To compare the performance and cost-effectiveness of the key absorbent product designs to provide a more solid basis for guiding selection and purchase. Also to carry out the first stage in the development of a quality of life (QoL) instrument for measuring the impact of absorbent product use on users' lives.Three clinical trials focused on the three biggest market sectors. Each trial had a similar crossover design in which each participant tested all products within their group in random order. SETTING, PARTICIPANTS AND INTERVENTIONS: In Trial 1, 85 women with light urinary incontinence living in the community tested three products from each of the four design categories available (total of 12 test products): disposable inserts (pads); menstrual pads; washable pants with integral pad; and washable inserts. In Trial 2a, 85 moderate/heavily incontinent adults (urinary or urinary/faecal) living in the community (49 men and 36 women) tested three (or two) products from each of the five design categories available (total of 14 test products): disposable inserts (with mesh pants); disposable diapers (nappies); disposable pull-ups (similar to toddlers' trainer pants); disposable T-shaped diapers (nappies with waist-band); and washable diapers. All products were provided in a daytime and a (mostly more absorbent) night-time variant. In these first two trials, the test products were selected on the basis of data from pilot studies. In Trial 2b, 100 moderate/heavily incontinent adults (urinary or urinary/faecal) living in 10 nursing homes (27 men and 73 women) evaluated one product from each of the four disposable design categories from Trial 2a. Products were selected on the basis of product performance in Trial 2a and, again, day time and night-time variants were provided. The first phase of developing a QoL tool for measuring the impact of using different pad designs was carried out by interviewing participants from Trials 1 and 2a.Product performance (e.g. comfort, discreetness) was characterised using a weekly validated questionnaire. A daily pad change and leakage diary was used to record severity of leakage, numbers of laundry items and pads. Skin health changes were recorded weekly. At a final interview preferences were ranked, acceptability of each design recorded, and overall opinion marked on a visual analogue scale (VAS) of 0-100 points. This VAS score was used to estimate cost-effectiveness. In addition, a timed pad changing exercise was conducted with 10 women from Trial 2b to determine any differences between product designs.Disposable inserts are currently the mainstay of management for lightly incontinent women (Trial 1) and they were better for leakage and other variables (but not discreetness) and better overall than the other three designs. However, some women preferred menstrual pads (6/85) or washable pants (13/85), both of which are cheaper to use. Washable inserts were worse both overall and for leakage than the other three designs (72/85 found them unacceptable). For disposable inserts and disposable diapers, findings from the community (Trial 2a) and nursing home trials (Trial 2b) were broadly similar. Leakage performance of disposable inserts was worse than that of the other designs for day and night. Pull-ups were preferred over inserts for the daytime. The new T-shaped diaper was not better overall than the traditional disposable one. However, there were important differences in performance and preference findings for men and women from both trials. Pull-ups (the most expensive) were better overall than the other designs for women during the day and for community-dwelling women during the night. Although disposable diapers were better for leakage than disposable inserts (the cheapest), women did not prefer them (except in nursing homes at night), but for men the diapers were better both overall and for leakage and were the most cost-effective design. No firm conclusions could be drawn about the performance of designs for faecal incontinence. Nursing home carers found pull-ups and inserts easier to apply (in the standing position) and quicker (in the pad change experiment) than the diaper designs; the ability to stand was associated with preference for pull-ups or inserts. The T-shaped diaper was not easier or quicker to change than the diaper. The washable products (Trial 2a) gave diverse results: they were better for leakage at night, but were worse overall for daytime than the other designs. Three-quarters of the women (27/36) found them unacceptable, but nearly two-thirds of men (31/49) found them highly acceptable at night. Findings from the two community trials (Trials 1 and 2a) showed that there were many practical problems in dealing with washable products but, together with the less effective and less expensive products, such as menstrual pads, they were more acceptable at home (and, in the case of washables, at night). This suggests that cost-effective management may involve combining products by using more effective (for a given user) but more expensive designs (e.g. pull-ups) when out and less effective but less expensive designs when at home. The interviews examining the impact of pad use on QoL provided themes and domains that can be further developed into a tool for further evaluation of absorbent products.This study showed that there were significant and substantial differences between the designs of absorbent products and for moderate/heavy incontinence some designs are better for men/women than others. There was considerable individual variability in preferences and cost-effective management may best be achieved by allowing users to choose combinations of designs for different circumstances within a budget. Further research is needed into the feasibility of providing choice and combinations of designs to users, as well as into the development of more effective washables and of specifically male disposable products. QoL measurement tools are needed for users of absorbent products, as are clinical trials of designs for community-dwelling carer-dependent men and women with moderate/heavy incontinence.