Age determination

SPEAKER 1: Welcome to “Justice Today,” the official podcast of the Department of Justice’s Office of Justice Programs, where we shine a light on cutting-edge research and practices and offer an in-depth look at what we’re doing to meet the biggest public-safety challenges of our time. Join us as we explore how funded science and technology help us achieve strong communities. 

JIM DAWSON: Hello. I’m Jim Dawson, a science writer with the National Institute of Justice. This is the second half of my conversation with Kyra Stull and Danielle McLeod-Henning about subadult skeletal remains. Kyra holds a Ph.D. in Biological Anthropology and is an anthropologist and forensic researcher at the University of Nevada, Reno, and Danielle is a physical scientist and program manager with NIJ.

In Part 1, we discussed the definition of subadult, generally anyone less than 20 years of age, and how to tell the age at death from subadult skeletal remains. The age-at-death estimates are forensically challenging, and our current conversation picks up with how they’re done.

So, if I go back to the, you know, what you hear on the crime shows, you know, within five years one way or the other, you clearly can’t do that. So, when you get done, if you're looking at a seven-year-old, do you have a range when you're done, or can you say this is a seven-year-old, or is it six-, seven-, or eight-year-old? How precise can you be, and is that evolving as you do more research and get this figured out?

KYRA STULL: You know, I would love to be able to tell you that I could say it is exactly a seven-year-old. [Laughs] Unfortunately, I am not able to say that. I do think that--so we always present age estimation as an interval because human variation exists, and we have to be able to capture that human variation and provide a level of uncertainty with it because it always exists, right? You have some kids that grow faster than others, some kids that grow slower than others, and we know just based on growth charts that the pediatricians use that there is some variability all the time. And forensic anthropologists absolutely want to make sure that you capture that variability because otherwise you may end up in a situation where you're not actually facilitating identification, right, and you're impeding identification. So, we always are providing an age range. That age range, unfortunately, is--starts fairly small and increases in width as you get older because--one of the ways that I like to explain it to people is that when you are alive longer, your body has this wonderful ability to interact and respond to the environment more and more and more and more, and as you are increasing in age, you have greater exposure to that environment. That relationship between your genetics and environment changes, and so our variation as humans also increases until we get to be adults. And people who study adult age estimation will tell you that, you know—[chuckles]—it is wide and stays wide for all of life after that. [Laughs] 

JIM DAWSON: I like the there’s no best indicator notion, too. We talked about that, and it seems to make your job a little more difficult, though, because you have to figure out which indicators--you’ve got to find the indicators that are relevant, as opposed to just, you know, starting with a set-formula kind of thing, it seems to me. 

KYRA STULL: Well, you know, the beauty is that if you can give--so we created a graphical user interface, which maybe we'll get to and I'm jumping the gun, but that individuals can put all the indicators into and get an age estimate so that it is not at least a subjective choice by the practitioner to use, and it is something that the model can choose, which is really helpful because it removes the subjectivity from those age estimations, but--so if you put all the data forward, it will tell you the best result. 

JIM DAWSON: OK, and this is the KidStats database. Is that correct? KidStats and KSCollect and—

KYRA STULL: Yeah, KidStats has become a hub because, as we have started to build age estimation methods and sex estimation methods, we have realized that KidStats actually changed. Its first version was from my dissertation a long time ago, with just looking at South African data. But now, KidStats has actually become a mother hub, and inside of that, we are making all of the GUIs available. So, the one for age estimation that is being published hopefully in the next two weeks. It's called mcp-s-age, so that will be available there. Sex estimation will be available there, and so KidStats is now a hub for all the statistical methods that help improve the subadult biological profile. 

JIM DAWSON: So, are researchers uploading new information to it, as well as using it? I mean, how does this work in the real world with forensic researchers or crime investigators, too, I guess? 

