OBJECTIVE: To determine whether the behaviour of participants based on perception of treatment group in a randomised trial contributed to clinical outcome. DESIGN: A double blind randomised controlled trial of the effect of daily application of SPF 17 broad spectrum sunscreen cream (or placebo) on solar keratoses. SETTING: A rural city in Victoria, Australia. Residents aged 40 years or over were invited by letter to attend for a skin cancer screening check. Of these, 588 people with between one and 30 solar keratoses enrolled in the trial and 431 completed the trial, which extended over a six month period that included summer. Participants' perceptions of their treatment allocation, adherence with the treatment regimen, adoption of other sun protection behaviours, side effects, and perceptions of change in condition were measured at two monthly intervals. RESULTS: There were no significant differences between those who completed the study and those that did not for sex, age, treatment group, skin type, number of solar keratoses or correct perception of treatment group. Thirty per cent of those completing the study correctly guessed their treatment allocation, and people were just as likely to be right as to be wrong when they stated their opinion about their treatment allocation (z = 1.04; p = 0.15). Study group, skin type, amount of time spent outdoors, presence of side effects, perceptions of change in skin condition did not significantly predict correct perception of treatment allocation. Multivariate analysis of variance indicated that adoption of other sun protection and adherence with cream use were not significantly affected by actual treatment allocation, correct perception of treatment allocation nor by their interaction. Poisson regression analysis showed a significantly lower difference ratio of solar keratoses in the sunscreen group compared with the placebo base cream group (OR 0.55; CI = 0.46, 0.64), and for women compared with men (OR = 0.76; CI = 0.63, 0.93) but no independent effect of any of the indices of other sun protection or adherence. CONCLUSIONS: A sufficient level of commitment to study procedures was achieved, so that trial participants did not adopt other behaviours that affected treatment outcomes. It is recommended that the potential threat to validity posed by the behaviour of participants be recognised at an early stage in planning of clinical trials, so that strategies to deal with this can be integrated into study protocols.