2 Biomedical markers and the assessment of chronological age

An age of uncertainty

Inquiry into the treatment of individuals suspected of people smuggling offences who say that they are children

Chapter 2: Biomedical markers and the assessment of chronological age

1 Introduction

Since 2001, the primary method of assessing whether an individual is under the age of 18 years in the context of criminal proceedings in Australia has been through the analysis of an x-ray of the young person’s wrist. As this Inquiry is considering the treatment of young Indonesian males suspected of people smuggling who say that they are children, it is important to consider the appropriateness of this and other age assessment processes in this context. Consequently, this chapter considers whether it is appropriate to adduce expert analysis of a wrist x-ray, or any other biomedical marker, as evidence in a criminal proceeding on the issue of whether a young male is under the age of 18 years. As noted in Chapter 1, the Commission is not aware of any female person having been suspected of people smuggling.

This chapter considers the use of wrist x-ray analysis as a means of assessing age, and whether it is sufficiently informative of whether a young male person has attained 18 years of age. The question of how wrist x-ray analysis has been used in practice during investigation and prosecution processes, and the impact of its use in specific cases, is considered in Chapter 4

Although dental x-ray analysis has not been used as a means of assessing age in the context of criminal proceedings in Australia, the process has been offered to a number of individuals suspected of people smuggling offences whose age was in doubt. Also, as recently as November 2011, the Commonwealth was considering specifying dental x-rays as a prescribed procedure for determining age for the purposes of the Crimes Act 1914 (Cth). Consequently, consideration is additionally given in this chapter to the question of whether it is appropriate to use dental x-ray analysis as a means of assessing age in criminal proceedings.

This chapter also gives brief consideration to the analysis of other biomedical markers for the purposes of age assessment in criminal proceedings. It specifically considers the following:

• the use of wrist x-rays for the assessment of chronological age
• the use of dental x-rays for the assessment of chronological age
• the ethical implications of the use of x-rays for the assessment of chronological age
• the use of other biomedical markers for the assessment of chronological age
• the use of multifactorial medical approaches for the assessment of chronological age
• the use of a multi-disciplinary approach for the assessment of chronological age.

2 The use of wrist x-ray analysis for the assessment of chronological age

Given the heavy reliance, since 2008, on wrist x-ray analysis as a means of assessing chronological age in people smuggling matters where age is in doubt, one of the most important questions for this Inquiry is: how informative is wrist x-ray analysis for assessing whether a young male person is over 18 years of age?

As discussed in Chapter 1 above, the chronological age of an individual suspected of a people smuggling offence is an important issue for the following reasons.

First, it will ordinarily be inconsistent with the Prosecution Policy of the Commonwealth for a juvenile to be prosecuted unless the seriousness of the alleged offence or the circumstances of the juvenile concerned dictate otherwise.^[72] This policy position presumably informed the more particular policy position articulated by the Australian Government that juveniles suspected of people-smuggling offences will only be prosecuted in exceptional circumstances on the basis of their significant involvement in a people smuggling venture or multiple ventures.^[73]

Second, there are mandatory minimum penalties for certain people smuggling offences, including the ‘aggravated offence of people smuggling (at least 5 people)’, the offence with which most individuals suspected of a people-smuggling offence are charged.^[74] The minimum penalty for this offence is, for a first offence, a sentence of imprisonment of at least five years with a non-parole period of at least three years.^[75] Importantly, however, this mandatory minimum penalty has no application ‘if it is established on the balance of probabilities that the person was aged under 18 years when the offence was committed’.^[76]

Hence, if wrist x-ray analysis is not adequately informative of whether a young person suspected of a people smuggling offence was over the age of 18 years when the alleged offence was committed, the potential consequences of reliance on such analysis are serious. They include prosecuting a juvenile contrary to the Prosecution Policy of the Commonwealth and a more specific policy position adopted by the Australian Government; prosecuting a juvenile in an adult court; the imposition on a juvenile of a mandatory sentence of imprisonment applicable only to adults; and the detention of a juvenile in an adult facility. Each of the above potential consequences would, if realised, involve a failure to respect rights recognised by the Convention on the Rights of the Child (CRC).

There is also an ethical dimension to a decision to subject any person, but particularly a child, to radiation, as radiation is potentially harmful. This ethical dimension will be the more acute when the exposure to radiation is for an administrative, rather than a medical, purpose and where the benefit from the exposure is doubtful. This issue is considered further below.

This section of the report discusses the process for taking and analysing a wrist x-ray in Australia. It then considers the following questions:

• Is wrist x-ray analysis using the Greulich and Pyle Atlas (GP Atlas) informative of whether a person has attained 18 years of age?
• How wide is the normal variation in the age at which young people generally achieve skeletal maturity?
• Does the GP Atlas have limitations when used for assessing the chronological age of a population dissimilar to the study sample on which the GP Atlas is based?
• Can errors of interpretation impact on the accuracy of wrist x-ray analysis of skeletal age?

2.1 The process for taking and analysing a wrist x-ray in Australia

The use of wrist x-ray analysis as evidence of age has relied on opinion evidence, ordinarily given by a radiologist. This evidence concerns the probable chronological age of an individual based upon the expert’s assessment of the skeletal maturity of his wrist as shown by the x-ray.

The ordinary process of obtaining this evidence involves the following steps:

• After the required consents are obtained, the Australian Federal Police (AFP) organises for a radiographer to take a wrist x-ray, usually of the individual’s left wrist.
• This image is interpreted by a radiologist who, after comparing the x-ray with the plates contained in the GP Atlas, gives an estimation of the individual’s likely skeletal age.
• Unless that estimation is under the age of 19 years, the AFP then requests a detailed expert report from a second radiologist which can be relied on in a legal proceeding; this report calculates a statistical probability that the individual is under the age of 18 years.

In Australia, when wrist x-rays are taken for the purposes of age assessment, they are usually interpreted with the aid of the publication A Radiographic Atlas of Skeletal Development of the Hand and Wrist, published in 1950 by Stanford University Press and Oxford University Press. The second edition of this Atlas was published in 1959. This Atlas is commonly referred to as the ‘Greulich and Pyle Atlas’ as it was compiled by Professor William Walter Greulich of Stanford University School of Medicine and Dr Sarah Idell Pyle, Research Associate, Departments of Anatomy, Western Reserve University and Stanford University Schools of Medicine. The second edition of this Atlas, which has been relied on in every case in which a wrist x-ray of an individual suspected of a people smuggling offence has been placed before a court in Australia, will be referred to as the ‘GP Atlas’.

The GP Atlas consists of a series of plates of standard hand-wrist x-rays for specified skeletal ages.^[77] The standard plates in the GP Atlas were selected from x-ray films of healthy, white children of North European ancestry in the United States whose families may be assumed to have been somewhat above the average in economic and educational status. Each standard is based on a group of 100 children of that chronological age. The x-rays on which the GP Atlas is based were originally obtained as part of a study by the Brush Foundation which was conducted between 1930 and 1942.^[78]

While there are other atlases that have been developed to help assess skeletal age, including the TW3 manual and its TW2 predecessor,^[79] the GP Atlas is most commonly used, and appears to be the only atlas to have been used for age assessment purposes in Australia.

Wrist x-ray analysis is most commonly used by medical practitioners to assess the skeletal development of a child whose chronological age is known. An assessment of this kind is undertaken for the purpose of evaluating other aspects of the child’s growth and development. Ordinarily, the medical practitioner wants to know how the child’s development compares with that of other children of the same sex and age.^[80] By comparing a child’s wrist x-ray with the standard plates contained in the GP Atlas, a medical practitioner can compare the skeletal development of that child with other children of the same sex and age for the purpose of forming a judgment about the child’s health and development status.^[81]

The specialised knowledge which informs reliance on the GP Atlas, and other comparable atlases, for the purpose of assessing skeletal development of a child whose chronological age is known includes the demonstrated close correspondence between the developmental status of the human reproductive system and the human skeletal system as disclosed by an x-ray of the hand and wrist. An x-ray of a young person’s wrist affords an objective measure of the amount of progress which the young person has made towards attaining physical (including skeletal) maturity.^[82]

It is important to bear in mind that the authors of the GP Atlas were not seeking to create a method of establishing the chronological age of young people. Rather, their concern was to establish a means of assessing the skeletal development of children whose chronological age is known. The authors of the GP Atlas state, in the text that precedes the standard plates, that in constructing the standards contained in the GP Atlas their first object was to select film which would provide an adequate record of discernible stages of normal development of the bones of the hand and wrist. Their second object was to relate those stages as accurately as possible to the chronological age at which they typically occurred in the children of their study sample. They indicate that ‘[i]n a sense, the first of these objectives is more important than the second’.^[83]

The critical question for this Inquiry, which is addressed below, is: how informative is an assessment of a wrist x-ray undertaken by reference to the GP Atlas for the purpose of determining whether an individual Indonesian male was over the age of 18 years at the time of his alleged offence?

2.2 Wrist x-ray analysis using the GP Atlas is not informative of whether a male has attained 18 years of age

Expert opinion evidence concerning a young person’s chronological age which is based on a wrist x-ray is ultimately dependent on statistics. The manner in which wrist x-ray analysis has been undertaken in the cases under consideration has involved the expert radiologist seeking to calculate the statistical probability of a person with the degree of skeletal maturity shown by the x-ray being under the chronological age of 18 years.

Statistics by their nature are concerned with populations, not with individuals. For this reason, great care must be taken when placing reliance on statistics to draw an inference about a particular individual. The extent to which it is appropriate to rely on statistics as evidence concerning an individual is discussed in more detail in the paper written by the President of the Commission which is reproduced in Appendix 5.

Even if this issue is put to one side, a limitation inherent in the use of a wrist x-ray to assess a young person’s chronological age is that a wrist x-ray is only informative to the point at which the individual achieves skeletal maturity. Thereafter the skeletal status of the young person’s wrist will remain unchanged. For this reason, if a wrist x-ray is to be used for the purpose of assessing whether a young male person has achieved the age of 18 years, it is critical to know the age at which, on average, young males achieve skeletal maturity.

The fact that a male person has a mature wrist will not indicate that he is over the age of 18 years if, on average, males achieve a mature wrist when under the age of 18 years or, alternatively, at an age sufficiently close to 18 years of age to render the GP Atlas, or any other source of comparative data, unhelpful.

(a) The GP Atlas does not consider the chronological age at which skeletal maturity is attained

As noted above, the GP Atlas depicts, by a series of standard x-ray plates, the degree of skeletal development that its authors considered representative of the group of children in its study sample, at successive chronological ages until the achievement of skeletal maturity.

The final standard male x-ray in the GP Atlas shows a mature wrist which is assigned the age of 19 years. It may be deduced from the method of selection used by the authors, which is described below, that among the individuals whose hand films were considered for inclusion in the GP Atlas for this standard, most were skeletally mature. The immediately preceding male standard is assigned the age of 18 years and shows a wrist extremely close to maturity. As the evidence of Professor Tim Cole discussed below indicates, it is appropriate to regard each of these plates as representative of a population of young males from which early maturing individuals (that is, for the plate for 18 years, those who were mature before they reached the age of 17 years and for the plate for 19 years, those who were mature before they reached the age of 18 years) had been removed.

The text which accompanies the plates in the GP Atlas reveals that the standards were selected from x-rays taken on, or relatively close to, the 18^th and 19^th birthdays of the individuals whose wrists they depict.^[84] The GP Atlas contains no male standard for any age between 18 and 19 years; that is, there is no plate, for example, for 18.5 years.

