and boxplots were employed to compare the mean ages reported across multiple studies. The present systematic review concludes that the McKern-Stewart method can be applied for the purpose of age estimation in skeletal remains. Broader age cohorts for higher scores, as well as, overlapping values for age ranges in relation to the cumulative scores, however, can be considered a limitation for its applicability in forensic case work.

Keywords

Forensic anthropology Age estimation McKern-Stewart method Applicability Skeletal remains

Introduction

Age estimation occupies a prominent niche in the identification process.¹ The role of age estimation in legal and medico-legal investigations includes, and is not limited to, cases dealing with child labour, pedopornography,² juvenile sentencing, and at other times, missing persons and illegal immigrants.³ When skeletal remains present for examination forensic anthropologists rely on various age markers distributed throughout the bony framework.⁴ Morphological changes ensuing within bony surfaces, and/or ossification and fusion of bones have often been employed to ascertain age-at-death of individuals.^5–12 The pelvis is of particular importance as it exhibits significant age-dependent developmental, as well as, degradative changes, allowing for age estimation across a broader age cohort.^{5,9,11,13–18} Within the pelvis, more commonly researched parameters are the pubic symphysis, and the auricular surface of the ilium.^19,20

The pubic symphysis displays high levels of preservation owing to the robusticity and taphonomic resilience conferred to this region within the pelvis. As an added advantage, transpiring pubic symphyseal changes encompass a broad age range, allowing for age estimation in sub-adults, as well as, adults.⁵ The first documented age estimation method using the pubic symphysis was laid down in early 1900s. Todd in 1920, proposed a ten phase pubic symphyseal age estimation method.²¹ Until the middle of the 20^th century, Todd's method remained the primary method employed for pubic symphyseal age estimation.²² Subsequently, multiple methods for age estimation through the evaluation of morphological changes within the pubic symphysis were developed.^5,18,23–28 In 1957, McKern and Stewart devised their component-based age estimation method in order to overcome the drawbacks associated with Todd's phase-based method.¹⁸ Morphological changes within the pubic symphysis were grouped into three distinct components; Dorsal plateau, ventral rampart, and symphyseal rim. Each component was graded on a scale of 0–5 based on a thorough analysis of transpiring morphological changes, following which, a cumulative score of 0–15 was computed. Obtained cumulative score was then employed to assign an age range to skeletal remains. McKern-Stewart's component-based analysis offers certain obvious advantages over its phase-based counterparts by allowing for an objective evaluation of pubic symphyseal changes.²⁹

Subsequently, a number of researchers have undertaken studies employing the use of the McKern-Stewart method, rendering it one of the principal pubic symphyseal age estimation methods.^30–44 A major share of this research attempted to age skeletal remains directly using McKern and Stewart's criteria, while, two studies incorporated the use of CT scans⁴² and casts,⁴⁴ respectively. Additionally, the majority of the research had reported mean ages pertaining to each cumulative score following the format prescribed by McKern and Stewart. Snow⁴³ alternatively, derived simple and polynomial models using the total score so as to confer greater statistical efficiency to the method. Studies conducted on geographically distinct Indian populations reported varying degrees of applicability for specific age groups.^30–41 The standalone CT-based study using the McKern-Stewart method for an Indian population generated regression models, and, reported greater accuracy for a specific age group of 15–40 years.⁴²

Klepinger et al.⁴⁴ evaluated the applicability of the McKern-Stewart and the Suchey-Brooks method, and, advocated the use of the latter in forensic age estimation. Previously undertaken systematic reviews using Suchey-Brooks method have indicated a good accuracy with direct skeletal examination,⁴⁵ as well as, CT-based examinations.⁴⁶ However, such an analysis is presently lacking for the McKern-Stewart method of pubic symphyseal age estimation. The present systematic review was aimed at establishing the applicability of the McKern-Stewart method for age estimation through a meta-analysis of findings reported across multiple studies. With a major share of published research undertaking an investigation of the McKern-Stewart method through a direct examination of skeletal remains and reporting their findings in the format prescribed by McKern and Stewart, the systematic review had to be restricted to include similar studies.

Scope and organization of the review

This systematic review was targeted towards isolating original research focussing on the use of the McKern-Stewart method for analysing pubic symphyseal changes.

