In a man with lower urinary tract symptoms, what level of PSA requires referral for further investigation (usually biopsy)? An otherwise excellent symposium report entitled ‘When should patients with BPH be referred?’ by Roger Kirby in The Practitioner (2008;252:29–33) rather ducks this key question, saying only that ‘PSA elevated for age’ is a ground for referral. That rather assumes there is consensus on age-related PSA reference ranges, which there is not. In contrast, Roddam et al. (BJU International 2007:100:514–7) studied the effect of lowering the PSA threshold for further investigation in men under 70 from 4 ng/mL to 2 ng/mL. Their data shows that this would markedly increase the number of referrals (from 110 to 230 per 1000 men tested) with a modest increase in the cancer detection rate (36–58 per 1000). Including a reflex test for free:total ratio or complexed PSA does little to improve the predictive value. The authors also point out that the extra cancers detected are likely to be clinically-localized and there is no evidence that treating them will improve disease outcome. Does not leave us much wiser, but at least we are better informed.

Have you ever finished a presentation at the weekly Grand Round crammed with data and evidence-based conclusions, only to be completely destroyed by some ancient clinician in the back row who starts off ‘In my experience…’ and then states the exact opposite of your carefully crafted message? Me too. So, it was immensely gratifying to find a paper in Medical Education (2007:41:965–7) entitled ‘The majority of bold statements expressed during grand rounds lack scientific merit’. Linthorst et al. studied bold statements made by senior medical staff during grand rounds at an Amsterdam teaching hospital over a four-month period. Examples of the sort of thing included ‘Crohn's disease always exacerbates during pregnancy’ and ‘Legionella pneumonia never causes pleuritic pain’. The statements were then checked against the evidence base. In only 32% of cases were the statements supported by evidence. There was no relevant literature to confirm or refute the statement in 16%, and in the remaining 52% of cases, the literature actually proved the statement to be wrong! The authors point out that exotic expert opinions expressed by top doctors at grand rounds are usually assimilated by junior staff as medical fact, and advise a much more critical approach. To repeat Deming's immortal words: ‘In God we trust. All others bring data’.

Instructions to journal authors (including those in the Annals) generally focus strongly on how to format articles, and have little to say about how to make them easier to read. Readability can be estimated using a predictor score such as the Flesch Reading Ease scale, which is buried in your Microsoft^® Word software (Tools > Options > Spelling and Grammar > Show Readability Statistics). The Flesch score runs from 0 to 121, and the higher the score, the more readable the document. It's based on word and sentence length – shorter words and shorter sentences make for better readability. Anything under 30 is deemed ‘very difficult’ and said to require a college education to read. Standard readability level is between 60 and 70, and insurance documents are normally pitched about 40–50. In the January issue of Br J Surg (2008;95:119–24), Hayden looks at the readability of articles in the Journal. Readability is better in manuscripts from authors who do not have English as their first language (either because they use shorter and less complex sentences or they get professional help). Hayden also shows convincingly that the editing process shortens manuscripts by about 8% and makes them significantly more readable (sound of collective ‘phew!’ from journal editors…). The mean Flesch score of published manuscripts was 23.8, but the range was from 8.6 (which is gobbledegook territory) to a relatively accessible 40.4. I'll tell you what this scores when I finish it…

The Br J Gen Pract (2008;58:62–3) begins what looks like a very useful series entitled ‘Top Tips in 2 Minutes’. The brief to authors is to imagine that they are trapped in a lift with a bunch of GPs and have two minutes before they are released to impart the essential pearls of wisdom on a subject dear to their heart. (Piscator doubts that his first instinct in that situation would be to deliver a lecture, but we'll let that pass.) The result is published on a single page. It kicks off with a concise summary of chronic kidney disease by John Firth from Cambridge, which is well worth having within reach.

Piscatrix is on well-deserved holiday. This column comes to you from our emergency backup Piscator, stored in a dusty warehouse in the Midlands and deployed when required. Its Flesch score is 43.3, so a second career in the insurance industry is on the cards.