The method of selection of the standards in the GP Atlas is described in the text as follows:

Each of the standards in this Atlas was selected from one hundred films of children of the same sex and age. The films of each of the series were arrayed in the order of the relative skeletal status, from the least mature to the most mature. In most cases the film chosen as the standard is the one which, in our opinion, was most representative of the central tendency of the particular array. The anatomical mode was frequently, but not always, at or near the midpoint of the distribution of the one hundred films.^[85]

The GP Atlas does not give the average age at which the children in the study sample achieved skeletal maturity. This was not the purpose of the authors’ study. Interestingly, however, Table III of the GP Atlas shows the variability of the skeletal age of boys included in the Brush Foundation Study. The data presented in this table are derived from the x-rays that formed the basis of the GP Atlas, but they are based on an assessment of those x-rays made by reference to an earlier set of standards.^[86] The authors of the GP Atlas conclude that ‘there is no reason to believe that the variability would be significantly greater than that recorded in the tables if those assessments had been based on the standards presented [in the GP Atlas]’.^[87] Some information about the average chronological age at which subjects achieved a specific skeletal age can be inferred from this table.

The following is an extract from Table III of the GP Atlas:

Table III concludes at the chronological age of 17 years.^[88]

Table V of the GP Atlas provides the mean and standard deviation for skeletal age for boys whose growth and development were studied over a long period of time by researchers at the Harvard School of Public Health in Boston. The data presented in this table are based on an assessment of the x-rays using the standards contained in the GP Atlas.^[89]

The following is an extract from Table V of the GP Atlas:^[90]

Table V also concludes at the chronological age of 17 years.^[91]

The failure of these two tables to provide information for the age of 18 years gives rise to an inference that the authors of the GP Atlas did not consider that a mean or standard deviation for skeletal age was meaningful at the chronological age of 18 years.

Support for this inference is found in the written submission and in the oral evidence to this Inquiry by Professor Cole, Professor of Medical Statistics, University College, London. The application of statistics to human growth has been Professor Cole’s main research focus for the past 30 years.^[92] Professor Cole argues that the reason each of the above tables concludes at the age of 17 years is that there were too few children with older bone ages to be included; that is, that most children older than 17 years have mature wrist x-rays. He draws additional support for this conclusion from the drop in the number of children in the Brush Foundation Study, particularly after the age of 14 years.^[93]

He explains that the GP Atlas did not attempt to document the different ages at which a wrist might mature, as the purpose of the GP Atlas was to plot all the stages of skeletal development up until wrist maturity. At the Inquiry’s hearing for medical experts Professor Cole explained:

Well this is the fundamental question. How can we get a handle on the age of attainment of a mature x-ray. That’s why I went to TW3 [an alternative database to the GP Atlas], because it had a table which gave ages of attainment. There’s nothing about that in Greulich and Pyle because Greulich and Pyle were not remotely interested in mature x-rays as needs emphasising. They really were not interested in them.^[94]

Thus, Professor Cole argues, the GP Atlas does not help a medical practitioner assess whether a person with a mature x-ray might be under or over the age of 18 years.

(b) The Commonwealth’s primary witness relied upon the GP Atlas to calculate the probability of attaining skeletal maturity prior to the age of 18

In contrast to Professor Cole, Dr Vincent Hock Seng Low, the witness most commonly called by the Commonwealth Director of Public Prosecutions in people smuggling matters where age is in dispute, has used the GP Atlas as the basis for developing a probability of a male person who has a mature x-ray being under the age of 18 years. Dr V. Low is a Consultant Radiologist at the Insight Clinical Imaging Group, Western Australia and former Head of the Radiology Department and Consultant Radiologist at the Sir Charles Gairdner Hospital in Nedlands, Western Australia.

The import of Dr V. Low ’s analysis of wrist x-rays in people smuggling matters where age was in dispute can be seen in the Joint Commonwealth submission provided to the Inquiry:

Based on expert advice the Commonwealth has sought, the wrist X-ray procedure can assist in determining whether a person is 19 years or older as male wrist skeletal maturation occurs on average at that age. The CDPP will only rely upon wrist X-rays in circumstances where the wrist X-ray indicates that skeletal maturity has been reached and an expert radiologist states that there is the highest level of probability that the person is an adult. Accordingly, only those cases where there is the highest probability that the defendant was 18 years or older at the time of the offence are brought before the courts.^[95]

Dr V. Low is of the opinion that:

In males, skeletal maturity at the hand is reached at approximately 19 years of age. This means that at this point in time, all the growth plates have fused.^[96]

To calculate the probability that a male person showing skeletal maturity is of a particular chronological age or younger, Dr V. Low uses the standard deviation for the age of 17 years identified in Table V of the GP Atlas. Dr V. Low starts by making an assumption that skeletal maturity is attained on average at 19 years of age. Using the larger of the standard deviations provided in the GP Atlas for the age of 17 years, he then extrapolates to conclude that there is a 21.79% probability that a person with a mature wrist is 18 years or less.^[97]

Dr V. Low ’s evidence on this topic did not attract support from any witness at the public hearing for medical witnesses conducted as part of this Inquiry; nor has support for it been found in any submission made to the Inquiry by a medical expert.

For example, Dr James Christie, a Specialist Radiologist at the Children’s Hospital at Westmead, New South Wales, explained why the assumptions on which Dr V. Low bases his calculations are wrong in the following way:

The use of such precise numbers suggests to the reader of the report that there is scientific accuracy to the estimate, where none exists. It is scientifically wrong to suggest that a standard variation curve can be applied to the end point of a population distribution, in this case the 19 year standard. For example both a 17 year old boy or a 60 year old man may both have the same skeletal age of 19. By assessing only the skeletal age, and using Dr V. Low ’s assumptions, you could come to the clearly false conclusion that each of them has exactly the same probability of being 18 years old.

The skeletal age value assigned to each standard in the [GP Atlas] is not the average or even median age at which the skeleton reaches this appearance, but merely the x-ray that the authors subjectively felt most closely represented that age. Clearly the 19 year value cannot be an average of the values between 15 years and 60 years. It also is not the median age for maturity, (that is the value where 50% will be above and 50% will be below), but is merely a descriptor that is higher than the 18 year value. [Greulich and Pyle] could have called this 20 or 30 or just Mature. The actual number means very little.^[98]

Professor Cole, who unlike Dr V. Low is an expert bio-statistician, expressed the opinion that Dr V. Low ’s conclusions based on the GP Atlas are wrong and should be dismissed.^[99] As noted above, Professor Cole stresses that the authors of the GP Atlas were not concerned with identifying the average age at which skeletal maturity is attained.

Professor Cole argues that the real question is: what is the average age at which a person might have a mature wrist x-ray? In his written submission, Professor Cole asserted:

This leads to the following questions: what proportion are mature at younger ages than 19, and what is the youngest age that adult x-rays are seen? These questions relate to the age of attainment of a subject’s mature x-ray, which is quite distinct from their current age. Since most x-rays are mature by age 19, the age of attainment must for most subjects be earlier than 19.^[100]

(c) Alternative analysis demonstrates that wrist x-rays are insufficiently informative of whether someone has reached 18 years of age

Professor Cole suggests that it is possible to calculate the distribution of the age of attainment of skeletal maturity in males by reference to a more recent publication, the TW3 bone age manual published in 2001. This manual addresses directly the question of the age at which skeletal maturity is attained.^[101] Professor Cole has calculated that the average chronological age at which a male achieves skeletal maturity is 17.6 years; that is, in Professor Cole’s opinion, on average males achieve skeletal maturity earlier that the age of 18 years.

Professor Cole’s written submission illustrates the basis of his calculation:

The distribution of the age of attainment of skeletal maturity can be estimated from the three centiles in Table 8 of TW3, reproduced here.

Assuming a normal distribution of the mean is about 17.6 years and SD 16.5 months. From this the probability of attaining maturity before age 18 is about 61%.^[102]

On Professor Cole’s analysis, there is a 61% statistical probability that a young male will have a mature wrist on his 18^th birthday. In other words a mature wrist x-ray is not informative of whether a young male is over the age of 18 years as more than half of all young men will have achieved skeletal maturity before that age.

Professor Cole further suggests that it is important to look beyond a simple probability figure to the relevant ‘likelihood ratio’. Professor Cole gave evidence that:

One might think that the 61% probability of being mature before age 18 is what should interest the court. In a sense it is, but more generally the court wants to decide which of the two alternative scenarios – the individual being either over 18 or under 18 – is better supported by the evidence. For this the court needs to compare two different probabilities – that of being over 18 with a mature x-ray versus that of being under 18 with a mature x-ray. Ideally the probability should be close to 100% over 18 and close to 0% under 18, and the ratio of the two probabilities is a concise summary of the evidential value of the x-ray. This ratio is known as the likelihood ratio (LR), and the further it is from 1 then the more informative the x-ray is.^[103]

Professor Cole calculates the relevant likelihood ratio as 2.64 and explains that:

To put this in context, [a likelihood ratio] of less than 5–10 in medical decision-making is viewed as weak – the degree of misclassification is too high. Here the [likelihood ratio] is well below 5, a cogent argument that the evidential value of the mature x-ray is poor. If relied on it would lead to too many minors being incorrectly assessed as adult.^[104]

In other words, by this additional calculation, Professor Cole seeks to show that, statistically speaking, a mature wrist x-ray is not informative of whether a person has reached 18 years of age.

(d) Other expert opinion confirms that wrist x-ray analysis is uninformative of whether a male has reached 18 years of age

Other material before the Commission provides support for the conclusion that, on average, males attain skeletal maturity either before, or at about the time, that they attain the age of 18 years.

Dr Ella Onikul, Director of Medical Imaging at the Children’s Hospital at Westmead, New South Wales, gave oral evidence to the Inquiry on behalf of the Royal Australian and New Zealand College of Radiologists (RANZCR). She expressed her support for the opinion of Professor Cole that the statistical probability of a male attaining a mature x-ray before the age of 18 years is 61%.^[105] Dr Christie, in a written submission, similarly supported the opinion of Professor Cole.^[106]

Professor Sir Al Aynsley-Green, a paediatric endocrinologist and Professor Emeritus of Child Health, University College, London, has shown that the method devised by Tanner and Whitehouse in 1962 to assess skeletal maturity (the TW2 method) indicates that 50% of boys will have achieved ‘adult’ appearances by the age of 17 years, 50% having yet to reach that stage; but some 10% of normal boys will have yet to achieve full maturity at the age of 19.5 years whilst 3% will have achieved this at the age of 16 years.^[107]

Additional support for a conclusion that, on average, males achieve skeletal maturity at, or earlier than, 18 years of age is found in a short communication from 2006 published in Forensic Science International: ‘[t]he skeletal development of hand bones is complete at the age of 17 years in girls and at the age of 18 years in boys’.^[108] This statement is inconsistent with males achieving skeletal maturity on average at 19 years of age, as posited by Dr V. Low .

As illustrated above, the evidence before this Inquiry strongly favours the conclusion that on average males achieve skeletal maturity at, or slightly before, the 18^th anniversary of their birth. This conclusion is of itself sufficient to render medical opinion evidence based on an assessment of skeletal maturity as shown by a wrist x-ray unhelpful for the purpose of determining whether a particular male person is over the chronological age of 18 years.

In other words, on Dr V. Low ’s approach, a mature wrist x-ray can be relied upon as evidence that the subject was likely to be at least 18 and therefore an adult at the time of the offence. The evidence before the Inquiry, however, strongly suggests that at least 50% of males attain skeletal maturity at or before the age of 18, leading to a very real risk, on Dr V. Low ’s approach, of minors being incorrectly assessed to be adults.

