Into the New Year…A PhD Update

My last blog was at Christmas and I’ve been meaning to do another one for the last few weeks but never seemed to get around to it – so here it is finally!

Since the Christmas break (which was lovely and lazy!) and settling back into work I’ve really started to get going with my project. The first step of my project was to propose an adjusted method for recording dental wear in human molars, which was completed before Christmas. This also included identifying the aging method, using human dentition, that would most be the most accurate and easy to apply and came to the conclusion of using the AlQahtani et al (2010) London Dental Atlas. About 2 years ago I wrote another blog piece discussing the accuracy of three different dental age estimate charts, including the London Atlas. The article describes the AlQahtani et al (2010) method in brief but the atlas itself can be viewed here.

The next step was to test out the method that I had created to assess how it worked when practically applied to human skulls. In order to do this I needed a good skeletal collection of the correct time period for my project (neolithic to medieval) that included juvenile and adults skulls. Luckily for me there is such a collection at Southampton and so it was very easy for me to get started! You can find out a bit more about the teaching collection and courses that Southampton University at the following blog: BOS, curated by the wonderful Ellie!

So it’s now the beginning of February and I’ve managed to apply my method to the juvenile skulls twice and the adults once. My plan is to do this a few more times for each set of skulls and there are a few reasons for this. The first is that this will enable to tweek my method where necessary so that it is as easy and accurate to use as possible. The second reason is so that I can become comfortable with handling and identifying teeth, but also so that I get used to the process of recording the data. This will be particularly useful when I move on to different skeletal collections, where I will only have a short period of time with them.

Finally, and probably most importantly, recording the data from the same set of skulls, multiple times will allow me to test the repeatability of my method. In order for this method to be robust and accurate it must produce statistically similar results when applied to the same collection, this will ensure that any results obtained using this method can be comparable. When recording the data from the same collection of skulls I am making sure that there is a period of time in between each data collection. This ensures that I cannot remember any particular individual and all of the collected data is ‘new’ again.

So far, the data collection is going well and I am getting through the skulls in the time that I wanted. Another key aspect of my project is to produce a method that is quick and easy to apply, and at the moment that’s going to plan so fingers crossed it stays that way!

I probably won’t do another blog post on the progress of my PhD for a little while as I won’t have much new to say until I’ve finished my data collection. However, I do plan to do a new blog post on molar identification (seeing as I’m getting pretty good at it now!). Hopefully, I’ll sort that out soon and get it posted!

A Look at the Accuracy of Dental Age Estimation Charts

A child's skull before all of the deciduous teeth are lost.

A child’s skull before all of the deciduous teeth are lost.

So it’s now the 4th of July and I haven’t created a new skull of the month and I don’t think I will. Over the past few months I’ve been really rubbish at updating my blog and the reason behind that – I’ve been applying for jobs and working out what I’m going to do at the end of July (my current contract is ending and I have to decide whether to re-sign it or not!)

Anyway, because of all this I’ve decided that for July I’m going to try and catch-up with some of the articles that I’ve wanted to share over the past couple of months. Some relate to various skulls of the month whilst others are just articles of interest. Ideally, I’m going to do a couple a week (depending on how many applications I have outstanding!) so fingers crossed!

To start with I’m going to read an article by AlQahtani, Hector and Liversidge called ‘Accuracy of dental age estimation charts: Schour and Massler, Ubelaker and the London Atlas’. During my osteology courses at university I used both the Schour and Massler and Ubelaker for estimating age from dental remains. I came across the article a while back and I just haven’t got round to reading it. 

This particular study caught my eye because it’s aim was to compare the accuracy of estimating age from developing teeth from the above methods. When using these types of methods, using illustrations and charts to form an estimate, it is important that they are reassessed over time to test their accuracy. This is essential in subjects like archaeology and anthropology where the outcome of the result could be very fundamental.

I have discussed in the past how teeth are used for aging skeletal collections. These answers help to provide a profile of the individual and in forensic environments could potentially lead to the identification of a missing person. In archaeology age at death is used as an indicator of a populations health, providing an insight into past life and communities. In both situations obtaining an accurate and reliable age estimate is key, and therefore it is necessary that the common methods used for this process are assessed periodically.