McKern and Stewart in their study scored the three components of dorsal plateau, ventral rampart, and symphyseal rim on a scale of 0–5. For the dorsal plateau the scores 0–5 represented the following morphological changes: dorsal margin absent; a slight margin formation first appears in the middle third of the dorsal border; the dorsal margin extends along the entire dorsal border; filling in of grooves and resorption of ridges to form a (beginning) plateau in the middle third of the dorsal demi-face; the plateau, still exhibiting vestiges of billowing, extends over most of the dorsal demi-face; billowing disappears completely and the surface of the entire demi-face becomes flat and slightly granulated in texture, respectively. Similarly, the ventral plateau was scored from 0–5, with the scores representing the following features: ventral bevelling is absent; ventral bevelling is present only at superior extremity of ventral border; bevel extends inferiorly along the ventral border; the ventral rampart begins by means of bony extensions from either or both of the extremities; the rampart is extensive but gaps are still evident along the earlier ventral border, most evident in the upper two-thirds; the rampart is complete, respectively. The third component of symphyseal rim was graded from 0–5 based on the presence of the following features: the symphyseal rim is absent; a partial dorsal rim is present, usually at the superior end of the dorsal margin, it is round and smooth in texture and elevated above the symphyseal surface; the dorsal rim is complete and the ventral rim is beginning to form, there is no particular beginning site; the symphyseal rim is complete and the enclosed symphyseal surface is finely grained in texture and irregular or undulating in appearance; the rim begins to break down and the face becomes smooth and flat, the rim is no longer round but sharply defined with some evidence of lipping on the ventral edge; further breakdown of the rim (especially along superior ventral edge) and rarefaction of the symphyseal face along with disintegration and erratic ossification along the ventral rim, respectively. Described morphological features have been represented in Figure 1.

Figure 1.

Morphological features defined by McKern-Stewart.¹⁸

Relevant papers were retrieved (see subsections on search strategy and eligibility criteria), their overall quality was evaluated, and the risk of bias was established. The mean ages reported in these studies were extracted and subjected to meta-analysis to evaluate the applicability of the method for age estimation.

Materials and methods

The review question “To determine the applicability of the McKern-Stewart method for age estimation from pubic symphyseal changes” was constituted. Methodology for the present systematic review was developed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines.^47,48 The review was subsequently registered with the PROSPERO database (International Prospective Register of Systematic Reviews) under the ID number CRD42020159557.

Search strategy

The search strategy was formulated with the objective of deriving scientific papers employing the McKern Stewart method to estimate and/or verify the age-at-death of skeletal remains. Search equations generated using a combination of MeSH terms and relevant free terms in amalgamation with specific Boolean operators, were keyed into four different databases- PubMed, CENTRAL, Google Scholar and ScienceDirect by two independent researchers (VW and RS). The final search equations generated for extracting relevant studies were:

“age estimation” AND “pubic symphysis”

“forensic age estimation” AND “pubic symphysis”

“age estimation” AND “McKern Stewart” AND “pubic symphysis”

“forensic age estimation” AND “McKern-Stewart” AND “pubic symphysis”

“accuracy” AND “reliability” AND “McKern-Stewart”

Eligibility criteria

Observational studies pertaining to the McKern-Stewart method, conducted in or before 2019 were included for review and meta-analysis. Editorials, literature reviews, case reports and research articles employing methods other than the McKern-Stewart method for pubic symphyseal age estimation were excluded.

No language restrictions were defined at the time of data search and extraction. Duplicate articles across the different databases were excluded. Each of the two researchers (VW and RS) conducted an initial assessment of all extracted titles. Articles wherein the title and abstract did not present sufficient information were retrieved to assess its inclusion/exclusion in the review. From these retrieved articles, original studies which did not employ the McKern-Stewart method for pubic symphyseal age estimation were excluded. Reference lists of relevant original studies were re-evaluated so as to identify any original research overlooked during the initial search. All the articles which addressed the review objective and satisfied the pre-decided inclusion and exclusion criteria were subjected to qualitative and quantitative analysis. Unanimity between the two reviewers was ensured at each stage of this systematic extraction of research articles.

Data extraction

A spreadsheet was created to standardize the process of data extraction. Information pertaining to the title of the paper, research objectives described within the abstract, study population, sample size, URL, description and details of the articles extracted, were noted in a systematic manner. Data in terms of age ranges, mean age and standard deviation pertaining to the total cumulative score ranging from 0–15, as well as, individual scores for each component (0–5) defined by McKern-Stewart was extracted wherever possible. Relevant articles with restricted access, or, where pertinent data was missing were marked. Authors of such original articles were contacted, and, at the end of a nine-month waiting period available data was subjected to qualitative and quantitative assessment.