There are, however, other problems in relying on a wrist x-ray which shows skeletal maturity as evidence that its subject is over the age of 18 years. The first of these is that there is a wide normal variation in the age at which young people generally achieve skeletal maturity.

The evidence is overwhelming that using skeletal age to assess chronological age is an imprecise technique. Young people develop skeletally at different rates and reach skeletal maturity at varying ages. The mean and standard deviation figures contained in Table III and V of the GP Atlas referred to in section 2.2(a) above are one measure of this variation.

The authors of the GP Atlas themselves observed that ‘[i]n the study of any aspect of the growth and development of children one is constantly bedevilled by their variability’.^[109] They state:

The system described in this Atlas ... is intended to provide merely useful estimates of skeletal status – and will do so, if it is properly used. Unfortunately, as in many other similar procedures, there is a tendency to attribute to and to expect from it a greater degree of precision than intended by those who designed it or, indeed, than is permitted by the nature of the changes about which it is designed to measure.^[110]

Expert opinion evidence has been given in Australian criminal cases which calculates to two decimal points the probability that a person showing skeletal maturity is a particular chronological age.^[111] This level of precision finds no support in the GP Atlas and tends to suggest (wrongly) that the calculation has a high level of scientific accuracy.

Australia’s Chief Scientist, Professor Ian Chubb AC, in a brief enclosed with a letter dated 11 January 2012 addressed to AGD advised:

Radiological based determination of skeletal maturity does not allow for a precise determination of chronological age. Outcomes of radiological assessments of bones vary with ethnicity and socio-economic conditions (nutrition and disease status). There is observed variation in skeletal maturity of 2 years within each gender.^[112]

The majority of medical experts who gave evidence to the Inquiry confirmed that there is significant variation in the rate at which normal children develop. For example, Professor Cole gave evidence that:

developmental age in general, and bone age in particular, is only weakly linked to an individual’s chronological age. The age of puberty based on bone age has a standard deviation (SD) of over 15 months. So the range of chronological ages seen in 95% of the population at this developmental stage extends over 5 years (±2 SDs), and two boys at this stage could be 5 years apart in chronological age. The same observation applies to other developmental markers such as age at peak height velocity or dental age.^[113]

Furthermore, a number of Australasian medical professional associations made a joint submission to the Inquiry in which it is stated:

Variations in the tempo of physical maturation have long been noted. For instance some children enter puberty before 10 years of age while others are not in puberty until late teenage years. ... Pubertal variation has a major impact on bone age estimation. ... Like most biological variables puberty occurs in a ‘bell-shaped’ curve. ... The tempo of puberty is usually similar among individuals and takes 3 to 4 years to complete. Thus there is approximately a 4 to 5 year range for normal puberty to start in both genders and, as expected, a similar difference in bone age estimations is also observed during this time.^[114]

It may therefore be concluded that wrist x-rays can never provide the basis for an accurate prediction of chronological age. This is because, as the authors of the GP Atlas themselves observed, there is substantial variability in the growth and development of children.

It is no doubt because of the inability of skeletal development to define age precisely that the Study Group of Forensic Age Estimation of the German Association of Forensic Medicine, in its publication Guidelines for Age Estimation in Living Individuals in Criminal Proceedings, specifies that:

The expert report has to quote the reference studies on which the age estimation is based. For each feature assessed, the report must state the most probable age including the range of scatter of the reference population. What must also be noted is that this range may increase further by an empirical observer’s error.

The age-related variations resulting from application of the reference studies in an individual case such as different genetic/geographic origin, different socioeconomic status and their potential effect on the developmental status ... as well as diseases that might affect the development of the individual examined must be discussed in the report including their effect on the estimated age. If possible, a quantitative assessment of any such effect should be given.^[115] (citations omitted, emphases added)

The ‘improved process for age determination in people smuggling matters’ jointly announced by the then Attorney-General and the then Minister for Home Affairs and Justice in July 2011 was purportedly based on this publication from the year 2000.^[116] It may be observed, however, that no expert report relied upon in age determination proceedings regarding an individual suspected of a people smuggling offence complied, or even substantially complied, with the above guidelines.

It can be concluded from the evidence before this Inquiry that the normal variation in the chronological age at which males generally achieve skeletal maturity is not less than two years and is possibly as high as five years.

This conclusion is an important one. This is because, as noted above, any assessment of chronological age which is based on skeletal age as shown by a wrist x-ray ceases to be informative upon the attainment of skeletal maturity; the individual’s wrist x-ray will remain unchanged thereafter. Even if it be assumed, contrary to the overwhelming weight of the evidence, that skeletal maturity is achieved by males on average at the age of 19 years, normal variation of two years means that no conclusion can safely be drawn from skeletal maturity that a particular male person is not under the age of 18 years. Normal variation of even two years renders his skeletal maturity entirely consistent with his being between 17 and 18 years of age.

2.3 The GP Atlas is of limited use for assessing the chronological age of populations dissimilar to that of the study sample

Another problem associated with using wrist x-rays as evidence of chronological age is that the population in the study sample used in the GP Atlas differs from the population from which the young Indonesians come.

The standard plates in the GP Atlas were selected from x-ray films of healthy, white children of North European ancestry in the United States whose families were somewhat above the average in economic and educational status.^[117] The authors of the GP Atlas observe that ‘[t]hese standards can be expected to fit reasonably well with other children of comparable genetic and environmental background’.^[118] They explain, however, that variability is especially marked in a ‘country whose people are as heterogeneous in national and even in racial antecedents as those of many parts of the United States’.^[119] They observe that, as a result, one usually cannot apply findings based on studies on children in one section of the country to children of another without some modification.^[120] They note that one should be reluctant to attribute the relative skeletal retardation of children in some other parts of the world to racial differences as illness and deprivation might provide a more likely explanation.^[121] They illustrate this point by reference to a study of American-born Japanese children living in California.^[122] The boys in that study were found to be significantly more advanced skeletally than Caucasian children from Cleveland at 13, 14, 15, 16 and 17 years of age.

On this issue, the Joint Commonwealth submission notes that:

The CDPP has been advised that while there is racial difference in skeletal size, there is no change across races in skeletal development. Accordingly, the Greulich and Pyle Atlas can be applied as a standard for making forensic age determinations in ethnic groups that differ from the reference population.^[123]

In late 2010 a senior officer of the Office of the CDPP identified two papers that argued that GP Atlas is valid for use across racial and ethnic groups.^[124] In addition, Dr V. Low provided the Office of the CDPP with a report on this issue in May 2011. Dr V. Low ’s report referred to the studies comparing Japanese and American children from the 1950s, studies from the 1990s which he suggested indicate that the GP Atlas can be used with confidence for young Thai people, and a meta-analysis conducted by Schmeling which found that socio-economic status rather than ethnicity is likely to affect the rate of skeletal maturation.^[125]

However, other studies confirm that certain populations of children may develop at different rates from the study sample upon which the GP Atlas is based. For example, an appraisal of the GP Atlas for skeletal assessment in Pakistan found significant differences between skeletal age assessed by the GP Atlas and chronological age in a subset of Pakistani children. In Pakistani males, skeletal age was advanced during early childhood, delayed during middle and late childhood, and again, advanced during adolescence.^[126]

A study has also been undertaken to determine if variation between bone age and chronological age exists in children of different ethnic groups within the United States population. It concluded that the standards of the GP Atlas to determine bone age must be used with reservation, particularly in Asian and Hispanic boys in late childhood and adolescence, when bone age may exceed chronological age by nine months to 11 months 15 days.^[127] Other studies have revealed racial differences in Middle Eastern, African and African American populations with bone age disparities between six to 12 months depending on when the children were assessed.^[128]

Dr V. Low has drawn to the attention of the Commission a study published in 2008 entitled ‘Skeletal Maturation in Indonesian and White Children Assessed with Hand-Wrist and Cervical Vertebrae Methods’.^[129] It does not appear that this study assessed skeletal maturity by reference to the GP Atlas, and it only studied boys between the ages of 10 to 17 years. The Indonesian children the subject of the study were school children from Jakarta and its vicinity, and the radiographs were taken during routine orthodontic examinations. Data for the white children came from pre-existing radiographs taken as part of routine orthodontic examinations at the University of Rochester Medical Centre, Rochester, New York, USA.^[130] It may therefore be assumed that both groups probably came from relatively affluent families. The study showed that during the pubertal growth spurt period there were differences between the Indonesian and white children with the white children reaching each skeletal maturity index stage earlier than the Indonesians. However, at ‘about the age of 17 years, the Indonesian children appeared to equalize with the white ones’.^[131] The study thus seems of some, albeit limited, relevance for present purposes. It appears that no other bone studies have been done of an Indonesian population.

Additionally, it may be the case, as Dr Paul Hofman, President of the Australasian Paediatric Endocrinology Group, speculated in his evidence to the Inquiry, that the population of individuals with which this Inquiry is concerned may vary genetically from the Indonesian male population generally. Dr Hofman spoke of the possibility that a person who was seeking to recruit a sailor to crew a boat intended to be used for people smuggling would be more likely to choose a person who appeared mature, even though young, so that the recruiter would have, intellectually and emotionally, a young person who could do more work.^[132]

While this possibility, as Dr Hofman appropriately acknowledged, is purely speculative, it draws attention to the complexities involved in seeking to determine whether an individual suspected of a people-smuggling offence who says that he is a child comes from a population that is relevantly comparable with the study sample on which the GP Atlas is based.

The differences in opinion outlined above indicate that there is no conclusive evidence about the degree to which any relevant genetic difference between the population on which the GP Atlas is based and the population with which this Inquiry is concerned might impact on the accuracy of age assessments. As Professor Cole put it at the medical hearing:

I think the important point to make is that although ... it is known that socio-economic status, ethnic group and nutritional status may all impact on the bone age, it’s very hard to actually come up with a number that one can use to make any sort of adjustment. All one can do really is wave one’s hand in the air and say whatever certainty we had based on a European population, that certainty decreased substantially when we moved to talking about an Indonesian fisherman population.^[133]

2.4 Errors of interpretation may impact on the accuracy of wrist x-ray analysis of skeletal age

Another unquantifiable factor which may impact on the reliability of an age assessment based on a wrist x-ray is the subjective nature of the assessment of an x-ray.

The authors of the GP Atlas draw attention to the subjective nature of an assessment of a wrist x-ray. They also point out the need for practice before one becomes efficient in duplicating by subsequent independent assessments one’s previous estimates.^[134]

The joint submission to the Inquiry by the Australasian medical professional associations states:

All bone age estimation methods have error involved. In other words if the same X-ray is assessed either by the same or different assessors the assigned bone age may vary. Intra-observer error refers to the variation of one clinician’s assessment while inter-observer error refers to the variation between different clinicians. ... A number of studies have investigated these effects and in summary have demonstrated an average intra-observer error of between 2 and 9 months and an average inter-observer error between 1 and 12 months. However, these were average errors and the error range in these studies was 0 to over 2 years. Combining both the intra- and inter-observer, variation differences of over 12 months frequently occur.^[135] (Citation omitted)

The Guidelines for Age Estimation in Living Individuals in Criminal Proceedings referred to above include two guidelines explicitly concerned with quality assurance.^[136] The first requires annual ring experiments for continual quality assurance. A ‘ring experiment’ in this context apparently involves a number of experts undertaking the same assessment for purpose of comparing the results and thereby enhancing their skills. The second guideline is that an expert may request an evaluation of an age estimation before his or her report is written.

