An occasionally irregular blog about orthodontics

Are historical controls valid? Secular traits in orthodontics.

Are historical controls valid? Secular traits in orthodontics.

Are historical controls valid? Secular traits in orthodontics

“Every year is getting shorter, never seem to find the time”  I am sorry to stay on the Pink Floyd theme, but this next blog is also about time and secular traits in orthodontics.

One of the more traditional methods that has been used for orthodontic research is to carry out a study that compares a convenience sample of patients who have received recent treatment with a group of untreated subjects from a historical control.  A significant number of these studies have been published over the last 20 years.  This methodology has been used because it has the following characteristics:

  • It avoids the problems and difficulties of carrying out a randomised controlled trial
  • The cases can be obtained from our stores of study models and cephalograms
  • The costs are less than a prospective trial
  • The project can be quickly and easily carried out and is often in the scope of an orthdontic Masters project.

The disadvantages are:

  • There is an inherent selection bias because there is no control of the record collection.
  • It is not possible to measure outcomes that are relevant to the patients, for example, self esteem and satisfaction.
  • Questions have been raised about the validify of historical controls because of the possibility of secular change in growth

This latter point is important because it is possible that the children who were included in the historical control samples have a different growth pattern, or rate of growth, than a child who is currently undergoing orthodontic treatment.  This study attempted to clarify this question.

DOI: 60-66 First published online: 4 August 2014



The authors set out to determine whether secular changes  occurred in a number of cephalometric variables over time.

What did they do?

They made use of the American Association of Orthodontists Foundation Craniofacial Growth Legacy.  This includes 138 sets of records obtained from the nine growth studies carried in the USA and Canada between 1930and 1982.  They selected 8 cephalometric variables and measured them at 9, 10 and 15 years.  They then analysed the effect of year of birth, growth study, age and gender on change in the variables.

What did they find?

They found that all variables changed with age, which was not surprising.  But it was very interesting that five of variables (SNA, ANB, S-N, Co-A) showed a signicant effect of year of birth and one decreased (SNB).  They also showed that the growth rate varied between the studies.

While they reported that the differences were statistically significant, the effect size was rather small, for example, SNA would decrease by 0.2 degrees/year for children born in 1923, but it would decrease by -0.025 degrees/year for a child born in 1965.  However, we also need to remember that most of the studies that used historical controls detected similar small differences, that are then described in detail.

They concluded

“It appears that secular trends are likely to result in modern craniofacial traits that are generally larger in size and different growth pattern than those of historical subjects.  This may also explain why modern RCTs report smaller differences than studies using historical controls”.

What did I think?

I thought that this paper was very interesting and relevant to the results of a large body of orthodontic research. I must admit that I found it difficult to follow. This is probably due to my mental block with cephalometric studies!

While the authors have drawn some firm conclusions, they have also drawn attention to some potential problems with their methodology. These are that the sample sizes were low and there was a wide distribution of skeletal classes.

Nevertheless, I think that this paper does start to add to evidence on the validity of the historical control.  As a result, I feel that the findings of retrospective studies using historical controls should be interpreted with caution.  This also continues to reinforce my opinion that the use of cephalometric in contemporary orthodontic research is not too useful.   More valid and clinically useful data can be obtained from outcomes such as patient perception, socio-psychological measures and aesthetics.  I have discussed this previously on this blog.  I know that I beginning to sound like a stuck record….time to listen to my old copy of Dark Side of the Moon….

I am taking a holiday break this week, so there may not be posting next week. This depends on the weather in the North of England, if it is wet and windy then there will be a post.

ResearchBlogging.orgAntoun, J., Cameron, C., Sew Hoy, W., Herbison, P., & Farella, M. (2014). Evidence of secular trends in a collection of historical craniofacial growth studies The European Journal of Orthodontics, 37 (1), 60-66 DOI: 10.1093/ejo/cju007

(Visited 375 times, 1 visits today)

Tags: , , , , ,

Post a Comment

Your email address will not be published. Required fields are marked *


Pin It on Pinterest