KYRA STULL: Yeah, so right now, we have--on KidStats, we’ll have two GUIs available, and people can--anyone can access them. They're freely available to use, and essentially, one of them is coming from one of my students. It's called the Ontogenetic Subadult Sex Estimation System. Again, I'm--I should have--[chuckles]--Stephanie Cole is doing it, and that is her dissertation, so she's looking at pelvic and cranial morphology of children. And then the mcp-s-age, which is my GUI, will be up as well. And we have at least two more that are in development right now, so they are--anyone can technically use KidStats as a place to upload GUIs for subadult techniques. I 100% believe in open access and open science, so the more collectively we work together to improve our methods, the better we are as a field at the end, so it would be available for anyone to upload anything onto it. The Subadult Virtual Anthropology Database is the same way, so technically, my research team is the only team that has contributed data to it, but we have--we encourage all other colleagues and researchers to contribute data to it, and as part of that, we also created what's called the Subadult -- or, well, it’s the Subadult Virtual Anthropology Database, and it’s on a Zenodo community. So, Zenodo is actually a platform to access open science. You can put publications, you can put datasets, you can put protocols, and so we created a community so that others can also use that as a platform to make data available or actually link publications directly to subadult research. 

JIM DAWSON: OK, and what's the response now? You've had this out there for a bit. Are you getting a good response and people responding? It seems to me, from my early reading, that there wasn't a lot there before you. There was some researchers, but not a lot and…

KYRA STULL: Yeah. It's—

JIM DAWSON: So, what's your reaction? 

KYRA STULL: Well, it's good. I will say that most of the--[chuckles]--the reaction is very positive, although, when you work in a land of working with complex computational models, your field generally asks for it to be made in the most simplest form possible, and so really bridging--I mean, I have had amazing feedback from colleagues helping me kind of bridge that gap even more in terms of complex methods and really user-friendly approaches, and not only just being able to articulate the science and statistics behind it, but just also making it user-friendly. So, the feedback has been great. You know, I think that there is a learning curve…that comes with complex modeling, and age estimation, in particular, is just in a deep world of complex modeling. [Laughs] So that has been something that--but the actual, you know, feedback has been great. It is very exciting to be able to help researchers get answers to when they are faced with subadult skeletal remains. You know, I think one of the previous times we talked, you know, I had some international colleagues reaching out about cases, and being able to help provide reference data, provide age estimations, it really does--it is extremely rewarding and mainly because there really, like you said, hasn’t been a lot out there to use previously, so… 

JIM DAWSON: And, Danielle, I am guessing that NIJ will continue its interest in this, so your--how does NIJ look at these kind of, these sort of theme projects? You know, you found this area of research, and it’s really been built up a lot through NIJ grants. How do you view that, and is there a goal with this? Do you have markers for it? 

DANIELLE McLEOD-HENNING: Yes, we will certainly continue to fund in this area. Kind of our overarching goal of our subadult research portfolio, kind of--it started back in 2008 with a project to Mercyhurst College with Dr. Stephen Ousley. And, you know, we've kind of been continuing to invest in this area over the past decade-plus to kind of develop and evolve analytical approaches to create this database that Kyra is talking about that allows investigators and forensic science practitioners to accurately determine age, sex, and population affinity of skeletal remains of infants, children, and adolescents. So, the work of Stull and other researchers, you know, has established skeletal reference collections of subadults, and that should significantly improve forensic science practice, aid in the investigation of death, assist in bringing justice in cases of violent death, and bring closure to loved ones. So, the research, I mean, is just absolutely invaluable, and we thank Stull and all of our researchers that we've been able to invest in this very, very important work.

JIM DAWSON: OK, and, Kyra, I have--if I'm a young girl or boy doing science or thinking about anthropology, how did you wind up in this field? This is not--you know, most anthropologists, I don't think, are doing this kind of thing. What--was it a drift into it, or did you make a decision? How did it happen? 