The expert reports that are before this Inquiry demonstrate variations of assessment between different radiologists. In at least eight cases, Dr V. Low assessed a wrist x-ray as showing a mature wrist when another radiologist considered that the wrist was not yet mature.^[137]

2.5 Wrist x-ray analysis could be relied upon to conclude that an individual with an immature wrist is under the age of 18 years

As discussed above, wrist x-ray analysis does not assist in determining whether a male young person is over 18 years of age. This is because on average males attain skeletal maturity at or just before the age of 18 years.

However, several submissions to the Inquiry argued that wrist x-ray analysis should be able to be used to show that a person is under 18 years of age. For example, Legal Aid Queensland submitted that:

the use of the technique should not be abandoned completely. Wrist x-rays have in the past been able to facilitate the speedy repatriation of children to Indonesia in cases where their x-ray’s showed clear skeletal immaturity. For this reason the wrist x-ray technique may still have a valuable [role] in determining the status of a person as a child. However, they should ... not be relied upon as sufficient by themselves to establish an individual as being an adult.^[138]

Victoria Legal Aid made a similar argument:

VLA does not support abandoning the use of x-ray techniques altogether. The problem with prohibiting the use of x-ray analysis is that we know that x-rays are sometimes a quick and effective mechanism to determine the veracity of a child’s claim about their youth. Children have been appropriately and quickly returned to Indonesia in such situations. To avoid children being unnecessarily detained, a preferable position to outlawing the use of x-ray analysis completely would be to prohibit the reliance on such evidence in isolation should the analysis point to the claimant being an adult.^[139]

Because a male person will attain a mature skeleton, on average, at or just before the age of 18 years, it is statistically probable that a male person who has an immature wrist is a minor. For this reason and applying the principle of the benefit of the doubt, it may be concluded, subject to the ethical considerations discussed below, that the practice generally followed by the Commonwealth of discontinuing the investigation and prosecution of any individual whose x-ray shows an immature wrist should continue.

3 The use of dental x-ray analysis for the assessment of chronological age

In July 2011, the Australian Government announced an ‘improved process for age determination in people smuggling matters’ which was to include offering voluntary dental x-rays to individuals suspected of people smuggling whose age was in doubt.^[140] In addition, at about this time the working group of Commonwealth agencies recommended to the then Attorney-General that dental x-rays be prescribed in the Crimes Regulations 1990 (Cth) so as to make both wrist x-rays and dental x-rays prescribed procedures for the purpose of s 3ZQB of the Crimes Act.^[141] The Joint Commonwealth submission notes that:

The Commonwealth is considering adding dental X-rays as a prescribed procedure in the Crimes Regulations, which would allow investigating officials to seek an order from a court to conduct a dental X-ray.^[142]

If this were to occur, dental x-ray analysis could be used for age assessment purposes in the same circumstances in which wrist x-ray analysis can now be used. The former Attorney-General has said that, if prescribed, dental x-rays would supplement wrist x-rays in all matters where age is in dispute.^[143]

Several dental forensic experts have given evidence to the Inquiry that, while a dental x-ray analysis cannot provide a fully accurate age assessment, it is more informative than any other method for assessing whether a person has reached the age of 18 years.^[144] Others making submissions to the Inquiry have suggested that dental x-rays suffer from the same weaknesses as wrist x-rays.^[145]

The question is whether dental x-rays are an appropriate substitute for, or supplement to, wrist x-rays or, indeed, any other method of assessing chronological age. It appears that, while dental x-ray analysis may be more informative of chronological age than wrist x-ray analysis, it suffers from many of the same issues that make wrist x-ray analysis insufficiently informative of chronological age to be appropriate for use within the context of a criminal prosecution.

Further, there is no evidence that a dental x-ray together with a wrist x-ray will provide a more accurate assessment of age than either one of these methods alone.

This section of the report will discuss the process of taking and analysing a dental x-ray. It will then consider the following questions:

• Is dental x-ray analysis informative of whether a person has attained 18 years of age?
• How wide is the normal variation in the age at which young people generally achieve dental maturity?
• Does ethnicity, socio-economic status and nutrition impact on the attainment of dental maturity?

3.1 The process for taking and analysing dental x-rays for the purposes of age assessment

There are several different types of dental examination that could be used to assess the age of persons in different age ranges. This Inquiry has focussed only on those methods that might be useful for the purpose of determining whether an individual is under, or alternatively over, 18 years of age.

There appears to be agreement that the only viable examination method for this purpose is to take an OrthoPantomoGraphic (OPG) x-ray to analyse the development of the third permanent molar (wisdom tooth).^[146] Essentially this means taking an x-ray by a procedure that is wholly external to the mouth (as opposed to placing x-ray films inside the mouth), to determine whether the person has a wisdom tooth that has reached a certain stage of maturity.

Unlike wrist x-rays, there appear to be some dental databases expressly developed for the purposes of assessing chronological age (as opposed to skeletal age). The most widely used of these databases is the Demirjian dental age assessment scale, which was developed in 1973 based on a French-Canadian study sample.^[147] In the context of determining adulthood, a forensic odontologist needs to assess whether the OPG indicates that the wisdom teeth have reached ‘Stage H’ on the Demirjian scale.^[148]

3.2 Experts disagree on how informative dental x-ray analysis is of whether a person has attained 18 years of age

Some submissions to the Inquiry asserted that dental x-rays are sufficiently accurate to be useful to a court making a determination of age on the balance of probabilities. For example, the Australian Society of Forensic Odontology cites recent research on the use of third molar development which finds ‘analysis of third molar development to be accurate and sufficiently correlated with chronological age to be of forensic value’.^[149]

However, Dr Anthony Hill, the President of this Society, gave evidence that a dental x-ray will only be able to tell whether a person is 18 years old ‘plus or minus 1.2 years’. He said:

Despite the incredible amount of robust research that has been done and the investigation and scientific research that has been undertaken over the years, looking at and focusing on the development and the maturation of the wisdom tooth, we are unable to state definitively the age of the individual. In other words we can’t say this person’s 18, we can’t say this person’s 18.5, we can’t say if this person’s 19 at all. We are not able to do that. But what we are able to do with ongoing research and ongoing science is that we are able to reduce the parameters of our confidence intervals and our standard deviations to the point where we are comfortable with stating that a person is 18 years of age plus or minus 1.2 years.^[150]

Similarly, forensic odontologist, Dr Stephen Knott, gave evidence that:

It is not possible to give a specific age of 18 years, the age assessment would need to include a confidence interval.^[151]

Dr Hill further stated:

I don’t use confidence levels because basically what we’re saying here is that in 61% of the cases we’ve got it right, but in 39% of the cases we’ve got it wrong. I don’t want to be sitting in the 39% of cases where I’m wrong and I’ve offered this opinion. I mean that’s not really good odds. What I’m saying is that we should not be talking about these confidence intervals, we should be talking about this person is 18 plus or minus.^[152]

The Inquiry has received two submissions which attempt to conduct a statistical analysis on the reliability of dental age assessments at the 18 year threshold; one from Professor Cole from University College, London (who also provided an analysis in the context of wrist x-rays), and the other from Professor Graham Roberts from the King’s College London Dental Institute.^[153] The two methodologies lead to different results.

Professor Cole uses recently gathered data on over 2600 individuals attending the Eastman Dental Hospital in London. Their teeth were x-rayed and staged using the Demirjian classification and information on their four third molars was extracted. The data was collected by a PhD student being supervised by Professor Roberts. From this dataset, Professor Cole calculates the mean age of attainment of one or more Stage H third molars to be 19.6 years with a standard deviation of 1.3 years. Using these figures he calculates that there is a 24% chance of a person having a Stage H wisdom tooth before age 18.^[154]

Professor Cole acknowledges that this is a better probability result than with the wrist x-rays, but he goes on to calculate a likelihood ratio which suggests that the result is still highly uninformative. He stated:

one should rely on the [likelihood ratio] to make the judgement. In the presence of a mature third molar, comparing the probabilities of being under 18 and over 18 gives [a likelihood ratio] of just 5.2 ... which is only marginally informative. ... Thus dental age suffers from the same lack of precision as bone age for forensic age assessment. Ages older than 16 cannot convincingly be excluded using either method.^[155]

In the most recent figures provided by Professor Roberts (using an updated version of the dataset used by Professor Cole), he has calculated the mean age of attainment of Stage H of the Lower Left third molar to be 20.7 years with a standard deviation of 2.3 years. From these figures he calculates that there is a 12% chance of having a Lower Left Stage H wisdom tooth before the age of 18.^[156]

Professor Roberts and Professor Cole have engaged in a detailed exchange of views regarding their respective results and statistical methodologies, both in the context of this Inquiry^[157] and, in 2008–09, in the British Dental Journal.^[158] Briefly summarised, Professor Roberts suggests that Professor Cole has neither used the most appropriate dataset nor applied the most appropriate statistical methodology to determine his statistical probabilities.^[159] Professor Cole, in turn, suggests that Professor Roberts has not considered the appropriate data and that his statistical methodology is wrong.^[160]

The debate between Professor Cole and Professor Roberts over the statistical accuracy of dental x-rays has not been subjected to the same scrutiny as the debate between Professor Cole and Dr V. Low in relation to wrist x-rays. Consequently, there is insufficient additional information before the Commission for a conclusion to be reached as to the relative accuracy of the calculations of these two acknowledged experts.

It may, nonetheless, be concluded that neither of the figures advanced by Professor Cole and Professor Roberts is particularly informative as to whether an individual is, or is not, over 18 years of age. Within any population, it appears that either 24%, or alternatively 12%, of the population would be incorrectly assessed as being an adult if reliance were placed solely on whether he had attained dental maturity. The difficulties inherent in relying on statistical evidence as evidence concerning a particular individual is examined in Appendix 5.

3.3 There is a wide normal variation in the age at which young people generally reach dental maturity

As with skeletal maturity, there appears to be a wide normal variation in the age at which young people generally reach dental maturity.

For example, the Australian Society of Forensic Odontology gave evidence that wisdom teeth can start developing ‘from mid-teens to early 20s’.^[161] This suggests a wide range of ages over which dental maturity is normally achieved. The Society goes on to state that ‘[p]recise determination of age is not possible due to human variation; an age range, with confidence intervals is the best expression of age estimation’.^[162]

The Victorian Institute of Forensic Medicine confirms that:

biological variation in human development means than any assessment of age based on analysis of anatomical growth markers is only an estimate, and thus will contain a degree of error.^[163]

Similarly, The George Institute for Global Health states more generally that:

Caution should be exercised when using methods of biological age assessment (i.e. skeletal age, dental age or bio markers) to determine chronological age in adolescents. The range of biological ages within a chronological age group may exceed 2 years even with the most accurate of these methods.^[164]

3.4 Ethnicity, socio-economic status and nutrition may impact on the attainment of dental maturity

The Demirjian scale is based on a study sample of French-Canadian girls and boys. This raises the question of whether assessing the teeth of Indonesian boys against that age assessment scale might be subject to inherent error based on differing ethnicity, socio-economic status and nutrition.

There appears to be substantial scientific debate about the extent to which the development of wisdom teeth is affected by these factors. Some studies suggest that the variation in age estimates between different groups are quite significant and that population-specific standards would increase accuracy.^[165] Others suggest that there are differences but they are insignificant.^[166] Some studies suggest that third molar mineralisation is standard across population groups.^[167] However, it appears that at least some of these studies focus on age groups younger or older than 18 years of age.^[168] It is therefore not clear how directly applicable this research may be at the age of 18 years.

