Nasty processes produce nasty guidelines

The guideline from the National Institute for Health and Clinical Excellence (NICE) on urinary tract infections (UTI) in childhood¹ was welcomed by the BMJ.^{2 3} Most readers will assume it was based on evidence correctly analysed by medical statisticians, robustly peer reviewed, and openly debated. As this is a controversial subject, dependent more on small studies than randomised controlled trials, many will imagine that it represented consensus following wide consultation, as stated.¹ Sadly, all these assumptions are wrong.

The NICE guideline committee signed highly restrictive secrecy agreements, and its two paediatric nephrologists did not consult with the British Association for Paediatric Nephrology, whose members hold diverse views. I was a peer reviewer but was not treated as one. My first draft review identified major flaws, was supported by the association, and delayed publication by six months. However, I was allowed to see the committee's adjustments only after strong insistence, signing a secrecy document, and accepting that it would ignore my responses. The errors persist.

The guidelines were derived from an inadequate review of the literature. The authors misused statistics and reached beyond the evidence to make erroneous conclusions based on flawed logic. Some seemed to reflect opinion rather than fact. The committee's own figures showed that nitrite screening has a mean sensitivity of about 50%, so will miss half the cases, yet it¹ and Watson² advise its use unreservedly. Similarly, both promote the use of ultrasound rather than dimercaptosuccinic acid (DMSA) scans, despite their own data showing DMSAs to be much more sensitive; on average ultrasound misses half the scars. They also view DMSA as invasive even though it requires only a single venepuncture and has the radiation burden of one abdominal x ray. Both advise a temperature cut off of 38°C for investigating infants' urines without clear evidence, and both assume that a lack of evidence for prophylactic antibiotics equates to evidence against their benefit, which many paediatricians dispute.

NICE guidelines result in uniformity of practice; clinicians “are expected to follow them.”⁴ Unifying practice before a consensus emerges is absurd. Scientific debates are not resolved by secrecy and decree but by patient research and genuinely open discussion. The premature imposition of inappropriate guidelines will stifle new clinical developments. For example, our unit runs a direct access service,⁵ which seems to be reducing renal scarring rates (despite Watson's assertion that most scars are congenital²). If we are all forced into one mould based on poor analysis of evidence, we will miss the opportunity to make important advances.