KYRA STULL: [Chuckles] It's a really great question. I was really lucky to--I mean, I think that I haven’t always had an interest in being what I--working in forensics. I thought I wanted to be a forensic pathologist, which--I still love forensic pathology, but I was really lucky to end up at the University of Tennessee, Knoxville for my undergrad, and within a semester, went from pre-med to an anthropology major. [Laughs] So, I had a sorority sister who was an anthropologist at the time, and she told me that I immediately needed to have a meeting with Lee Jantz as soon as possible. And so, as a naive freshman, I went to go meet with Lee, and she got me set up with volunteering with the Forensic Anthropology Center at a very young age, and I had an extraordinary experience as an undergrad at UT, Knoxville. I got to--I mean, so many of my colleagues now even come from the relationships I established as an undergrad. So, in the latest NIJ grant in 2019, one of my co-PIs, Cates Bradley, has known me since I was an 18-year-old and she was a Ph.D. student there, so, you know, these are longstanding relationships. And I consider the Jantzes, in particular, really important academic parents in the field. So, I was pretty lucky there, and then I continued to be able to work with Joe Hefner, and then that helped me get to Mercyhurst and working with Dennis Dirkmaat and Steve Sims and Steve Ousley. I was part of the grant in 2008 to make Patricia, which Steve did, so, you know, I will say that specifically working with subadults, I did not know until I was at a forensic case with Dennis Dirkmaat. 

It was the summer in between my first and second year in grad school, and they are wonderful in term--they, as in Mercyhurst, were wonderful in terms of integrating students really into working the forensic cases as soon as you got there. And I--that summer, I worked on a case with an infant and, you know, I just remember being really--well, I thought I did exactly what I should have done, and then Dennis told me that because of all these problems with subadult age estimation that that wasn't the best estimate that I could have come up with. So then, of course, being the student that I was, I looped Steve Ousley into the conversation, and Steve Ousley was like, “Well, you need to go find those original publications,” which happened to be from 1955, and it blew my mind that we were using something from--so old. And really, the gap in the field that existed to me at that time and that we didn't have anything more up-to-date just--it became so apparently a problem for me and, you know, while you said the statistics earlier in terms of missing children, the bottom line is that most forensic anthropologists don't deal with subadult remains as frequently as they do with adult remains, but in my opinion, we should really be proactive scientists rather than reactive, right? You don't ever want to be in a situation where you're faced with a ton of skeletal remains of children, and you're unable to help identify them because we don't have the methods available to do it. And so, there is just a huge gap in the field, and at that same time, I was working as an autopsy tech in Buffalo, and I was organizing their radiograph closet--[laughs]--and seeing all of these radiographs of children, which is very sad, I mean, you know. But I kept seeing, talking with Steve Ousley like, “I think we have a resource. I think there's a way that we can start to access variables in children that we previously can’t in skeletal collections.” And so, somehow, I think I have just been--I hate to call it “lucky enough,” but lucky enough to have a lot of things present themselves and me take advantage of them and, you know, really find my love in building the methods to remove that gap in our field. 

JIM DAWSON: That's terrific. All right, and as we wrap this up, let me get to one of the things I've learned over the years writing about science, that there's never an end point to this, right? You don't have a specific goal and then you're done and go into a new field. So how do you measure your progress? What are your--is your future “I want to achieve this by next year, expand the database”? How do you come at this to move into the future? 

KYRA STULL: You know, I wish--[laughs]--I could say that I want to achieve this by next year. There's a lot of things still, so--[chuckles]--but I do have a plan. I'm technically on sabbatical right now, and my goal during sabbatical is to finish a book, and that book is going to be essentially a visual atlas for other colleagues to use in their lab and in classes. One of the things that we've been lucky enough with especially, you know, having access to data from around the world and to a lot of them being CT scans is that we--my research team, AMIA--have been exposed to really the human variations surrounding growth and development. And so, one of the immediate things that will occur is the development of this book that will essentially provide a visual reflection of growth and development of each element from birth to 20, so that people can truly see the variation that occurs throughout ontogeny, and then we'll also provide some fusion data and growth and development in human variation information inside that book as well. 