However, there seems to be a consensus among those experts who gave evidence to the Inquiry that there are differences between different population groups. Any controversy seems to concern the significance of the impact of those differences. For example, the Victorian Institute of Forensic Medicine suggested that ethnicity is not a big concern in the context of wisdom teeth, stating:

To the extent this has been researched, there appears to be some robustness in the development of the third molar between different populations.^[169]

Dr Hill stated:

There are subtle differences but they are of very little consequence to what we are doing in our age assessments.^[170]

Dr Knott also submitted that there may be variations between ethnic groups but said that there would need to be extreme medical problems for there to be a significant effect. His evidence was that:

Biological development will vary slightly between sexes, ethnic groups and an individual’s overall health. ... Variation in chronological age assessment due to medical conditions are minimal. The individual would need to be suffering from a noticeable extreme medical problem to have any significant effect, eg: starvation, syndromal variations, severe Vitamin D deficiency, etc.^[171]

On the other hand, The George Institute for Global Health believes that variations in development on the basis of ethnicity could be quite substantial. Its submission stated:

A recent systematic review of third molar age estimation in various American population groups reinforced a requirement for the use of population specific studies when estimating age from dental x-rays. Within a number of noted American populations, varying rates of third molar development were seen. The review concluded that undoubtedly, additional and larger population specific studies are needed.^[172] (Citation omitted)

3.5 There is no evidence that a wrist x-ray plus a dental x-ray improves reliability in the assessment of age

Given the Australian Government’s announced intention to specify both wrist and dental x-rays as prescribed procedures for the purposes of s 3ZQB of the Crimes Act, it is appropriate to consider whether there is any evidence that the accuracy of age assessments will be significantly enhanced by the taking of both wrist x-rays and dental x-rays.

The Commission is aware of some expert medical opinion that using a combined method of physical examination as well as wrist x-rays, dental x-rays and clavicle x-rays can increase diagnostic accuracy. However, the authors of the report in which this opinion is expressed state that:

For each examined feature, the report must indicate the most probable age and the range of scatter of the reference population. Furthermore, it should be noted that the range of tolerance may be increased by an empirical observer error. The age-relevant variations resulting from the application of the reference studies in an individual case such as deviating genetic/geographic origin, different socio-economic status, and with that a possible difference degree of acceleration, developmental disorders of the individual, have to be discussed in the report including their effect on the estimated age and, if possible, a quantitative assessment of any such effect should be given.^[173] (Citations omitted)

The Commission is not aware of any evidence which demonstrates that combining dental and wrist x-ray analyses leads to a more reliable indication of age than one or other of these analyses alone. Further, the radiation exposure involved in two x-ray procedures raises even greater ethical concerns than a single x-ray procedure.

4 The ethical implications of the use of x-ray analysis for the assessment of chronological age

The radiation exposure associated with x-rays gives rise to important ethical issues.

As noted in a recent publication regarding age assessment of individuals subject to immigration control in the United Kingdom:

In considering the ethics of radiography it is necessary to weigh up the actual or potential benefits of radiography with the potential damage that might be caused to a group of children and young people who are potentially vulnerable as a consequence not only of their age but also their background and experiences.^[174]

There is no doubt that the amount of radiation exposure involved in wrist or dental x-rays is low. The Australian Radiation Protection and Nuclear Safety Agency (ARPANSA) states that the radiation exposure associated with wrist and dental x-rays is minimal and that there is wide acceptance in the scientific community that the radiation dosage from a wrist x-ray poses negligible risks to a person’s health. ARPANSA estimates that the radiation dose received from a single x-ray examination of the hand is approximately 0.01 millisievert (mSv), a dosage that the Joint Commonwealth submission notes is comparable to flying from Darwin to Singapore.^[175] Further, the Australian Society of Forensic Odontology provides a table in its submission suggesting that a dental OPG exposes a patient to 0.02 mSv, which the submission shows is equivalent to three days background radiation. This exposure level is classed ‘negligible’ under the category of ‘additional lifetime risk of fatal cancer from examination’.^[176] Nonetheless, there is wide agreement that all deliberate exposures to radiation should be justified and subject to control.

The International Atomic Energy Agency (IAEA) publication, Radiation Protection and Safety of Radiation Sources: International Basic Safety Standards – Interim Edition (General Safety Requirements: Part 3) published in 2011 includes a number of general requirements for protection and safety. Requirement 10 is concerned with ‘Justification of Practice’. The opening statement under this requirement is: ‘The government or the regulatory body shall ensure that only justified practices are authorized’.^[177] Paragraphs 3.16 and 3.18 of this publication, which form part of Requirement 10, state:

3.16 The government or the regulatory body, as appropriate, shall ensure that provision is made for the justification of any type of practice and for review of the justification, as necessary, and shall ensure that only justified practices are authorized. ...

3.18 Human imaging using radiation that is performed for occupational, legal or health insurance purposes, and is undertaken without reference to clinical indication, shall normally be deemed to be not justified. If, in exceptional circumstances, the government or the regulatory body decides that the justification of such human imaging for specific practices is to be considered, the requirements of paras 3.61–3.64 and 3.66 shall apply.

Paragraph 3.61 from this publication sets out the requirements of the justification process when radiation is to be used for a purpose other than for medical diagnosis or medical treatment or as part of a program of biomedical research. The paragraph calls for a justification process that includes the consideration of the benefits and detriments of implementing the type of human imaging procedure and the benefits and detriments of not implementing the type of human imaging procedure.

Further, in Australia, the Code of Practice adopted by ARPANSA requires medical practitioners to apply the As Low As Reasonably Achievable (ALARA) principle in any decision relating to radiographic imaging.^[178]
It is important to note that the ALARA principle requires that radiation be justified, regardless of the dosage of any radiological procedure. ARPANSA is clear that ‘the use of wrist and dental x-rays for age determination purposes must satisfy internationally accepted principles of radiation protection, in particular the principles of justification and optimisation’.^[179]

RANZCR gave the following evidence to the Inquiry:

One of the most important principles underpinning medical imaging is the ALARA Principle (As Low As Reasonabl[y] Achievable). This requires three factors to be met in the performance of imaging examinations, with a particular emphasis on those using ionising radiation – justification, optimisation and dose limitation. Justification requires any proposed imaging examination to yield a sufficient benefit to society to justify the risks incurred by the radiation exposure, and is based on the hypothesis that any radiation exposure, no matter how small, carries with it a certain level of risk. ...

Determining the risk of the level of radiation involved with a procedure must involve consideration of not just the amount of radiation involved, but also the clinical benefit of the procedure to the subject. If a procedure is without benefit and its application is not evidence based, any level of radiation exposure is considered unacceptable, no matter how ‘trivial’ the radiation dose may be considered.^[180]

The majority of medical experts who made submissions to the Inquiry argued that the use of x-ray analysis to assess age cannot be justified and therefore is unethical. For example, the joint submission of Australasian medical professional associations states that:

the accuracy of X-ray methods to determine age is not reliable and the benefit of their use is not proven; therefore such examinations are not consistent with the ALARA principle and cannot be justified.^[181]

In his submission, Dr Christie goes further and submits that he:

deplore[s] the use of a non-validated and unproven x-ray technique that exposes these young people to radiation for administrative purposes, without any clear opportunity to improve their health or lives.^[182]

Importantly, ARPANSA advised the Commonwealth that:

Current international best practice would require that any use of ionising radiation for the purpose of dental or wrist X-rays for age determination must be subject to a formal process of justification, to demonstrate that there is a net benefit from the exposure.^[183]

As discussed in Chapter 3, there is no evidence that such a formal justification process has been undertaken.

5 The use of other biomedical markers for the assessment of chronological age

The Commission is aware of a number of other processes that involve using biomedical markers for the assessment of chronological age, including physical examination, clavicle (collar bone) x-ray, magnetic resonance imaging, and ultrasound of the wrist or elbow. However, the Commission is not persuaded that any of these methods constitutes a more informative means of assessing age than the analysis of wrist x-rays.

5.1 Physical examinations suffer the same weaknesses as x-rays and raise serious ethical issues

Physical examinations require a medical practitioner – ideally a paediatrician – to assess height, weight, skin and visible signs of sexual maturity. The indicators of pubertal development in boys include penile and testicular development, pubic hair, axillary hair, beard growth and laryngeal prominence.^[184] These assessments would need to be compared to reference data to allow an approximate age to be calculated.^[185] Thus, while the examination would be external, it would involve the examination of genitalia.

The main advantage of this method is that it is relatively simple and does not require any radiation exposure. However, these are the only advantages of this method and they are outweighed by the disadvantages.

The disadvantages are, first, that these types of physical examinations suffer from the same inherent unreliability as x-rays and other biomarkers including:

• the wide variation in the rate of pubertal development between individuals within any given group
• the impact of ethnicity, socio-economic and nutritional background as well as illness
• the absence of current and culturally relevant reference sets.^[186]

Dr Knott submitted that visual age assessment is patently unreliable.^[187] Courts in the United Kingdom have also urged great caution in relying on paediatric reports, particularly where they purport to pinpoint an exact age.^[188]

Second, there has been considerable controversy in the United Kingdom over the appropriateness of including paediatric reports in official age assessment processes. In particular, there are concerns about the absence of rigorous protocols and the subjective nature of paediatric reports.^[189] One United Kingdom court decision held that a report from a paediatrician cannot generally ‘attract any greater weight than the observations of an experienced social worker’.^[190]

Third, physical assessment of a person’s genitalia raises ethical issues.^[191] Ethically, it is questionable whether such intrusiveness is justified for administrative purposes.^[192]

The Commission notes that the Australian Government has rejected physical assessment of genitalia as an age assessment technique. In June 2011, the then Attorney-General Robert McClelland advised the President of the Commission:

I do not consider paediatric examinations appropriate for the Commonwealth to utilise as part of the age determination process. This is because of the invasive nature of the procedure. The working group has investigated this procedure and reported that it involves examination of a person’s genitals and is likely to cause significant distress, fear, embarrassment and discomfort.^[193]

5.2 Assessing age using an x-ray of the clavicle requires further research

Some scientists have recommended an examination of the ‘ossification status of the medial epiphysis of the clavicle’ to estimate the age of people who are assumed to be older than 18 ‘because all other developmental systems under examination have completed their growth by this time’.^[194] Thus, this procedure may have some use for those individuals wishing to challenge an assessment of adulthood. The necessary examination would involve taking an x-ray or CT scan and there is some research suggesting that magnetic resonance imaging (MRI) or an ultrasound could be used in the future.^[195] However, it appears that further research is required before this method can be relied upon for age assessment.^[196]

5.3 Assessing age using magnetic resonance imaging requires further research

The Commission understands that assessing age through MRI has potential for future use, but that a significant amount of research needs to be conducted. The George Institute for Global Health’s submission to the Inquiry states:

The clinical use of MRI in the assessment of growth plate maturity is currently limited. There are some preliminary studies using MRI for the assessment of the extension of the growth plate; specific closure patterns of the normal physis around joints and physial arrest; however, at present none of these methods is widely used.^[197] (Citations omitted)

The George Institute for Global Health reports that the method has been used in the context of age assessment for the purposes of international sporting competition, but concludes that there is no evidence to support the use of this technology for age determination of young people below 14 years and above 17 years and that more research into the methodology is required.^[198]

5.4 Assessing age using an ultrasound of wrist and elbow requires further research

The George Institute for Global Health has also suggested that there is some potential for using ultrasounds of the wrist and elbow as a tool for age determination.

