Introduction
Colposuspension has conventionally been considered the gold standard surgical treatment for genuine stress incontinence. However, the resource use associated with this form of management is considerable, in particular the length of stay in hospital. TVT (Gynecare, Edinburgh) is an alternative intervention which promises to be as effective an intervention with a lower cost per patient. A recent randomised controlled trial of colposuspension versus TVT provided an opportunity to estimate the differential cost and cost-effectiveness of the two forms of management. This report focuses of differential hospital costs.

Methods
Data were collected prospectively, for all 316 patients undergoing surgery in the trial, on the duration of the operation, time in recovery room and overall length of hospital stay. Details of the number and grade of theatre staff, consumables, drugs and gases during each type of operation were taken from interviews with relevant staff at the main recruiting centre in the trial . Resource use was costed using appropriate UK unit costs at 1997-98 prices. Drug costs were based on British National Formulary prices; consumables were costed at manufacturers' list prices; staff costs were based on the mid-points of appropriate national pay review body scales; ward costs and theatre overheads were based on an earlier cost analysis in gynaecology, uprated for health service inflation.

Results
The mean duration of the operation was 40 minutes in the TVT group compared to 52 minutes for colposuspension. Mean time in the recovery room was also lower in the TVT group: 52 versus 96 minutes. Mean total length of stay in hospital (pre and post operatively) was 2.24 days in the TVT group compared to 6.54 days in the colposuspension group. The mean hospital cost per patient was £856 in the TVT group compared to £1,159 in the colposuspension group.

Conclusions
The hospital cost of the initial surgical treatment of GSI is likely to make up the bulk of the cost of this form of management. The mean hospital cost of TVT is lower than that of colposuspension. In order for TVT to be considered more cost-effective than colposuspension, the mean differential cost of £303 per patient needs to be related to trial data on differential outcomes. Interpretation of these differences needs to be viewed in the light of local hospital costs and average duration of hospitalisation in different healthcare systems. KW was supported by a grant from Johnson and Johnson, who also provided materials and additional support to collaborating centres.