I am hoping to--I mean, there are so many research questions and so many projects going on to this that I am sure I will be working on disseminating this information for years. The part of it that's difficult is it's extraordinarily exciting, so new projects are kind of always blossoming out of it, but I really think that truly understanding--I think one of the biggest things that will come from it is truly understanding the complex interactions that occur in growing children. You know, I think previously we have tended to think about it really independently, as one part of your body at a time. And I will say, currently, my research team is really invested in looking at it as a holistic individual and really taking it as, like, the whole picture, information, and trying to really better understand the biological relationships that are occurring all at one time, both with the cranium growing, the long bones growing, and really the expression of those traits. So, I think that we just are in such a place to really make a big impact, but I have no idea what that timeline is, except for I'm going to do this for a long time--[laughs]--and I encourage people to come help me. So--[laughs]. 

JIM DAWSON: OK. Well, I do appreciate it. This has been--and I thought I knew a little bit about this, but this has really been enlightening. I appreciate it. And so thank you, Danielle, thank you, Kyra. It's been terrific. I appreciate it. 

Thanks for listening to Part 2. If you missed Part 1 of this discussion, please check out the link in the episode description and stay tuned for future episodes from NIJ. 

SPEAKER 2: To learn more about today's topic or about NIJ, visit the links in the episode description and join us for new episodes every month.

SPEAKER 1: Welcome to "Justice Today," the official podcast of the Department of Justice's Office of Justice Programs, where we shine a light on cutting-edge research and practices and offer an in-depth look at what we're doing to meet the biggest public safety challenges of our time. Join us as we explore how funded science and technology help us achieve strong communities.

JIM DAWSON: Hello, I'm Jim Dawson, the science writer for the National Institute of Justice. Today I'm talking with Danielle McLeod-Henning and Kyra Stull about subadult skeletal remains. Kyra holds a Ph.D. in Biological Anthropology and is an anthropologist and forensic researcher at the University of Nevada, Reno. Danielle is a physical scientist and program manager with NIJ. I want to thank both of you for being here, and first, let me ask Kyra how one would define subadult. When you say that to people, they often look confused, so, if you could define it up front, I would appreciate it.

KYRA STULL: Yeah, thank you, Jim, for that introduction. I'm excited to be here, and I'm happy to try to provide a foundation for the rest of the conversation. So, in biological anthropology, subadult generally is referring to anybody that is an immature individual, which, of course, comes with its own complications. We roughly think about that as someone being between ages of birth and 18 to 20 years of age. Of course, when you then bring in legal thresholds and what that means in a socio-cultural context, that is--kind of changes as you are in different locations. But generally, we're talking about anyone less than 20 or 18 years of age.

JIM DAWSON: OK. And I don't mean to be overly grim, but what this really is, is when you find remains of children for forensic purposes, right, and you're looking at skeletal remains of a 4-year-old or a 9-year-old or a 14-, 15-year-old, how do you tell the age at death. And I guess you work from there. And I think a lot of us in the world have grown up with, well, you can tell the difference between boys and girls because of their pelvis and haven't given it much more thought of than that. But so let me get to that in a minute. 

But first I want to give an idea of how big the problem is. When I was looking at statistics on this, there are something like 460,000 children reported missing in the U.S. every year. And most of them come back quickly and are found, and it isn't a big deal, but about one in 10,000 of the kids that go missing are found not alive, I think is the right term for it. So that's the focus of this is that. And NIJ--clearly Danielle looked at this a number of years ago and started focusing on grants. And I wanted to get a feeling from her on why this is important for NIJ.

DANIELLE McLEOD-HENNING: Thank you, Jim. So, as you both know, so NIJ is the research, development, and evaluation agency of the U.S. Department of Justice and is one of the lead funders of forensic science research, encompassing all forensic research disciplines. But of those, we also, of course, include forensic anthropology, where research funds--or where NIJ funds research to improve technologies, techniques, and the application of forensic anthropology to assess biological profile, identify injuries, and estimate time since death to assist in identification and investigation. So, research into subadult biological profile estimation is an identified need of the forensic science community, and it also aligns with NIJ's recently published Forensic Science Strategic Research Plan, which is available now on nij.gov. These NIJ investments are paramount to assisting in the identification of unidentified subadult remains and aiding in death investigation. Not only is this research a need of the forensic science community, it's also a need of the public to bring justice in cases of violent death and to bring a sense of closure to loved ones.

