A smartphone app improves orthodontic patients oral hygiene and gingival health?

This is an interesting new study on the effectiveness of a UK based smartphone app used to manage oral hygiene frequency and duration. This showed that it worked!

There have been several studies on using smartphone technology to improve our patient’s oral hygiene.  I have reviewed a couple of these before. Generally, these studies show that there is some improvement in oral hygiene in the short term.  This new study looked at an established smartphone app called Brush DJ. It is recommended by the UK National Health Service.

A team from Iran did this study. The British Dental Journal (online) published the paper.

What did they ask?

They wanted to find information on

“The efficacy of a smartphone app compared to conventional methods on improving the oral hygiene of orthodontic patients in fixed appliances”.

What did they do?

They did a single-blind randomised controlled trial with a 1:1 allocation.  The PICO was

Participants: Orthodontic patients who received their orthodontic appliances 12 weeks after entering the study. They also had to have access to a smartphone.

Intervention: The Brush DJ app (I will give more details later).

Control: Treatment as usual with oral hygiene instruction reinforced with videos and leaflets.

Outcomes:  The primary outcomes were the Plaque (PI) and Gingival indices (GI). Secondary outcomes were questionnaire data on duration and frequency of toothbrushing.

Brush DJ is a smartphone app that was developed by a UK based National Health Service dentist. The app is essentially a timer that plays two minutes of music from the device or a streaming service.  It is designed to make toothbrushing more enjoyable and contains other useful information on oral health.  Further details are on this website: Brush DJ.

They did a sample size calculation that was clear and relevant. This showed that they needed to enrol 60 participants in each group.  They used pre-prepared block randomisation. But I could not find any information on how the sequence was generated. Importantly, they did not include information on allocation concealment.

They collected the data at baseline (they stated that this was before the intervention). Then at 4 weeks (T1), 8 weeks (T2) and at 12 weeks (T3).

They analysed the data using univariate and multivariate statistics.

What did they find?

They enrolled 120 participants in the study.  I spotted that the gender was not balanced in the two groups as there were more males in the control group than the Brush DJ group.

There were no significant differences between for PI between the groups at the start of the study.

One important finding was that there was a reduction in plaque accumulation in the Brush DJ group while remaining the same in the control group.  At the end of the 12 week study period, the PI for the Brush DJ group was 67.84 (SD= 12.33), and for the control group, it was 80.82 (SD= 10.05). The authors did not provide 95% CIs.

When they looked at GI, they found that the control group had a poorer GI than the Brush DJ group.  We need to remember that there were more males in this group than the Brush DJ group. This means that it is likely that the groups were unbalanced.  As a result, they analysed this data with repeated measures ANCOVA. This statistical test takes into account the effects of covariates, for example, gender and start values.  This showed a reduction in GI in the Brush DJ group and an increase in the control group.

Finally, when they looked at the duration and frequency of brushing. The authors found that these were not significantly different.

What did I think?

I thought that this study was fascinating and very relevant.  It was great to see a trial looking at this crucial area. Furthermore, the trial showed some positive effects of the intervention.

However, I would be keener and more effusive about these results if the paper was easy to read. I had to work really hard at interpreting the method and data.

Notably, the authors, and the journal, did not follow the CONSORT guidelines for the presentation of clinical trials.  This was disappointing.  I was surprised that the journal did not follow the recommendations.  The authors analysed a great deal of data, and this was reasonably clear. However, several times they drew attention to differences between the groups, and then stated that they were not statistically significant. This is confusing and plainly wrong.  If a difference is not statistically significant, there is a high probability that it occurred by chance. As a result, authors should not claim there are differences.

Furthermore, it was not clear if the clinicians applied the interventions at the bonding appointment.  The authors stated that the baseline was “prior to the intervention”.   As a result, it was not clear if the intervention started at bond up or later in the treatment process.  This is important because if this was the case, there needs to be information on the orthodontic treatment stage to assure us that the groups were balanced for this critical variable.

Other issues were that the trial was done for only 12 weeks of orthodontic treatment.  I would like to see a similar study run for the duration of treatment. This is because the novelty of using an app may wear off in the later stages of treatment.

Final thoughts?

I am really keen on the use of free technology to help our patients.  It is great to see that this was developed by a general dentist and offered to patients at no cost.  I feel that the results suggest that it affects tooth brushing and gingival health for orthodontic patients. As a result, I would use this for my patients, as it is a very low-risk intervention.

However, I would be more convinced if the publication was more precise and easier to interpret. Unless I am starting to lose my critical appraisal ability?