Assessment of risk of bias of studies

The methodological quality of the available studies was established independently by two researchers (VW and RS) using the QUADAS-2 tool.⁴⁹ QUADAS-2 uses four parameters to evaluate the risk of bias of any study: patient selection, index test, reference standard, and flow and timing. The first three parameters additionally indicate the applicability of these selected studies in answering the review question. Risk of bias was established by answering specific signalling questions in the form of “Yes” or “No” or “Unclear,” provided within the tool. The risk of bias analysis was evaluated using ROBVIS.⁵⁰

Meta-analysis

Meta-analysis of the extracted data was carried out using IBM Statistical Package for Social Sciences (SPSS v 23.0). Differences between mean ages corresponding to each cumulative score (0–15) reported across different studies were analysed using the Kruskal-Wallis test For studies which reported mean ages for scores of each component (0–5), an additional Kruskal-Wallis test was performed. Boxplots corresponding to the same representing the median, minimum, maximum values and interquartile range were plotted.

Results

An initial search across the four databases yielded a total of 1711 articles. These were subsequently reduced to 1040 articles following the removal of duplicates. 1010 articles did not fall within the limits of the inclusion criteria pertaining specifically to the type of study, and were thus excluded. An exhaustive screening of the remaining 30 papers based on the information contained within the title, and, abstract further reduced the search results to 10 original articles. From these remaining 10 observational studies, three additional research articles had to be excluded due to restricted access to their complete versions. Thus, a total of seven original articles fulfilling the inclusion criteria, and, addressing the review question were obtained. These seven observational studies were then subjected to a rigorous qualitative evaluation. Figure 2 illustrates a flow chart with various steps of the Systematic Review (in accordance with PRISMA guidelines).⁴⁷

Figure 2.

Flow chart with various steps of the systematic review along the guidelines proposed by the preferred reporting items for systematic reviews (PRISMA).

In order to determine the applicability of the McKern-Stewart method, data extracted from the seven studies was meta-analysed alongside the data contained within McKern-Stewart's original study. Thus, meta-analysis was conducted on a total of eight studies.

Characteristics of included studies

All studies included herein, barring the primary study by McKern and Stewart are prospective cross-sectional studies carried out within the Indian subcontinent. The age groups included within these studies ranged between 12–75 years. McKern and Stewart¹⁸ analysed 349 skeletal remains of the male sex pertaining to a varied age range. Their obtained findings were represented in terms of age range, mean age, mode and standard deviation, for, cumulative scores (0–15). Sinha and Gupta³⁰ applied the McKern-Stewart method on male skeletal remains and demonstrated their observations in the form of mean age, standard deviation and associated age ranges. Sharma et al.³¹ analysed remains of both sexes using the McKern-Stewart method and commented on the obtained mean ages and standard deviation. Kumar et al.,³² established the mean age for each cumulative score (0–15). Kumar³³ reported the mean age, standard deviation and corresponding age range of 32 remains belonging to the male sex. Prasad et al.³⁵ described the mean age and standard deviation for scores 0–15 based on a thorough examination of 80 male skeletal remains. Singh et al.,³⁴ and Selvamurugan et al.³⁶ reported mean ages for scores 0–15 after a thorough analysis of 53 and 100 skeletal remains, respectively. Thus, a total of 8 studies had reported mean ages for scores 0–15, including McKern-Stewart's original research.

McKern-Stewart in their original study had additionally reported the mean age for scores 0–5 for each of the three components.¹⁸ Similar work was undertaken later by Kumar,³³ Singh et al.,³⁴ and Selvamurugan et al.³⁶ Thus, a total of 4 studies had reported mean ages corresponding to scores 0–5 for each of the three components.

Meta-analysis had to be restricted to mean age data obtained from each study, as, other associated descriptive statistics of range and standard deviation were not always reported.

Risk of bias of studies

All seven studies included for qualitative analysis had reported their patient selection procedure, and inappropriate exclusion of samples was avoided in all these studies. Sharma et al.³¹ and Singh et al.³⁴ reported blinding at the time of analysis. Proof of age, wherever employed, included police and municipal records, often followed by confirmation from relatives of the deceased. All seven studies being prospective cross-sectional studies, information pertaining to sex, wherever applicable, was known to the investigators. Across all seven papers, there was adequate information regarding flow and timing. None of the studies reported a high risk of bias, with most of them indicating some concerns regarding risk of bias wherein relevant information had not been provided in order to comment on certain aspects. The assessment of risk of bias is shown in Table 1.

Table 1.

Risk of bias assessed using QUADAS-2.