The George Institute for Global Health lists the main advantages of ultrasounds as including: that it is a radiation-free imaging technique; it is relatively inexpensive and widely available; it can be easily applied using portable systems; and that patient compliance is generally good. They note that disadvantages include: there is likely to be higher inter- and intra-rater error; difficulties in standardising documentation and imaging transfer; and very little data is available regarding age determination from the ultrasound of a wrist. They thus conclude that further validation is needed before this technique is preferred over any other method.^[199]

Professor Aynsley-Green also highlights that, but for the radiation exposure, the weaknesses of wrist x-rays will carry over to ultrasounds of the wrist.^[200]

6 The use of multifactorial medical approaches for the assessment of chronological age

Most submissions to the Inquiry openly acknowledge that there is no single reliable scientific method for determining a person’s age. However, some go on to suggest that a ‘multifactorial’ approach will provide more reliable assessments. A multifactorial approach involves employing a combination of medical age assessment processes.

Submissions from medical and forensic experts that supported a multifactorial approach included those from Associate Professor Daniel Franklin from the Centre for Forensic Science^[201] and the Victorian Institute of Forensic Medicine.^[202] Associate Professor Franklin argues that there ‘appears to be strong evidence showing that multifactorial techniques increase accuracy and help control for variation that may occur in any one single age indicator’.^[203] He also notes that ‘[t]here does not yet, however, appear to be any general consensus as to which methods should be combined, if they should be weighted, and how this can be achieved’.^[204]

The Joint Commonwealth submission states that the Commonwealth’s July 2011 decision to adopt a combination of different age determination procedures was based on academic literature suggesting that multiple procedures would ‘increase diagnostic accuracy’.^[205] However, it is important to note that the studies cited in the Joint Commonwealth submission, and several of the other medical experts, openly acknowledge that they are assuming an improvement in the quality of results. They are quite clear that there is no way of quantifying those improvements. For example, one publication cited in the Joint Commonwealth submission states:

for age diagnoses obtained with a combination of methods there is still no satisfactory way to scientifically determine the margin of error. ... If independent features are examined as part of an age diagnosis that combines several methods it may be assumed that the margin of error for the combined age diagnosis is smaller than that for each individual feature. However it has not yet been possible to quantify this reduction.^[206]

Similarly, the most recent study from the international interdisciplinary Study Group on Forensic Age Diagnostics (based in Germany), which has focused on the question of age assessments for criminal prosecutions, found that when age estimations from multiple sources are discussed critically in an individual case ‘it can generally be assumed that the range of scatter is reduced. So far, however, this reduction could only be estimated’.^[207]

The Commission has heard criticism of a multifactorial approach on ethical grounds. The more procedures that an individual is subject to, the greater their radiation exposure. As Professor Aynsley-Green has observed, there is risk associated with administering any amount of radiation.^[208] Further, there must be consideration of whether it is ethically justified to ask individuals to undertake more procedures when it has not been established that this will make a significant difference to obtaining an accurate assessment of chronological age.

The Commission is also concerned at the practicality and the cost involved in a multifactorial medical approach. For example, the Australian Society of Forensic Odontology has suggested that to achieve a more accurate age assessment from dental x-rays, a ‘panel of specialists experienced in the interpretation of both the OPG and dental development’ is required.^[209] This recommendation was explored further with Dr Hill at the hearing in the following exchange:

Dr Anthony Hill: I am proposing that the investigation should be headed up by a dentist simply because we are focusing on the dentition, teeth, and that is our field of expertise. I believe we should have radiographers with us on a panel, we should have paediatricians with us on a panel, I believe we should have orthodontists on that panel, I believe we should have other experts within this field, so that you are not going to get a skewed or biased dental opinion. You will get an opinion across the board from various experts within this field.

Catherine Branson: Dr Hill from what you’ve said I think you’re envisaging a panel of six or more expert medical practitioners or dental practitioners of one sort or another, is that right?

Dr Anthony Hill: Yes.

Catherine Branson: Have you turned your mind to what the cost to the public purse might be of using such a panel, with respect to every Indonesian national suspected of people smuggling who asserted that he was a child?

Dr Anthony Hill: We need to have one dentist, who is examining the client. We need one radiographer who will be taking one OPG. That OPG can be digitalised and can be sent around Australia. We do not need to convene a board of people; we do not need to convene anyone. This can all be done digitally and securely and that opinion can be sourced and can be arrived at within a matter of days, very simply. Then a report would be written by the forensic dentist, the forensic odontologist or whoever’s heading this panel and can be again sent to a commission, sent to a court, sent electronically.

But the cost to the community – well the cost to the community quite frankly, if we get it wrong, is far outweighed by what it’s going to cost us and what it’s going to cost us to set up the taking of an X-ray. Every state and every capital and every area in Australia where these people are housed, imprisoned, has radiographic facilities. So it’s not as though we are expecting extra funding to do this.^[210]

Finally, the Commission has heard robust criticism of this approach, with Professor Cole commenting that it is ‘pie in the sky’ to suggest that multiple assessment methods will provide better answers. He observed:

There’s a belief that if you collect extra information you can come up with a more precise estimate of age. I mean the first thing to say is that this is optimistic because many of the measurements that you will be making will be correlated with each other so they’ll all tend to give the same answer anyway, but the much more important point is that if you are going to combine information from lots of different directions then you have to operate to a very tightly developed and validated protocol so that you cut out personal bias in the way that individuals assess whatever particular marker they’re going to assess.

So I would argue (a) you’re not likely to improve precision by using a multidimensional assessment tool and the difficulty involved in setting it up, developing it and validating it would rule it out in practice in anyway. So I would say very bluntly that the idea that you can take lots of different measurements and come up with a better answer is pie in the sky.^[211]

It must be noted that if a conclusion is drawn that neither wrist nor dental x-ray analysis is sufficiently informative of chronological age to be of use in the context of criminal proceedings, then it is highly unlikely that in combination they will be appropriate as an age assessment method in this context. The Commission has seen no evidence that combining these procedures would lead to better results.

7 The use of a multi-disciplinary approach for the assessment of chronological age

Several submissions and some of the research papers in this area advocate for a ‘multi-disciplinary’ or ‘holistic’ approach to age assessment. While there is no consistent description of what this might mean, it appears that the proponents of this approach are suggesting a mix of medical and non-medical approaches to age assessment.

As well as medical procedures, a holistic approach might include conducting focused age interviews such as those that are discussed in detail in Chapter 5 below.

A number of submissions to the Inquiry supported a ‘holistic’ approach, including Professor Aynsley-Green who recommends that:

Multi-professional assessment – a ‘holistic’ approach – involving a team of social workers, educationists, paediatricians and psychologists working in specialised Age Assessment Referral Units or within the existing structures for child protection would seem to be a pragmatic way forward in order to obtain a consensus decision on age.^[212]

Several medical bodies also suggest a multi-disciplinary approach. RANZCR states that:

Whilst there is no single medical way to accurately determine an individual’s age, the government should consider developing a process where age is assessed in a number of ways; this is often referred to as ‘holistic’ age assessment. This approach incorporates narrative accounts, physical assessment of puberty and growth, and cognitive, behavioural and emotional assessments.^[213]

The Royal Australasian College of Physicians submitted that it would be appropriate to investigate:

comprehensive assessments which may include psychological, cognitive, developmental and cultural factors, as well as comprehensive efforts to source accurate documentation of age where it exists.^[214]

The Australian Society of Forensic Odontology argued that, even in the context of medical evidence, there should be a multi-disciplinary medical team. It advised:

Current thinking would suggest that age estimation is best practiced as a multi-disciplinary specialty, in that practitioners engaged should be familiar with the theory and practice of forensic anthropology, forensic odontology, medical imaging, human growth and development, and anatomy. To obtain the most accurate age estimates, it is evident that practitioners from different disciplines need to work together and reports should be written following consultation from a panel of experts who have examined all relevant data. This would maximise the accuracy of age estimations.^[215]

International commentary on age assessment also supports a holistic approach. For example the Separated Children in Europe Progamme’s Statement of Good Practice states that:

The [age assessment] procedure should be multi-disciplinary and undertaken by independent professionals with appropriate expertise and familiarity with the child’s ethnic and cultural background. They must balance physical, developmental, psychological, environmental and cultural factors.^[216]

However, while a broader approach to age assessment appears to be common sense, the Commission is not aware of any evidence that supports the view that a physical and psychosocial assessment will give any better result than either one of those methods alone. Individual differences in physical and social maturity will remain, and both approaches are vulnerable to the subjective views of the assessor.

Further, it is possible that in a ‘holistic’ assessment that includes both medical and non-medical age assessment techniques, preference will still be given to the results of medically-based processes due to a perception that they are scientifically based.

For these reasons, if a ‘holistic’ age assessment process is conducted, it is critical that a wide margin of benefit of the doubt is afforded to the individual whose age is being assessed. The importance of affording the benefit of the doubt when conducting focused age interviews is discussed further in Chapter 5 below.

8 Findings

8.1 Findings regarding wrist x-ray analysis for the assessment of chronological age

The overwhelming weight of the evidence before this Inquiry is that, on average, males achieve skeletal maturity as shown by a wrist x-ray, before they reach the age of 18 years or, alternatively, at about the age of 18 years.

Dr V. Low ’s evidence that, on average, males achieve skeletal maturity of the wrist at 19 years of age is inconsistent with all other evidence on this issue before the Commission. It also appears to be inconsistent with the GP Atlas itself; the very tool on which Dr V. Low places reliance when giving opinion evidence as to age. This conclusion can be drawn from the authors’ description of the method by which the standard plates were selected for inclusion in the GP Atlas and from the absence of data for the age of 18 years in Tables III and V of the GP Atlas. Dr V. Low ’s evidence in this regard cannot reasonably be accepted.

It may therefore be concluded that an expert assessment, made by reference to the GP Atlas, that a wrist x-ray shows a mature wrist is not informative for the purpose of establishing that the individual is over the age of 18 years. Having a mature wrist is quite consistent with a person being under the age of 18 years.

It does not follow from the above conclusion that an assessment that a wrist x-ray shows an immature wrist is not informative for the purpose of establishing that an individual is under the age of 18 years. As the above discussion makes plain, subject to the matters discussed below, an immature wrist is statistically consistent with a person being under the age of 18 years.

Even if the above conclusion concerning the age at which males on average achieve a mature wrist is put to one side, problems remain in relying on a mature wrist x-ray as evidence that an individual is over the age of 18 years. The most significant of these problems arises from the extent of normal variation in the chronological age at which males generally achieve skeletal maturity. The extent of this normal variation is not less than two years. For this reason, expert opinion that a young man is an adult that is based upon analysis that he is skeletally mature, is of limited, if any, probative value. Even if it were the case, contrary to the strong weight of expert evidence, that on average young men achieve skeletal maturity at the age of 19 years, the range of normal variation would reach to at least the age of 17 years.

Another complexity is that it cannot be known whether the individuals suspected of people smuggling who say that they are children come from a population that has a different mean age of skeletal development than the study sample on which the GP Atlas is based. Should this be the case, the potential margin of error in age assessments based on the GP Atlas could be increased.

Additionally, it may be concluded that the potential for errors in the use of wrist x-rays as evidence of chronological age is real – particularly where the medical practitioner who interprets the wrist x-ray does not regularly involve himself or herself in a quality assurance program.

The above conclusions demonstrate that expert opinion evidence based on wrist x-ray analysis should not be accepted for the purpose of establishing that an individual is over the age of 18 years; evidence of this character is not probative of this issue. Consequently, wrist x-rays should not remain a ‘prescribed procedure’ for the purposes of s 3ZQB of the Crimes Act unless the purpose for which they, or evidence based on them, may be adduced in evidence is limited to establishing that a young person is under 18 years of age. That is, wrist x-rays should only be able to be relied on, where appropriate, as evidence tending to establish that an individual is under the age of 18 years.

