One of the challenges in delivering an journal like E&P is getting the balance right between articles on conditions, tests or skills that you, the readers, will need on a daily basis, and those which you might need only once in a while but have other appeal. This other appeal may have a number of aspects. They may simply be interesting, or contain important information, or they may stimulate other serendipitous thoughts.

This issue has a range of articles which I hope will do all of these for you. This month's editor's choice is a thought provoking paper about the measurement of urine catecholamines by Daniel Erdelyi, Martin Elliott and Bob Phillips (see page 107). You'll probably suspect a little bit of ‘Old Boys Club’ with this choice – me focusing on another paper co-written by a fellow editor, but let me try to explain what this paper does for me. First, it gives me a good, but brief, review of neuroblastoma – which I can always do with. Second, it gives me something to take back to my local lab – in short, we should be doing spot urinary catecholamines, since they are as helpful as a 24-h collection, and an order of magnitude more convenient to collect. It seems that it is only inertia – and perhaps distance from the practical challenge of having to collect that much urine – that stops us from moving to spot urine collection. Third, I should stop worrying about phaeochromocytoma. Which, of course, is an overstatement, but I take it as particularly telling that during the earlier drafts of this paper the tertiary oncologist authors responded to questions about phaeo along the lines of ‘Oh; it hadn't occurred to us that you'd seriously consider that as a differential – it's so rare, and the clinical features are so clear‘.

Where's the serendipity then? That's in table 2 of the paper, and I'd strongly suggest taking a few minutes to digest what it's saying. What the authors have done is apply this test to a series of populations or clinical scenarios with different prevalence of the condition, and done some simple arithmetic to look at what a positive or negative test actually means in terms of subsequent likelihood that the patient has neuroblastoma. This isn't a lesson that can be repeated too often. I run a short teaching session based on (meaning ripped off from) an article from Bad Science, Crystal Balls and Positive Predictive Values, which you can find here: http://bit.ly/CrystalBalls. I did the pencil and paper version of this myself probably five times and was astounded at the outcome each time, and find that people I take through the exercise are equally shocked; that good tests, badly used can give frighteningly bad results. Take a look back at the top line of table 2 and think about what this means in terms of what you say to the patient's family: ‘Um, I think your child has a possibly pretty unpleasant diagnosis. But I'm only two-thirds sure that my test result, which is reading positive, actually is positive’.

We have some quite eclectic other papers for you this month. There's a discussion about how we should be assessing analgesia in trials in neonates (see page 112). There's a review of the long-term care of patients after liver transplantation (see page 82). There's a paper on how to treat malaria in the UK – which, for our international readers, should probably be read along the lines of ‘in the developed world’ (see page 87). On this subject, I've worked with generations of registrars who have found our approach to treatment of malaria genuinely perplexing, given what they've learnt in their country of origin; this short article helps frame treatment in a UK context. Starkey and Sammons review the pharmacology behind sedation (see page 101) in a great article to read alongside the recent publication of National Institute for Health and Clinical Excellence guidelines on sedation in children which can be found here: http://www.nice.org.uk/CG112. And I can usually find something in the issue which I've never even heard of before; this month it is specific immunotherapy to prevent progression of allergic rhinitis (see page 91).

Finally we've an interesting Picket paper this month about an important area – which has, in the past been relatively unpopular to study (see page 120). None of the three antiepileptic medicines used to treat absence seizures is high cost, and therefore some of the drivers to conduct a randomised controlled trial have been missing. This is the sort of thing that the Medicines for Children Research Network in England is intended to address – although this is not their study. The outcome? Well, read the Picket, and perhaps the original paper too. It's changing my practice.