Study	Male/ Female	Age (years)	Patient selection	Index test information	Reference standard information	Flow and timing information	Overall risk of bias
Sinha and Gupta³⁰	82/ --	12–75	Low	None	None	Low	Some concerns
Sharma et al.³¹	267/ 69	17 +	Low	None	None	Low	Low
Kumar et al.³²	60/ 52	20–50	Low	None	None	None	Some concerns
Kumar³³	32/ --	15–60	Low	None	None	Low	Some concerns
Singh et al.³⁴	53/ --	12–75	Low	None	None	Low	Low
Prasad et al.³⁵	80/ --	18–65	Low	None	None	None	Some concerns
Selvamurugan et al.³⁶	69/ 31	16–49	Low	None	None	None	Some Concerns

Quality of evidence assessed using GRADEpro

In order to confer the undertaken review additional authenticity, the quality of evidence was analysed and graded using GRADEpro.⁵¹ The present systematic review includes a total of 795 participants spread across seven cross-sectional studies. No serious indicators for high risk of bias, or, indirectness was observed. Lack of significant differences between mean ages for each score, across studies, indicates high precision and consistency. There is no direct evidence of publication bias or other associated biases, as, age estimation was carried out by an objective analysis of skeletal remains. An overall moderate to high evidence in favour of use of the McKern-Stewart method for age estimation in skeletal remains was obtained. (Table 2).

Table 2.

Quality of evidence evaluation of studies using GRADEpro.

Outcome	All included studies
No. of studies	7
No. of participants	795
Study design	Prospective cross-sectional (Non-RCT)
Risk of bias*	Not serious
Inconsistency*	Not serious
Indirectness*	Not serious
Imprecision*	Not serious
Other considerations (Publication bias)*	Possible publication bias
Overall Quality of evidence	Moderate

*Quality assessment parameters.

Meta-analysis

Kruskal-Wallis test for distribution of mean ages corresponding to cumulative scores 0–15 across different studies

The Kruskal-Wallis test yielded no significant difference between mean ages across studies for cumulative scores 0–15 (p > 0.05) as shown in Table 3. Additionally, no sexual dimorphism was observed for distribution of mean ages corresponding to scores 0–15 (p > 0.05).

Table 3.

Differences in mean ages for cumulative scores (0–15) and scores for each component (0–5) across different studies.

McKern-Stewart cumulative score		Dorsal plateau		Ventral rampart		Symphyseal rim
Score	p-value	Score	p-value	Score	p-value	Score	p-value
0	0.317	0	0.368	0	0.392	0	0.392
1–2	0.429	1	0.392	1	0.392	1	0.392
3	0.429	2	0.392	2	0.392	2	0.392
4–5	0.429	3	0.392	3	0.392	3	0.392
6–7	0.423	4	0.392	4	0.392	4	0.392
8–9	0.423	5	0.392	5	0.392	5	0.392
10	0.429
11–13	0.429
14	0.429
15	0.429

*Significance level of p=0.05.

Kruskal-Wallis test for distribution of mean age for each component defined by McKern-Stewart

The distribution of mean ages for scores 0–5, for each of the three components (dorsal plateau, ventral rampart, and, symphyseal rim), across studies, showed no significant difference (p > 0.05) as shown in Table 3.

Boxplots for distribution of mean age corresponding to scores 0–15

Boxplots for scores 0–15 representing the median, minimum and maximum values, along with the interquartile range associated with each cumulative score has been shown in Figure 3. The obtained minimum-maximum age range pertaining to a total score of 0 was 17.29–20.99 years. Morphological changes associated with score 1–2 were observed between 17.00–25.00 years. Score 3 yielded an age range of 17.00–25.25 years. For the cumulative score 4–5, the corresponding age range was 20.33–25.50 years. With a score of 6–7, an age range of 22.42–30.00 years was observed. Boxplots for the scores 8–9, and, 10 indicated ranges of 24.14–31.87 years, and 24.50–35.50 years, respectively. The age group showing the desired morphological characteristics associated with the score 11–13, across all the studies, was 29.18–38.83 years. For score 14, the obtained age range was 35.84–48.88 years. For a cumulative score of 15, the associated age range was observed to be 49.00–59.40 years, with 41.00 years representing the outlier for this specific group.

Figure 3.

Box and whisker plots showing distribution of mean age corresponding to the scores defined by McKern-Stewart¹⁸ across all studies.