JIM DAWSON: Thanks. Well, that's terrific. And that's--all right. And let me go to Kyra and get to the crux of this. When you're confronted with subadult remains, how are they different from adult remains? You know, we always see: body was found, and it was 35 years old, give or take 5 years, or something like that. And I assume there's standards for that. What are the unique problems with subadults that you face? Or that researchers and forensic people face when they're looking at this?

KYRA STULL: Yeah. So when we generally--when we interact with law enforcement, we generally as—we, as in forensic anthropologists, are usually asked to estimate a biological profile, which is that snapshot of an individual. So age estimation, sex estimation, population affinity, stature. And all of that is used to usually facilitate positive identification. But when we are faced with subadult remains, those individuals are still growing, right? So we haven't actually attained those adult phenotypes that we are all so—you know, we know so well. And even though with soft tissue, you can talk about how tall a child is or what that child looks like, we really don't have enough skeletal collections to have created those biological profile techniques for children. So there are a lot of biological and methodological complexities that come into actually making those estimations. But the real problem has been that we haven't had the data available to make those methods. 

So we, as forensic anthropologists, heavily rely on skeletal collections that are usually coming from donors or cadavers, right? So University of Tennessee, Knoxville has the Bass Collection, which is probably the most famous one that started back, you know, over 30 years ago now. But those are primarily individuals that have donated themselves to science or next of kin who donate themselves. And generally speaking, we don't have individuals that donate their children to science if they have died, which is absolutely understandable. But because of that, anthropologists have really been limited in method development to be able to build age estimation techniques or sex estimation techniques. So there's a lot of biological and methodological things that happen, but the first problem was that we actually didn't have data available to make the methods or test the methods or validate the methods. And historically, up until, you know, 5 years ago, we heavily depended on historic skeletons, which are individuals that have died, you know, 150 years ago, which we know have different growth patterns, have different environments, have different--come from a variety of situations that contemporary individuals aren't faced with today. So, using those historic skeletons are not generally recommended for use in a modern context.

JIM DAWSON: So, and as a result of this, you started--the--what--it's called the Subadult Virtual Anthropology Database, is one of the issues you've developed. Could you look at that? And I know this involves, just for people with a, you know, from, you know, not much education on this, you're looking--talking about bones, teeth, structure, growth plates, those kind of things in skeletons, all of those. And I know at different times, from your work and reading your papers, they come into play at different times. So what you look at for a 2-year-old is different than a 6-year-old. It's not just you look at growth plates and you look at the--I mean, these things evolve, and you have to figure that out. But how does that tie in? And you need--as you were talking with all good forensic science, it needs a database behind it. And so, talk about your database a little bit if you would, please.

KYRA STULL: Yeah, absolutely. So, even though back in 2015 when NIJ so kindly awarded me the first grant, we really had major outcomes of that grant to be creating new methods. And while we had planned to create a sample, that is really when we first came into realization that we can't do anything in terms of method development without sample-building first. And so at that time, that first grant went into working closely with the individuals at the Office of the Medical Investigator in New Mexico. So they now also were funded, I think a year later, to develop the New Mexico Decedent Image. They also were working towards the same problem. Anyway, so we worked with them first on an American sample. So, between them and the Office of the Chief Medical Investigator in Baltimore, Maryland, we collected data from about 1,300 individuals in the United States. Since then, we have worked with collaborators in seven more countries. So that is Colombia, Brazil, France, the Netherlands, Angola, South Africa, and Taiwan. And that was funded from the NIJ as well in 2017. So now we have data collected on these age indicators, and more now, from approximately 5,000 individuals, which is pretty remarkable. So that is, by far, the largest repository of skeletal and dental reference data that is available. 