8.2 Findings regarding dental x-ray analysis for the assessment of chronological age

Even if the current statistical analysis suggests that dental x-rays are more statistically reliable than wrist x-rays, it does not follow that dental x-rays should replace or supplement wrist x-rays for age assessment purposes in Australia. Nor, for the following reasons, does it follow that dental x-rays should become a prescribed age assessment procedure under the Crimes Act.

First, dental x-rays are not sufficiently informative of whether an individual is over 18 years of age for them to be relied upon as evidence of age in criminal proceedings.

Second, dental x-rays appear to share many of the inherent weaknesses of wrist x-rays as a means of assessing chronological age. In particular, the inescapable factor of substantial variation in dental development between individuals means that dental x-rays are unlikely to provide reliable evidence of age in individual cases.

Third, there is no evidence to suggest that a wrist x-ray together with a dental x-ray will allow more reliable age assessment than either of those methods alone.

Finally, as shown in Chapter 3, giving wrist x-rays the status of a prescribed procedure under the Crimes Act for the purpose of age determination has resulted in their being seen as far more informative for this purpose than they really are. There is a real risk that the same result will follow if dental x-rays are given the equivalent status.

8.3 Findings regarding the ethical implications of the use of x-ray analysis for the assessment of chronological age

As the IAEA has made plain, human imaging using ionising radiation when undertaken for occupational, legal or health insurance purposes, and without reference to clinical indication, should normally be deemed to be not justified.

Even where human imaging of this character can otherwise be justified, it will only be ethically acceptable, according to principles accepted both internationally and in Australia, if the particular use is subject to a formal process of justification.

Given the above findings that wrist and dental x-ray analyses can provide limited, if any, reliable information concerning whether an individual has attained the age of 18 years, it is very difficult to see how their use for age assessment purposes could be justified.

8.4 Findings regarding the use of other biomedical markers, multifactorial medical approaches, and multi-disciplinary approaches to the assessment of chronological age

There are inherent problems with the use of all other biomedical markers of age of which the Commission is aware. Physical examinations are not appropriate for use because of the wide variation in pubertal development between individuals within any age group, and because they involve examination of a person’s genitalia, an intrusive process that is not justified for administrative purposes.

The processes of assessing age through an x-ray of the clavicle, MRI, or through an ultrasound of the wrist or elbow all require further research before they can be appropriately used for age assessment purposes.

Although some submissions to the Inquiry indicate that a multifactorial medical approach might lead to a more accurate assessment of chronological age, there appears to be no consensus as to which methods should be combined or how they should be weighted.

Finally, while a multi-disciplinary approach to age assessment appears to be a common sense approach, and appears to have some support amongst both medical and non-medical authorities, the Commission is not aware of any evidence that a combined approach will provide a more accurate age assessment than either medical or non-medical approaches alone. Individual differences in physical and social maturity will remain, and both approaches are vulnerable to the subjective views of the assessor. There also remains a possibility that within a multi-disciplinary approach, greater weight will be given to medically-based processes. If a multi-disciplinary age assessment process is conducted, a wide margin of benefit of the doubt should be afforded to individuals whose age is being assessed.