Boxplots corresponding to scores 0–5 for dorsal plateau

The minimum-maximum age range for characteristics pertaining to score 0, across studies, was observed to be 15.67–19.30 years. For score 1, the obtained age range was 16.00–21.70 years. With score 2, an age range of 20.00–23.75 years was observed. For a score of 3, the defined morphological characteristics were observed within an age range of 21.50–28.90 years. A score of 4 yielded a corresponding age range of 26.10–34.42 years. Lastly, with the score 5, 32.90–51.87 years, was the obtained range. (Figure 3)

Boxplots corresponding to scores 0–5 for ventral rampart

A score of 0 yielded an age range of 16.87–18.30 years. For score 1, 18.90–23.00 years represented the associated age range. With score 2, the minimum-maximum age range for the defined set of morphological characteristics was observed to be 19.80–24.00 years. The obtained range for score 3 was 20.10–28.12 years. Score 4 yielded a range of 23.30–35.25 years and 31.90–49.84 years was the range corresponding to score 5. (Figure 3)

Boxplots corresponding to scores 0–5 for symphyseal rim

Score 0 yielded a minimum-maximum range of 16.20–19.20 years. With score 1, the obtained age range was 20.95–23.80 years. 25.00–31.00 years was the corresponding range associated with score 2. For score 3, 28.08–30.30 years was the obtained range. With score 4, onset and progression of the defined set of morphological characteristics was observed between 35.90–42.00 years. Lastly, for score 5 of symphyseal rim, the obtained age range was 41.10–53.80 years (Figure 3).

Discussion

The McKern-Stewart method¹⁸ renders pubic symphyseal age estimation a more objective and simpler approach. They methodically grouped morphological changes ensuing within the pubic symphysis originally described by Todd²¹ into three distinct components. The onset and progression of morphological changes transpiring within each component was further broken down and assigned scores ranging from 0–5. The cumulative score computed using the three components were employed to assign an age range to skeletal remains presenting for examination. Subsequently, numerous researchers embarked on age estimation studies using the McKern-Stewart method.^30–44

All the studies included in the present systematic review reported a low to moderate risk of bias, a plausible explanation for which is the lack of information concerning index and reference tests. An overall moderate to high evidential quality was indicated. Studies by Sinha and Gupta³⁰ indicated lower age ranges for cumulative scores 1–5, and higher age ranges for scores 6–13, when compared to McKern-Stewart's original study. An upper limit of 69.175 years was obtained corresponding to a score of 15. Component wise analysis by Kumar³³ similarly yielded higher mean age values for scores 0–5 than those defined by McKern and Stewart.

The distribution of mean ages for cumulative scores (0–15) across these studies, including McKern-Stewart's original work shows no significant difference, suggesting that there is a certain uniformity in the prevalence of age related pubic symphyseal changes. Component-wise analysis also points towards a certain concordance in the distribution of mean ages corresponding to scores (0–5) for each component, across different studies. Boxplots for mean ages pertaining to scores 0–15 for all included studies indicates that it is possible to assign an age range to remains based on the observed changes. Boxplots for scores 0–5 for each of the three components suggest a similar trend. Even though an overlap is observed for age ranges corresponding to scores 0–15, a minimum age can successfully be assigned to the remains in question. This is of particular forensic significance in medico-legal cases including missing person identification.

Kumar et al.³² reported a lack of sexual dimorphism in the distribution of mean ages for cumulative scores (0–15). This contradicts the findings of Sharma et al.,³¹ Berg²³ and Gilbert-McKern,²⁴ who reported a significant sex difference in the expression of morphological changes within the pubic symphysis. However, a significantly lower number of female remains were analysed within this study, necessitating the need to further scrutinize and address concerns regarding the applicability of the McKern-Stewart method for female skeletal remains.

Since the meta-analysed studies scrutinized the method on Indian populations, it can be assumed that the distribution of mean ages for this biogeographical region falls within the limits of McKern and Stewart's findings. Thus, the McKern-Stewart method can successfully be employed for age estimation on Indian populations. However, the sample size of observational studies included for meta-analysis comprised of a comparatively narrower age range. Additionally, specific sexual dimorphism targeted research, estimating the overall accuracy of the McKern-Stewart method are currently wanting. Further evidence pointing towards the applicability of the method for age estimation requires carrying out population specific studies around the globe. Age-related hormonal changes, modifications in bone texture, and, other lifestyle factors are known to influence the observed morphological changes alongside biogeography and demography, warranting the need to carry out studies targeting different populations.⁵²

