Expert opinion is that evidence is sufficient to recommend that urodynamic testing is not required before initiating conservative treatment in women with urinary incontinence.

• The National Institute for Health and Clinical Excellence (NICE) found 25 studies that compared diagnosis based on history alone with urodynamic findings on cystometry. Eleven of these studies (n = 2447) allowed sensitivity, specificity, and positive and negative predictive values to be calculated. Data were not pooled because of heterogeneity among studies. Instead, confidence intervals for each value (sensitivity, specificity, and positive and negative predictive values) were calculated. The median values and ranges of results were then calculated. Unfortunately, the studies were all considered to be of poor quality in terms of diagnostic accuracy because of unblinded urodynamic testing. Within the limitations of the studies, NICE concluded that:
• Urodynamic investigations offer little added diagnostic value:
• If a women reports pure stress urinary or pure urgency incontinence (that is, not mixed incontinence), the probability of finding urodynamic stress incontinence plus detrusor overactivity is low.
• If a woman does not report pure urgency incontinence, the probability of finding detrusor overactivity is low (10%).
• If a woman does not report pure stress incontinence, the findings are less clear-cut, mainly because of the poor quality of the trials and the lack of consistency between the negative predictive values. Between 15% and 51% of women who do not report pure stress incontinence will be found to have urodynamic stress incontinence on testing.
• A Health Technology Assessment undertook a systematic review and identified 129 studies from the 6009 articles reviewed. They found that a large proportion of women with stress urinary incontinence can be identified in primary care from clinical history alone. Pooled sensitivity and specificity values for diagnosis of urodynamic stress incontinence from clinical history found a high sensitivity (0.92) and a lower specificity (0.56) [Martin et al, 2006].