Tuesday 9 April 2013

Don't believe the hype

A story in the dental press caught my attention today, quoting research from America that has shown that injections of dental anaesthetic in young children might stop wisdom teeth developing in the future.

Wow, that seems pretty amazing doesn't it? That a simple injection might stop the wisdom teeth developing, reducing the risks of pain, infection and surgical complications in the future. The authors seem pretty clear that the difference is significant, but I've never heard of anything like this before and I'm sceptical.

Now before I continue I should say that I haven't got access to the full article which I would need if I was going to do a full critique, but just reading through the reports it's pretty easy to pick out some flaws in the research.

The researchers looked back at records of children who had a dental x-ray at 7 years old or over and examined them to see if the wisdom teeth were developing. Already I'm a bit concerned as before the age of 10 the wisdom teeth are unlikely to appear and up until the age of 12 I wouldn't be confident that they were definitely missing.

Then they divided the children into two groups depending on whether they'd had a dental injection in the lower jaw between the ages of 2 and 6 and compared them to see which had the most wisdom teeth. I guess they could only know that if they'd had treatment at the same clinic. What if they'd had treatment elsewhere they didn't know about? Asking the patient or parent is not reliable either (though they don't appear to have done this).

The total number of patients was 220, giving 439 sites where wisdom teeth might develop, so that would be one on either side of the lower jaw for each patient (there's a missing one somehow!). 63 sites had received injections (comparison group) and 376 hadn't (control group), not very equal groups. Comparing 'sites' rather than 'patients' could confuse the results, as the left and right sides are not independant. So if the left wisdom tooth is missing it's more likely that the right one is too. Hopefully in the full results they explain if or how they've allowed for this.

This is the 'headline' though, "In the control group, 1.9% of the sites did not have X-ray evidence of wisdom tooth buds. In contrast, 7.9% of the sites in the comparison group – those who had received anaesthesia – did not have tooth buds. The comparison group was 4.35 times more likely to have missing wisdom tooth buds than the control group." They've applied some statistical analysis and found that this is significant, though I can't see what analysis they've used, if there was a sample size calculation or what level of significance, again I hope this is explained in more detail in the full paper. Looks pretty good though doesn't it, 4.35 times more likely to have a missing tooth if there's been an injection?

How about the numbers though? 1.9% is just 7 sites in the control group that didn't have a developing wisdom tooth. 7.9% of the comparison group is 5 sites. Remember that's 'sites' not 'patients', so this could represent as few as 4 and 3 patients! These numbers are pretty small and I'd say too small to make statistics meaningful, though I'm surprised it's not higher considering the age of the children when xrayed.

I don't think this is a good enough piece of research to change habits and make people start injecting small children in the hope their wisdom teeth might not develop. In any case, I'd like to see an argument that it's desirable have them missing, what happens if other teeth fail to develop or are lost through dental disease or trauma? But what it should do is stimulate more research. Larger groups and prospective trials or follow up studies would be better. I don't think the stats are good enough to support a study that would randomise to two groups, one with injections, one without, and long term monitoring with xrays though, as well as the eithical issues that would go with administering injections and xrays to children with no dental disease!

I think it's a good reminder that when we see a piece of research that tells us something we want to hear, or even something we don't (how often do we hear conflicting reports about whether wine or chocolate is good or bad for us) that there is often much more to it than that. A little probing into the facts and figures can often turn up flaws in the findings and we must be cautious not to believe the headline straight away.