The component-based analysis, in consortium with the observed uniformity in the distribution of mean ages for cumulative scores (0–15) across studies points towards the reliability of morphological features described by McKern-Stewart. However, with increasing cumulative score, broadening of age ranges associated with these scores is observed. A similar broadening of age ranges is observed with increasing scores for each individual component described by McKern-Stewart. A possible explanation for this is that McKern and Stewart's study sample for establishing their method of pubic symphyseal age estimation inadvertently comprised of younger individuals. This resulted in narrower yet overlapping age ranges for scores 0–13, and extremely broad ranges for scores 14, and 15, dampening the role of the method in forensic age estimation. From a forensic point of view, accuracy and precision is, and continues to remain, a longstanding challenge faced by forensic anthropologists. Accurate age estimates, along with other parameters of sex and stature are vital to the process of human identification. Broader overlapping age cohorts associated with the McKern-Stewart method pose a drawback to ascertaining the identity of skeletal remains presenting for medico-legal investigations.

One way to overcome this drawback and achieve greater accuracy is through a multifactorial approach to age estimation, by employing additional age markers within the pelvis. The acetabulum and auricular surface of the ilium, present as favourable, durable evidences which allow for accurate age estimation in older individuals.^9,15–17 Such a multifactorial approach has the potential to greatly improve the reliability associated with age estimation.⁵³ Regression-based models offer another efficient alternative over the current application of the McKern-Stewart method.⁴² Such a regression-based approach can improve applicability of the method by reducing the associated error rate resulting from the use of broader age cohorts. Additionally, regression models generated using individual components can aid in age estimation without mandating the need to visualize and score even the highly degraded features. Population specific studies employing the use of such regression models need to be undertaken as they can help confer additional validity to the McKern-Stewart method in favour of its applicability in forensic investigations. An alternative approach is to estimate the varying accuracy of each morphological feature towards the overall process of age estimation. Taking into consideration such a differential contribution can help generate more accurate age estimation models using the McKern-Stewart method. The use of such models, in consortium with a CT-based age estimation approach can further improve the implementation process by doing away with the time and resource intensive maceration process.

Conclusion

The findings of the present systematic review suggest uniformity in the distribution of mean ages corresponding to a defined set of morphological characteristics defined by McKern and Stewart, across multiple studies. No significant differences between mean ages were obtained for each cumulative (0–15), as well as, individual score (0–5) in these studies. Thus, it can be concluded that the McKern-Stewart method can be applied successfully to an Indian population for the purpose of age estimation. Broader age cohorts for higher scores, as well as, overlapping values for age ranges in relation to the cumulative scores, however, can be considered a limitation for its applicability in forensic case work. It can be overcome through the development of population specific regression models, and by incorporating a multifactorial approach to the process of age estimation.

Footnotes

Acknowledgements

This research article is a part of the ongoing doctoral research being conducted by the principal author (Varsha Warrier) at the Department of Forensic Medicine and Toxicology, All India Institute of Medical Sciences, Jodhpur, India. The principal author is grateful to the University Grants Commission, New Delhi for awarding research fellowship (UGC-JRF) for pursuing PhD. The authors are also thankful to the individuals who participated in this study, and to the authorities of the institution for allowing us to conduct this research.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The review was methodologically registered under the PROSPERO database (International Prospective Register of Systematic Reviews) with the ID number CRD42020159557. No support in the form of grants.

ORCID iDs

Rutwik Shedge

Kewal Krishan

Tanuj Kanchan

References

Krogman

Iscan

. The Human Skeleton in Forensic Medicine. Springfield, Illinois: Charles C Thomas Pub Ltd, 1986.

Franklin

Flavel

Noble

, et al. Forensic age estimation in living individuals: Methodological considerations in the context of medico-legal practice [internet]. Res Rep Forensic Med Sci 2015; 5: 53-66. [cited 2019 Nov 27]. Available from: 5: 53–66. [cited 2019 Nov 27]. Available from: https://www.dovepress.com/forensic-age-estimation-in-living-individuals-methodological-considera-peer-reviewed-fulltext-article-RRFMS (Last accessed 25 October, 2021).

Schmeling

Dettmeyer

Rudolf

, et al. Forensic age estimation: methods, certainty, and the law. Deutsches Aerzteblatt Online [Internet] 2016; 113: 44–50. [cited 2019 Nov 13]; Available from: https://www.aerzteblatt.de/10.3238/arztebl.2016.0044.

Villa

Lynnerup

. Age estimation of skeletal remains: principal methods [internet]. Res Rep Forensic Med Sci 2014; 4: 3–9. [cited 2019 Nov 29]. Available from: https://www.dovepress.com/age-estimation-of-skeletal-remains-principal-methods-peer-reviewed-fulltext-article-RRFMS.

