Do transfer cases take longer to finish?

When we transfer a patient to another operator, we hope the treatment will proceed well. Unfortunately, our clinical experience tells us that this is not always the case.  This new study tends to support this concept. I thought that it was interesting.

Transferring an orthodontic patient to another operator can make the continuation of treatment difficult. This may be because the new orthodontist needs to understand the treatment plan and goals, perhaps, change the mechanics and build a rapport with the patient.  As a result, treatment times and success of treatment may be compromised.  A team based in Turkey looked at this question.  The AJO-DDO published the paper.

What did they ask?

They did this study to:

“Compare the treatment duration and results between patients whose treatment was transferred to those whose entire treatment was completed by a single operator”.

What did they do?

The authors made a retrospective comparison of case series.  In the first part of the study, the investigators searched the record archive of the department of orthodontics. Next, they looked for transfer cases who completed their treatment between 2013 and 2018.

The inclusion criteria were:

• Completed fixed appliance treatment
• All records were available
• No more than two operators treated the transfer patients
• There was at least a six-month period before and after transfer.

Then they selected a control group of patients whose treatment was done by a single operator.

Notably, the same Faculty members supervised all the patients.

The primary outcomes were the treatment duration and the ABO CRE.

I could not find a sample size calculation in the paper.

What did they find?

I thought that these were the relevant findings:

• The duration of treatment was longer in the transfer cases group (33.6 months) than in the single operator group (20.1 months). P<0.001.
• The mean total CRE score for the transfer cases was 37.8 (SD=9.8), and for the single operator group, this was 30.64 (SD=5.53) P<0.001.
• 74% of the transfer patients had a CRE score of greater than 30 points, 54.7% of the single operator patients had a score greater than 30 points.

In the paper, the authors also compared non-extraction and extraction treatment.  However, I do not have sufficient space to go through these. Furthermore, this is not the central question of the paper.

The author’s conclusions were:

“There were statistically significant differences in the quality of treatment results between the group. Furthermore, the mean treatment times were significantly longer for the transfer patients”.

What did I think?

I thought that this paper looked at a clinically relevant question.  I believe that the results are clinically relevant.  However, I have some concerns with the report and methods we should consider when interpreting the results.

Firstly, this was a retrospective study. This factor means that we should always consider the high possibility of selection bias. When I looked closely at the inclusion criteria I found that the transfer cases needed complete records. Also, the authors did not provide any information on the total number of transfer patient records they found compared to the number of transfer patients in the clinical database? This fact leads me to conclude that there is a high risk of selection bias.

I am also not sure whether using the ABO-CRE as an outcome in studying multiple cases.  This is because we do not know the validity of this measure. I concede that I may be biased because I was a junior member of the team led by Bill Shaw and Steve Richmond when they developed the PAR index. This index is weighted for validity and is, perhaps, more relevant than the ABO-CRE.  I wonder if the ABO-CRE is a better measure for individual cases presented as an assessment of a candidate in an examination.

Nevertheless, they also clearly showed that transferring patients contributed to an increase in treatment duration. It was unclear whether the process of transfer contributed to the delay. However, this is a “real world” problem, and we should accept this finding.

We also need to consider whether poor treatment progress resulted in the transfer as the resident student left the programme.  The skill of the resident may also contribute to the need to transfer. We are all aware of residents whose treatment takes longer than most of their colleagues!

Final comments

I am aware that I have been critical of this study.  Nevertheless, it is difficult to see how we may improve this design. Perhaps, we could carry out a prospective cohort study, but this would take many years and a large amount of resources.

It is, therefore, up to you as a critical reader to decide whether this study adds to your clinical knowledge. I cannot help feeling that it is helpful, and we should warn our patients about possible delays to their treatment when we arrange their transfer.