^Top

---

^[72] Commonwealth Director of Public Prosecutions, Prosecution Policy of the Commonwealth: Guidelines for the making of decisions in the prosecution process, para 2.15. At http://www.cdpp.gov.au/Publications/ProsecutionPolicy/ProsecutionPolicy.pdf (viewed 9 July 2012).
^[73] Australian Government, Joint submission, Submission 30, p 6.
^[74] Migration Act 1958 (Cth), s 233C.
^[75] Migration Act 1958 (Cth), s 236B.
^[76] Migration Act 1958 (Cth), s 236B(2).
^[77] WW Greulich and SI Pyle, Radiographic Atlas of Skeletal Development of Hand and Wrist (2^nd ed, 1959) (GP Atlas), p 27.
^[78] GP Atlas, above, preface.
^[79] Tanner and Whitehouse introduced a more complex process of wrist x-ray assessment in 1962 (TW2 method) in which every one of the 20 bones of the hand and wrist is scored against pictorial and written criteria from 2700 British lower and middle class children’s x-rays. The data for this method were updated in 1995 and 2001 (TW3 method), so as to reflect the secular changes that have occurred in the speed of bone development in adolescence: JM Tanner, RH Whitehouse, WA Marshall, MJR Healy and H Goldstein, Assessment of Skeletal Maturity and Prediction of Adult Height (TW2 Method) (1^st ed, 1975); JM Tanner, RH Whitehouse and N Cameron, Assessment of Skeletal Maturity and Prediction of Adult Height (TW3 Method) (3^rd ed, 2001) (TW3 Method).
^[80] GP Atlas, note 6, p 28.
^[81] GP Atlas, above.
^[82] GP Atlas, above, pp 15–18.
^[83] GP Atlas, above, p 33.
^[84] GP Atlas, above, pp 31, 51.
^[85] GP Atlas, above, p 32.
^[86] GP Atlas, above, p 49.
^[87] GP Atlas, above, p 49.
^[88] GP Atlas, above, p 51
^[89] GP Atlas, above, p 49.
^[90] GP Atlas, above, p 55.
^[91] GP Atlas, above, p 55.
^[92] Tim J Cole PhD ScD FMedSci, Professor of Medical Statistics, MRC Centre of Epidemiology for Child Health, Institute of Child Health, University College London, UK. In 2006 the British Royal College of Paediatricians and Child Health bestowed on him the title of Honorary Fellow for research of growth assessment in paediatrics.
^[93] Professor Tim Cole, Submission 5, p 5.
^[94] Professor Tim Cole, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 17.
^[95] Australian Government, Joint submission, Submission 30, p 11.
^[96] Dr Vincent Low, Submission 15, p 2.
^[97] Dr Vincent Low, Submission 15, p 2.
^[98] Dr James Christie, Submission 19, pp 4–5.
^[99] Professor Tim Cole, Submission 5, p 1.
^[100] Professor Tim Cole, Submission 5, p 4.
^[101] TW3 Method, note 8.
^[102] Professor Tim Cole, Submission 5, p 7.
^[103] Professor Tim Cole, Submission 5, pp 8–9.
^[104] Professor Tim Cole, Submission 5, p 9.
^[105] Dr Ella Onikul, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 16.
^[106] Dr James Christie, Submission 19.
^[107] Professor Al Aynsley-Green, Expert commentary on the age assessment of [WIL024] prepared for Fisher Dore Lawyers, Brisbane Australia, June 2011 (WIL024 – CDPP document 066.0023) p 8.
^[108] A Schmeling, W Reisinger, G Geserick and A Olze, ‘Age Estimation of Unaccompanied Minors, Part I. General Considerations’ (2006) 159S Forensic Science International S61, S62.
^[109] GP Atlas, note 6, p 40.
^[110] GP Atlas, above, p 44.
^[111] For example, Dr Vincent Low, Opinion regarding the use of skeletal age determination technique to estimate chronological age, CDPP, May 2011 (OFD030 – CDPP document 293.0505) (Dr V. Low Opinion for CDPP).
^[112] Chief Scientist, Office of the Chief Scientist, Letter to Deputy Secretary, National Security & Criminal Justice Group, AGD, 11 January 2012.
^[113] Professor Tim Cole, Submission 5, p 8.
^[114] Presidents of Australasian Paediatric Endocrinology Group, Paediatrics and Child Health Division – Australasian College of Physicians, Royal Australian and New Zealand Royal College of Radiologists, Australian and New Zealand Society for Paediatric Radiology, Australian Society for Adolescent Medicine, Paediatric Imaging Reference Group – RANZCR, Joint submission, Submission 9, pp 1–2.
^[115] A Schmeling, H-J Kaatsch, B Marré, W Reisinger, T Riepert, S Ritz-Timme,
FW Rösing, K Rötzscher and G Geserick, Study Group of Forensic Age Estimation of the German Association for Forensic Medicine: Guidelines for Age Estimation in Living Individuals in Criminal Proceedings (2000) (Guidelines for Age Estimation), p 3.
^[116] Hon R McClelland MP, Attorney-General, Correspondence to Dr A Cotterill, President, Australian Paediatric Endocrine Group, 18 October 2011 (AGD document CORRO-5), p 1.
^[117] GP Atlas, note 6, preface, p xii.
^[118] GP Atlas, above, p 40.
^[119] GP Atlas, above, p 40.
^[120] GP Atlas, above, p 40.
^[121] GP Atlas, above, pp 42–43.
^[122] WW Greulich, ‘A Comparison of the Physical Growth and Development of American-born and Native Japanese Children’ (1957) 15(4) American Journal of Physical Anthropology 489.
^[123] Australian Government, Joint submission, Submission 30, p 12.
^[124] A Schmeling, W Reisinger, D Loreck, K Vendura, W Markus and G Geserick, ‘Effects of Ethnicity on Skeletal Maturation: Consequences for Forensic Age Estimations’ (2000) 113 International Journal of Legal Medicine 253; RR van Rijn, MH Lequin, SG Robben, WC Hop and C van Kujik, ‘Is the Greulich & Pyle Atlas Still Valid for Dutch Caucasion Children Today’ (2001) 31 Pediatric Radiology 748, as cited in Senior Assistant Director, People Smuggling Branch, CDPP, People Smuggling Prosecutions Age Determination Issues, 15 December 2010 (CDPP document Attachment D document 4).
^[125] Dr V. Low Opinion for CDPP, note 40.
^[126] AM Zafar, N Nadeem, Y Husen and MN Ahmad, ‘An Appraisal of Greulich-Pyle Atlas for Skeletal Age Assessment in Pakistan’ (2010) 60(7) Medical Association of Pakistan Journal 552, 554.
^[127] FK Ontell, M Ivanovic, DS Ablin and TW Barlow, ‘Bone Age in Children of Diverse Ethnicity’ (1996) 167 American Journal of Roentgenology 1395, 1398.
^[128] K-H Chiang, AS-B Chou, P-S Yen, C-M Ling, C-C Lin, C-C Lee and P-Y Chang, ‘The Reliability of Using Greulich-Pyle Method to Determine Children’s Bone Age in Taiwan’ (2005) 17(6) Tzu Chi Medical Journal 417; A Zhang, JW Sayre, L Vachon, BJ Liu and HK Huang, ‘Racial Differences in Growth Patterns of Children Assessed on the Basis of Bone Age’ (2009) 250(1) Radiology 228; B Büken, AA Safak, B Yazici, E Bükenand AS Mayda, ‘Is the Assessment of Bone Age by the Greulich–Pyle Method Reliable at Forensic Age Estimation for Turkish Children?’ (2007) 173 Forensic Science International 146, as cited in Presidents of Australasian Paediatric Endocrinology Group, Paediatrics and Child Health Division – Australasian College of Physicians, Royal Australian and New Zealand Royal College of Radiologists, Australian and New Zealand Society for Paediatric Radiology, Australian Society for Adolescent Medicine, Paediatric Imaging Reference Group – RANZCR, Joint submission, Submission 9.
^[129] BM Soegiharto, SJ Cunningham and DR Moles, ‘Skeletal Maturation in Indonesian and White Children Assessed with Hand-Wrist and Cervical Vertebrae Methods’ (2008) 134(2) American Journal of Orthodontics and Dentofacial Orthopedics 217.
^[130] BM Soegiharto et al, above, 218.
^[131] BM Soegiharto et al, above, 221.
^[132] Associate Professor Hofman, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 23.
^[133] Professor Tim Cole, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 24.
^[134] GP Atlas, note 6, p 43.
^[135] Presidents of Australasian Paediatric Endocrinology Group, Paediatrics and Child Health Division – Australasian College of Physicians, Royal Australian and New Zealand Royal College of Radiologists, Australian and New Zealand Society for Paediatric Radiology, Australian Society for Adolescent Medicine, Paediatric Imaging Reference Group – RANZCR, Joint submission, Submission 9, p 1.
^[136] Guidelines for Age Estimation, note 44, p 4. As noted earlier, this is the publication on which the ‘improved process for age determination in people smuggling matters’ jointly announced by the then Attorney-General and the then Minister for Home Affairs and Justice on 8 July 2011 was purportedly based.
^[137] As discussed in Chapter 4, section 5.
^[138] Legal Aid Queensland, Submission 6, p 2.
^[139] Victoria Legal Aid, Submission 13, p 13.
^[140] Attorney-General and Minister for Home Affairs and Justice, ‘Improved process for age determination in people smuggling matters’ (Media release, 8 July 2011). At http://parlinfo.aph.gov.au/parlInfo/search/display/display.w3p;query=Id%3A%22media%2Fpressrel%2F906838%22 (viewed 9 July 2012).
^[141] Consultation with agencies, Crimes Act Regulations – Prescribing dental and clavical x-rays as additional age determination procedures, 15 November 2011 (AGD document IMP-18).
^[142] Australian Government, Joint submission, Submission 30, p 13.
^[143] Hon R McClelland MP, Attorney-General, Correspondence to Hon C Branson QC, President, Australian Human Rights Commission, 30 June 2011.
^[144] See for example, Victorian Institute of Forensic Medicine, Submission 18; Australian Society of Forensic Odontology, Submission 4; Dental Age Assessment Team, King’s College London Dental Institute, Submission 7.
^[145] See for example, Professor Tim Cole, Submission 5; Professor Al Aynsley-Green, Submission 38.
^[146] Australian Society of Forensic Odontology, Submission 4, pp 3, 7; Dr Anthony Hill, Transcript of hearing, Public hearing for key medical experts (9 March 2012), pp 31–32.
^[147] One study examining a variety of classification methods found Demirjian to be the most reliable. See A Olze, D Bilang, S Schmidt, K Werneke, G Geserick and A Schmeling, ‘Validation of Common Classification Systems for Assessing the Mineralization of Third Molars’ (2005) 119 International Journal of Legal Medicine 22.
^[148] Dental Age Assessment Team, King’s College London Dental Institute, Submission 7.
^[149] Australian Society of Forensic Odontology, Submission 4, p 7.
^[150] Dr Anthony Hill, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 30.
^[151] Dr Stephen Knott, Submission 17, p 5.
^[152] Dr Anthony Hill, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 33.
^[153] The King’s College submission to the Inquiry also includes a number of research articles calculating the reliability of dental assessments more generally, but none of those articles focus on the 18 year threshold – the closest study focuses on the 16 year threshold: Dental Age Assessment Team, King’s College London Dental Institute, Submission 7.
^[154] Professor Tim Cole, Submission 5, p 11.
^[155] Professor Tim Cole, Submission 5, p 11.
^[156] Professor Graham Roberts, Emails to the Australian Human Rights Commission, 5 March 2012 and 16 April 2012.
^[157] Professor Graham Roberts, Emails to the Australian Human Rights Commission, 5 March 2012 and 16 April 2012; Professor Tim Cole, Email to the Australian Human Rights Commission, 7 March 2012; Professor Tim Cole, Transcript of hearing, Public hearing for key medical experts (9 March 2012), pp 34–37.
^[158] TJ Cole, ‘Hot Potato Topic’ (2008) 205 British Dental Journal 581; GJ Roberts and VS Lucas, ‘Ethical Dental Age Assessment’ (2009) 207 British Dental Journal 251.
^[159] Dr Graham Roberts, Email to the Australian Human Rights Commission, 16 April 2012.
^[160] Professor Tim Cole, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 35.
^[161] Australian Society of Forensic Odontology, Submission 4, p 7.
^[162] Australian Society of Forensic Odontology, Submission 4, p 11.
^[163] Victorian Institute of Forensic Medicine, Submission 18, p 1.
^[164] The George Institute for Global Health, Submission 22, p 10.
^[165] See for example, A Olze, A Schmeling, M Taniguchi, H Maeda, P van Niekerk, K Wernecke and G Geserick, ‘Forensic Age Estimation in Living Subjects: The Ethnic Factor in Wisdom Tooth Mineralization’ (2004) 118 International Journal of Legal Medicine 170; DL Zeng, ZL Wu, MY Cui, ‘Chronological Age Estimation of Third Molar Mineralization of Han in Southern China’ (2010) 124 International Journal of Legal Medicine 119; H Crawley, When Is a Child Not a Child? Asylum, Age Disputes and the Process of Age Assessment, International Law Practitioners Association Research Report (May 2007) (When Is a Child Not a Child), pp 31–32; EF Harris and JH McKee, ‘Tooth Mineralisation Standards for Blacks and Whites From the Middle Southern United States’ (1990) 35(4) Journal of Forensic Science 859; A Demirjian, H Goldstein and JM Tanner, ‘A New System of Dental Age Assessment’ (1973) 45 Human Biology 221; A Olze, W Reisinger, G Geserick and A Schmeling, ‘Age Estimation of Unaccompanied Minors, Part II. Dental Aspects’ (2006) 159S Forensic Science International S65.
^[166] PW Thevissen, S Fieuws and G Willems, ‘Human Third Molars Development: Comparison of 9 Country Specific Populations’ (2010) 201 Forensic Science International 102.
^[167] HM Liversidge, ‘Interpreting Group Differences Using Demirjian’s Dental Maturity Method’ (2010) 201 Forensic Science International 95.
^[168] A Olze, W Reisinger, G Geserick and A Schmeling, note 94.
^[169] Victorian Institute of Forensic Medicine, Submission 18, p 5.
^[170] Dr Anthony Hill, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 40.
^[171] Dr Stephen Knott, Submission 17, p 4.
^[172] The George Institute for Global Health, Submission 22, p 8.
^[173] A Schmeling, C Grundmann, A Fuhrmann, HJ Kaatsch, B Knell, F Ramsthaler, W Reisinger, T Riepert, S Ritz-Timme, FW Rosing, K Rotzscher and G Geserick, ‘Criteria for Age Estimation in Living Individuals’ (2008) 122 International Journal of Legal Medicine 457, 459.
^[174] A Aynsley-Green, TJ Cole, H Crawley, N Lessof, LR Boag and R Wallace, ‘Medical, Statistical, Ethical and Human Rights Considerations in the Assessment of Age in Children and Young People Subject to Immigration Control’ (2012) British Medical Bulletin 1 (forthcoming).
^[175] Australian Government, Joint submission, Submission 30, p 14.
^[176] Australian Society of Forensic Odontology, Submission 4, p 23.
^[177] IAEA, Radiation Protection and Safety of Radiation Sources: International Basic Safety Standards – Interim Edition, General Safety Requirements: Part 3 (2011). At http://www-pub.iaea.org/books/iaeabooks/8736/Radiation-Protection-and-Safety-of-Radiation-Sources-International-Basic-Safety-Standards-Interim-Edition-General-Safety-Requirements-Part-3 (viewed 9 July 2012).
^[178] ARPANSA, Code of Practice: Radiation Protection in the Medical Applications of Ionizing Radition, Radiation Protection Series No. 14, May 2008, para 2.2.1.
^[179] ARPANSA, ‘Response to AGD re: human imaging for age determination’, 6 March 2012, Attachment – Email from Principal Legal Officer, AGD, to Australian Human Rights Commission, 6 March 2012 (ARPANSA Response to AGD), p 1.
^[180] The Royal Australian and New Zealand College of Radiologists, Submission 14, pp 2-3.
^[181] Presidents of Australasian Paediatric Endocrinology Group, Paediatrics and Child Health Division – Australasian College of Physicians, Royal Australian and New Zealand Royal College of Radiologists, Australian and New Zealand Society for Paediatric Radiology, Australian Society for Adolescent Medicine, Paediatric Imaging Reference Group – RANZCR, Joint submission, Submission 9, p 2.
^[182] Dr James Christie, Submission 19, p 2.
^[183] ARPANSA Response to AGD, note 108, p 1.
^[184] A Schmeling et al, note 37, p 17.
^[185] See A Schmeling et al, above.
^[186] See for example, When Is a Child Not a Child, note 94, p 33; T Smith and L Brownlees, Age Assessment Practices: A Literature Review and Annotated Bibliography, UNICEF Discussion Paper (2011) (Age Assessment Literature Review), pp 20–21.
^[187] Dr Stephen Knott, Submission 17, p 5.
^[188] A v London Borough of Croydon & SSHD; WK v SSHD & Kent County Council [2009] EWHC 939, as cited in UK Border Agency, Guidelines for Assessing Age, 2011, section 6.5.
^[189] Professor Al Aynsley-Green, Submission 38, p 19.
^[190] A, R (on the application of) v London Borough of Croydon [2009] EWHC 939, as cited in Refugee Studies Centre, Negotiating Childhood: Age Assessment in the UK Asylum System, Working Paper Series No 67 (2010), p 21.
^[191] See for example, Professor Al Aynsley-Green, Submission 38, pp 16–17; When Is a Child Not a Child, note 94, p 33; Age Assessment Literature Review, note 115, pp 21, 28.
^[192] Professor Al Aynsley-Green, Submission 38, p 16.
^[193] Hon Robert McClelland MP, Attorney-General, Correspondence to Hon C Branson QC, President, Australian Human Rights Commission, 30 June 2011.
^[194] A Schmeling et al, note 37, S62. See also Associate Professor Daniel Franklin, Submission 16, citing various studies; S Black, A Aggrawal and J Payne-James, Age Estimation in the Living: The Practitioner’s Guide (2010) (Age Estimation in the Living), p 142.
^[195] Age Estimation in the Living, above, pp 142–144.
^[196] Age Estimation in the Living, above, p 144.
^[197] The George Institute for Global Health, Submission 22, p 7.
^[198] The George Institute for Global Health, Submission 22, p 7.
^[199] The George Institute for Global Health, Submission 22, p 8.
^[200] Professor Al Aynsley-Green, Submission 38, p 28.
^[201] Associate Professor Daniel Franklin, Submission 16.
^[202] Victorian Institute of Forensic Medicine, Submission 18.
^[203] Associate Professor Daniel Franklin, Submission 16, p 2.
^[204] Associate Professor Daniel Franklin, Submission 16, p 2.
^[205] Australian Government, Joint submission, Submission 30, p 7.
^[206] Age Estimation in the Living, note 123, p 144.
^[207] A Schmeling et al, note 102, 459.
^[208] Professor Al Aynsley-Green, Submission 38, p 26.
^[209] Australian Society of Forensic Odontology, Submission 4, p 7.
^[210] Dr Anthony Hill, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 43-4.
^[211] Professor Tim Cole, Transcript of hearing, Public hearing for key medical experts (9 March 2012), p 44-5.
^[212] Professor Al Aynsley-Green, Submission 38, p 31.
^[213] The Royal Australian and New Zealand College of Radiologists, Submission 14, p 2.
^[214] The Royal Australasian College of Physicians, Submission 11, p 2.
^[215] Australian Society of Forensic Odontology, Submission 4, p 10.
^[216] Age Assessment Literature Review, note 115, p 12.