Brooks

Suchey

. Skeletal age determination based on the os pubis: a comparison of the acsádi-nemeskéri and suchey-brooks methods. Hum Evol 1990; 5: 227–238.

Shedge

Kanchan

Garg

, et al. Computed tomographic analysis of medial clavicular epiphyseal fusion for age estimation in Indian population. Leg Med (Tokyo) 2020; 46: 101735.

Shedge

Kanchan

Kumar Garg

, et al. Age estimation from sternebral fusion in an Indian population - A computed tomographic evaluation. Leg Med (Tokyo) 2021; 53: 101951.

Lovejoy

Meindl

Pryzbeck

, et al. Chronological metamorphosis of the auricular surface of the ilium: a new method for the determination of adult skeletal age at death. Am J Phys Anthropol 1985; 68: 15–28.

Rissech

Estabrook

Cunha

, et al. Using the acetabulum to estimate age at death of adult males. J Forensic Sci 2006; 51: 213–229.

10.

Ruengdit

Prasitwattanaseree

Mekjaidee

, et al. Age estimation approaches using cranial suture closure: a validation study on a Thai population. J Forensic Leg Med 2018; 53: 79–86.

11.

Passalacqua

. Forensic age-at-death estimation from the human sacrum. J Forensic Sci 2009; 54: 255–262.

12.

Ebeye

Okoro

Ikubor

. Radiological assessment of age from epiphyseal fusion at the wrist and ankle in southern Nigeria. Forensic Sci Int: Reports 2021; 3: 100164.

13.

Bartolini

Pinchi

Gualco

, et al. The iliac crest in forensic age estimation: evaluation of three methods in pelvis X-rays. Int J Legal Med 2018; 132: 279–288.

14.

Zhang

Dong

X-A

Fan

, et al. Age estimation based on pelvic ossification using regression models from conventional radiography. Int J Legal Med 2016; 130: 1143–1148.

15.

Calce

. A new method to estimate adult age-at-death using the acetabulum. Am J Phys Anthropol 2012: 148(1): 1–13.

16.

Buckberry

Chamberlain

. Age estimation from the auricular surface of the ilium: a revised method. Am J Phys Anthropol 2002; 119: 231–239.

17.

Osborne

Simmons

Nawrocki

. Reconsidering the auricular surface as an indicator of age at death. J Forensic Sci 2004; 49: 1–7.

18.

Skeletal age changes in young American males, analysed from the standpoint of age identification, by Thomas W. McKern, et al. | The Online Books Page [Internet]. [cited 2020 Jul 20]. Available from: http://onlinebooks.library.upenn.edu/webbin/book/lookupid?key=ha002075809.

19.

Miranker

. A comparison of different age estimation methods of the adult pelvis. J Forensic Sci 2016; 61: 1173–1179.

20.

Moraitis

Zorba

Eliopoulos

, et al. A test of the revised auricular surface aging method on a modern European population. J Forensic Sci 2014; 59: 188–194.

21.

Todd

. Age changes in the pubic bone. I. The male white pubis. Am J Phys Anthropol 1920; 3: 285–334.

22.

Sarajlić

Gradaščević

. Morphological characteristics of pubic symphysis for age estimation of exhumed persons. Bosn J Basic Med Sci 2012; 12: 51–54.

23.

Berg

. Pubic bone age estimation in adult women. J Forensic Sci 2008; 53: 569–577.

24.

Gilbert

McKern

. A method for aging the female os pubis. Am J Phys Anthropol 1973; 38: 31–38.

25.

Hanihara

Suzuki

. Estimation of age from the pubic symphysis by means of multiple regression analysis. Am J Phys Anthropol 1978; 48: 233–239.

26.

Hartnett

. Analysis of age-at-death estimation using data from a new, modern autopsy sample--part I: pubic bone. J Forensic Sci 2010; 55: 1145–1151.

27.

Fleischman

. A comparative assessment of the chen, et al. And suchey-brooks pubic aging methods on a north American sample. J Forensic Sci 2013; 58: 311–323.

28.

Chen

Zhang

Tao

. Determination of male age at death in Chinese han population: using quantitative variables statistical analysis from pubic bones. Forensic Sci International 2008; 175: 36–43.

29.

Shirley

Ramirez Montes

. Age estimation in forensic anthropology: quantification of observer error in phase versus component-based methods. J Forensic Sci 2015; 60: 107–111.

30.

Sinha

Gupta

. A study on estimation of age from pubic symphysis. Forensic Sci Int 1995; 75: 73–78.