And so, in these first two projects, we focused heavily on diaphyseal dimensions, epiphyseal fusion, dental development, and vertebral neural canal dimensions. And those first three variables are really the classic age indicators that we think about, and really they all have to do with how tall an individual is and how you go from having deciduous dentition to permanent dentition. And that is really the Holy Grail of all subadult age estimation techniques that have ever existed are from those three indicators. So since then, we also, like I said, have data from the vertebral neural canal, which gives us an idea of potentially a nonspecific stress indicator. And we now also have a grant from the NIJ from 2019, where we are collecting craniometrics, dental metrics, dental morphology, and cranial morphology from the United States sample. So the database is growing not only in individuals that comprise it, but also of the data within it. So I think at this point we have over--in some individuals, have over 450 data points collected on them, which is pretty remarkable.

JIM DAWSON: OK. It is. And let's explain to me then if you're looking--and let me--I can be a little simplistic here, but…

KYRA STULL: No, let's talk about it.

JIM DAWSON: Yeah, if you have three sets of remains and they're all subadults, what--how do you distinguish--I mean, I know they're different things at different stages, mentioned that, but what is it you're looking with when you're presented with this, if you walk up and there are the remains on the table? Do you start with size? How do you go about this? What do you look at? What do you look for?

KYRA STULL: Yeah, I think that that's a great question. In terms of those age indicators that we think of, they all kind of tell us different...or I should say they essentially put us in the correct location and ontogeny, and by ontogeny, I just mean growth and development, so how old that individual may be at certain times. So when we are very young, we can heavily rely on something like diaphyseal dimensions because they have a very strong correlation with chronological age. So, you know, if you think about how quickly kids grow when they're very young, that is how we can depend on those diaphyseal dimensions, because it's the length of their arms and their legs. But as you continue to increase in size and you grow older, the part of our body that's allowing us to grow is also that epiphyseal fusion. And so epiphyseal fusion actually occurs when you stop growing, right. So, we get through that wonderfully uncomfortable period in life of puberty, and then you've reached your maximum height. And during that really wonderful, uncomfortable time is when we can look at those fusion patterns, right? 

So where diaphyseal dimensions are really useful for young individuals, epiphyseal fusion is really helpful in individuals like from, you know, 10 years of age to 20 years of age. So you kind of have this trade-off between those two. And then dental development is really lovely because it occurs through at least the first two thirds of growth and development, right? So when you are a kid, you actually already have--well, I should say when you're in your first year of life, you already have your first molar growing inside of your jaws. And that first molar, so we would be able to look at that, you know, from the time it erupts at six and then through its root development three years later. And we all know what kids look like, you know, losing their teeth and getting their permanent teeth in. And that is--that mixed dentition and growth and development of their teeth, we can kind of use. Mostly, it's going to be the most informative from that 5- to 12-year-old period. So all of these indicators kind of tell us a different story about the individual, but then also, importantly, how they're interacting is going to give us the best idea. 

So I think, Jim, you and I talked about it for the previous publication, but because these are all telling us different stories, there's no really best age indicator that anthropologists can walk away from and say teeth are always going to be it or long bones are only going to be at or epiphyseal fusion is going to be perfect, because really it all depends on how old that individual is for which indicator is going to give you the best information, and also the interaction of the traits that you have available, right, because we generally are not faced with remains that are complete skeletons also. So we have to be able to work with what we have available.

JIM DAWSON: Thanks for listening to part one of this episode. Please stay tuned for part two of this conversation about subadult skeletal remains.

SPEAKER 2: To learn more about today's topic or about NIJ, visit the links in the episode description and join us for new episodes every month. 

Opinions or points of view expressed in this episode represent a consensus of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice. Any products and manufacturers discussed in this episode are presented for informational purposes only and do not constitute product approval or endorsement by the U.S. Department of Justice.