Lies, Damned Lies and Statistics…? by Penny McWilliams

The use of general anaesthetic for extraction of children’s teeth has reduced very considerably in Scotland in the past 20 years – quite right too, I’m sure most people would agree.

The Scottish government has arguably led the world in funding the Childsmile programme, which is intended to tackle the fundamental causes of poor oral health in children as early as possible, by providing multiple educational and preventative interventions in community and school settings.

Penny  3

Like most large scale health programmes and initiatives these days, it is accompanied by HEAT targets, one of which is for a reduction in the number of elective hospital admissions for tooth extractions for children aged under 3 years.

All very sensible and intuitive, and given the scale of the funding associated with Childsmile, it is hardly surprising that the impact on oral health should be monitored over time to see if the various initiatives have been effective. It would be fair to say that dental public health experts worldwide are interested in the long term success or otherwise of Scotland’s Childsmile project, to see if they should commissioning or implementing something similar. Changing people’s behaviours as opposed to increasing their levels of knowledge is notoriously difficult; looking back on issues such as smoking and drink-driving, we all know that changes have gradually occurred, but they have taken decades of health education and health promotion effort to bring about.

Penny 4The much improved access to dental care across D&G region since the mid-noughties, combined with the Childsmile activities by dental primary care and health improvement teams have been very successful in identifying those families most at risk of poor oral health. And we are much more successful than previously at providing appropriate dental treatment for young children with active tooth decay.

So paradoxically, the average age of many children’s first contact with a dentist has almost certainly come down, as many families now register with a dental practice for care. And the numbers of children aged 3-5 years being admitted to hospital for tooth extractions under general anaesthetic has certainly come down since 2003 rather than gone up, as the graph shows.

Chart 1 – DGRI Hospital admissions for tooth extractions in children aged 0-5 years old in 2004, 2009, 2013, and 2014 

Penny 1

But as you can see, the actual numbers of children under three admitted for tooth extraction are very low – only a couple a year on average. And some of these need an extraction because of trauma i.e. an accidental fall has damaged their front teeth, rather than tooth decay. And the average numbers admitted per annum have not really come down since 2009.

Very young children with painful tooth decay often now have much earlier contact with healthcare staff who successfully identify that they have dental treatment needs.

And those children for whom a tooth extraction is genuinely required, which can usually only be achieved by admission to hospital and use of a general anaesthetic, are arguably now more likely to get referred promptly.

If the whole point of HEAT targets from a government perspective is that ‘what gets measured is what gets done….‘ , where does that leave us when it comes to trying to achieve this particular target?

One way to achieve the HEAT target would be to leave the waiting times for admission long – children might be under three years old at the time of the decision to extract the tooth, but even with 18 week waiting time guarantees, most of them will be over three years old by the date of admission. Should we postpone provision of treatment in hospital for young children needing tooth extractions, because it would help us achieve the HEAT target? I don’t think anyone would advocate that, but could failure to achieve the reductions in numbers of hospital admissions be used to imply that oral health is not improving in Dumfries & Galloway region? Or NHS Dumfries & Galloway is not implementing the Childsmile programme properly?

We have already created local care pathways to ensure that the alternatives to extractions of deciduous teeth are available for young children, including active dental prevention strategies, and provision of more specialised paediatric dental treatment services. And because the risk associated with the use of general anaesthetic is very much higher than for routine dental treatment, dental general anaesthetic services are delivered in accordance with all of the currently available clinical guidelines, in as safe an environment as we can achieve.

I think it was Winston Churchill who talked about ‘lies, damned lies and statistics…’ and looking at the available figures general anaesthetics for dental extractions in children and Dumfries & Galloway over a period of years is pretty complicated. You can make an analysis of the statistics apparently illustrate almost anything you like.

Chart 2 below shows child admissions for general anaesthetic for dental extractions across Dumfries & Galloway region by quarter for 2013-14, arranged by age group. 

Penny 2

I could certainly argue for hours about what the figures and any mapping of overall trend does or does not tell us, based on this.

I’m sure that the original intent of the HEAT target was to see if oral health in very young children improves over time, particularly as it is well-known that children with complex physical, medical and social needs are at much higher risk of developing tooth decay. One can also assume it was intended to ensure that health authorities commissioned adequate local primary care dental services for families with young children.

But HEAT targets are very high-profile reporting measures, and failure to achieve them can very easily be misinterpreted, even by people who are entirely sincere and well-intentioned. Whatever the reasoning behind it, this one could become an unfortunate example of the misuse or misinterpretation of statistics.

Penny McWilliams is Director of Primary Care Dental Services for NHS Dumfires and Galloway

One thought on “Lies, Damned Lies and Statistics…? by Penny McWilliams

  1. A classic example of how targets can lead to unintended consequences and may in fact be detrimental to clinical care.
    Similar issues are being faced by the regional day centres who are being compelled to produce “outcome measures” of their effectiveness to justify continuing funding support in the new climate of social work / healthcare integration and joint funding.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s