31.

Sharma

Gargi

Kalsey

, et al. Determination of age from pubic symphysis: an autopsy study. Med Sci Law 2008; 48: 163–169.

32.

Kumar

Tyagi

Banerjee

. Estimation of age by pubic symphysis metamorphosis in east Delhi population. Undefined [Internet] 2009. [cited 2022 Feb 3]; Available from: https://www.semanticscholar.org/paper/Estimation-of-age-by-pubic-symphysis-metamorphosis-Kumar-Tyagi/7676115c16651509e9707550645efe7d3584f079.

33.

Kumar

. Estimation of age by morphological changes in symphyseal surface of pubis in males - a cross sectional study. Bangalore: Rajiv Gandhi University of Health Sciences, Karnataka, Bangalore, 2011.

34.

Singh

Aggrawal

Mittal

. A comparative study of symphyseal surface of pubic bone at autopsy for age estimation using McKern-Stewart criteria. Medico-Legal Update 2013; 13: 102.

35.

Prasad

Babu

Reddy

, et al. Reliability of McKern and Stewart method for estimating age in males: an autopsy study in a teaching hospital in tirupathi. J Evid Based Med Healthc 2015; 2: 3940–3944.

36.

Selvamurugan

Gokula Pandia Sankar

Tamilmani

, et al.

Component analysis of pubic symphysis morphology: is it still a gold standard tool of age assesment?

Indian J Forensic Med Toxicol 2019; 13: 53–56.

37.

Javvadi

Agarwal

Mestri

, et al. A cross sectional study of age related pubic symphyseal changes in 3 rd and 4 th decades of life. Medico-Legal Update 2016; 16: 114.

38.

Kumaran

Singh

Bansal

, et al. Applicability of three component system of age estimation in haryana population. Med-Leg Update 2019; 19: 36–41.

39.

Janardhan

Subhedar

Khan

. Determination of age by pubic symphysis metamorphosis in hyderabad population. Medico-Legal Update 2016; 16: 232–234.

40.

Pal

Tamankar

. Preliminary study of age changes in gujarati (Indian) pubic bones. Indian J Med Res 1983; 78: 694–701.

41.

Madanraj

Shetty

. Age estimation from morphological changes in pubic symphysis. Indian J Forensic Community Med 2021; 8: 161–165.

42.

Gorchiya

Kochar

Pathak

, et al. The application of McKern-Stewart's Method for age estimation using computed tomography imaging of the pubic symphysis of living individuals. Forensic Imaging 2020; 23: 200415.

43.

Snow

. Equations for estimating age at death from the pubic symphysis: a modification of the McKern-Stewart method. J Forensic Sci 1983; 28: 864–870.

44.

Klepinger

Katz

Micozzi

, et al. Evaluation of cast methods for estimating age from the os pubis. J Forensic Sci 1992; 37: 763–770.

45.

Schanandore

Wolden

Smart

. The accuracy and reliability of the suchey-brooks pubic symphysis age estimation method: systematic review and meta-analysis. J Forensic Sci 2022; 67: 56–67.

46.

Warrier

Kanchan

Shedge

, et al. Computed tomographic age estimation from the pubic symphysis using the suchey-brooks method: A systematic review and meta-analysis. Forensic Sci Int 2021; 325: 110811.

47.

PRISMA. [Internet]. [cited 2020 Jul 30]. Available from: http://www.prisma-statement.org/.

48.

Hermetet

Saint-Martin

Gambier

, et al. Forensic age estimation using computed tomography of the medial clavicular epiphysis: a systematic review. Int J Legal Med 2018; 132: 1415–1425.

49.

Bristol U of. QUADAS-2 [Internet]. University of Bristol; [cited 2022 Feb 3]. Available from: https://www.bristol.ac.uk/population-health-sciences/projects/quadas/quadas-2/.

50.

Risk of bias tools - robvis (visualization tool) [Internet]. [cited 2022 Feb 3]. Available from: https://www.riskofbias.info/welcome/robvis-visualization-tool.

51.

GRADEpro [Internet]. [cited 2022 Feb 4]. Available from: https://www.gradepro.org/.

52.

Herrera

Retamal

. Reliability of age estimation from iliac auricular surface in a subactual Chilean sample. Forensic Sci Int 2017; 275: 317.e1–317.e4.

53.

Lovejoy

Meindl

Mensforth

, et al. Multifactorial determination of skeletal age at death: a method and blind tests of its accuracy. Am J Phys Anthropol 1985; 68: 1–14.