Design: prospective multicentre diagnostic accuracy study with separate development and validation cohorts
Population: 601 men with lower urinary tract symptoms in 67 general practices in England and Wales
Primary outcome: prediction of urodynamic diagnoses of bladder outlet obstruction, detrusor underactivity and detrusor overactivity
Effect: validation c-index 0.82 for obstruction; 0.63 and 0.62 for underactivity and overactivity

Six hundred and one men presenting with lower urinary tract symptoms across 67 practices in England and Wales had a validated symptom score, a bladder diary, physical examination, prostate specific antigen, uroflowmetry and post-void residual measured, with invasive urodynamics as the reference standard. Models were built in 350 men and tested in a separate 251.

The bladder outlet obstruction model held up: a c-index of 0.80 on development and 0.82 on independent validation, which is the part that matters, since most prediction models lose their performance when they leave the cohort that produced them. The detrusor underactivity and detrusor overactivity models did not, at 0.63 and 0.62 on validation - barely better than a coin weighted by prevalence.

So the usable conclusion is specific. For the man whose symptoms suggest obstruction, a combination of ordinary primary care measurements carries real diagnostic information and can support starting treatment rather than waiting for a urology slot. For distinguishing an underactive from an overactive detrusor, it does not, and those are precisely the men in whom an antimuscarinic started empirically can precipitate retention. Uroflowmetry and post-void residual are not universally available in Indian general practice, and bedside ultrasound residual is the practical substitute.

A symptom score, bladder diary, examination, prostate specific antigen, flow rate and residual volume together predict obstruction usefully.
The same combination does not reliably separate detrusor underactivity from overactivity.
Be cautious starting an antimuscarinic empirically where obstruction has not been excluded.
Ask for a bladder diary before the next visit - it costs nothing and carries diagnostic weight.
Where uroflowmetry is unavailable, a bedside ultrasound post-void residual is the practical substitute.

Why it matters

It moves part of the urology assessment into the consultation that already happened, for the commonest presentation in older men.

The statistics, in plain English

A c-index of 0.82 on independent validation means the model ranks two men correctly about four times in five - genuinely useful. At 0.62 the model is close to guessing, and the gap between the two is the whole message. Validation in a separate cohort is what distinguishes this from the many prediction models that look impressive only in the data they were built from.