The article by AlQahtani, Hector and Liversidge first provides a small amount of background to each of the chosen methods under study.  This image of Schour & Massler should be familiar to any osteology student. It depicts 21 drawings of dental development from the age of 31 weeks in utero up until adulthood. The method was apparently criticized in the early days due to the lack of information about the material or the method to provide the images. Since the original production in 1941 revisions have been made of  Schour & Massler’s atlas with the use of radiographs.

Schour & Massler's 1944 chart, taken from A Test of Ubelaker’s Method of Estimating Subadult Age from the Dentition (E. Smith 1999( p.30.

Schour & Massler’s 1944 chart, taken from A Test of Ubelaker’s Method of Estimating Subadult Age from the Dentition (E. Smith 1999) p.30.

In 1978 Ubelaker attempted to improve the method for obtaining age estimates from dental development and used a number of published sources. This method was seen as one which covered a the range of variation that can been seen at each stage of development. Again, most osteologists will have seen the following chart.

Ubelaker's 1989 chart, taken from A Test of Ubelaker’s Method of Estimating Subadult Age from the Dentition (E. Smith 1999) p.31.

Ubelaker’s 1989 chart, taken from A Test of Ubelaker’s Method of Estimating Subadult Age from the Dentition (E. Smith 1999) p.31.

Finally, there is the London Atlas which has combined many resources to create a dental chart that includes 31 age categories. This cart is tooth specific and defines each tooth by the development of it’s enamel, dentin, and pulp cavity.  The other bonus is that this chart has been made freely available here.

Second page of the London Atlas showing the development of individual teeth avaiable here.

Second page of the London Atlas showing the development of individual teeth available here.

That’s the method under investigation briefly covered now onto the study. As I have already mentioned the aim of the study was to assess the accuracy of estimating age from developing teeth using the three methods above. In order to do this a large sample of skeletal remains of known age at death was used. This included 183 individuals who were aged between 31 weeks in utero to 4.27 years and a further 1323 individuals (649 male and 674 female) aged between 2.07 and 23.86 years. In addition to skeletal remains the archived dental radiographs of living patients were also used. From this the age estimates were made using the three methods chosen for study and compared to the chronological age. To determine the reliability of the aging methods statistical analysis, including a paired t-test was carried out.

Before moving onto the results it is worth mentioning that intra-observer error was taken into account. This means that the same individual contacted the age estimates multiple times of a sub-set of the study sample. From doing this it was found that the error rate was low, meaning taht there was excellent when repeated for all three of the methods.

So the final results. The method which produced the closest age estimate to the chronological age was the London Atlas. For the other two methods it was found that they both under-estimated by around 0.75 years. This result was attributed to the amount of categories produced by Schour & Massler and Ublekaer as their age categories are much larger as the individual ages. This potentially misses out important development phases which can provide a more accurate age estimate. This is true of the growth of the third molar. As can be seen in the dental charts above the development of this molar is not depicted in detail, unlike the London Atlas. Due to this AlQahtani et al. removed the age estimates using the third molar and discovered that both the Schour & Massler and Ubelaker charts still under-estimate the age but only by 0.5 years. In comparison the London Atlas had a mean difference of 0 years.

This study showed, that whilst all three methods of estimating age using dental development is fairly accurate the London Atlas performed the best. As stated above these tests are important to the study of osteology and anthropology. This is because determining the age at death is vital of identifying individuals and establishing the health of a population for their dead. These studies cause researchers to constantly check their methods and to further their work in order to obtain the most reliable system that is possible, which is extremely important.

References

S. J. AlQahtani, M. P. Hector and H. M. Liversidge (2014). ‘Accuracy of dental age estimation charts: Schour and Massler, Ubelaker and the London Atlas.’ In the American Journal of Physical Anthropology Volume 154, Issue 1, pages 70–78, May 2014

E. SMith (1999). ‘A Test of Ubelaker’s Method of Estimating Subadult Age from the Dentition’. A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science in Human Biology in the Graduate School of the University of Indianapolis May 2005.