The purpose of this study was to investigate the effect of childhood cancer and its treatment on self-reported health-status in 10,189 adult survivors of childhood cancer in Britain. Age-and sexadjusted scores on the SF-36 Mental and Physical Component Summary scales (MCS, PCS, respectively) were compared between survivors and UK norms, and between subgroups of survivors, by multiple regression. Survivors had comparable scores to UK-norms on the MCS scale (difference (D) 5 20.1, 99% CI: 20.5, 0.3). The difference in scores between survivors and UKnorms on the PCS scale varied by age (p heterogeneity < 0.001). Young survivors (16-19 years) scored similarly to UK-norms (D 5 0.5, (21.1, 2.2), whereas the age groups of 25 and older scored statistically and clinically significantly below UK-norms (all p-values < 0.0001), with Ds ranging between 22.3 (23.5, 21.2) and 23.7 (25.0, 22.4). Survivors of central nervous system (CNS) and bone tumors scored significantly (p-value at all ages <0.003) below UK-norms on the PCS scale. Specifically, these survivors were substantially more limited in specific daily activities such as, for example, walking a mile (40, 63%, respectively) when compared to UK-norms (16%). In conclusion, childhood cancer survivors rate their mental health broadly similarly to those in the general population. Survivors of CNS and bone tumors report their physical health-status to be importantly below population norms. Although self-reported physical health is at least as good as in the general population among young survivors, this study suggests that perceived physical health declines more rapidly over time than in the general population. ' 2007 Wiley-Liss, Inc.Key words: SF-36; health-status; childhood cancer survivors; epidemiology With the improvements in survival rates after childhood cancer, 1 in every 1,000 young adults currently has a history of childhood cancer, and this proportion is likely to increase.1 This increase, the potential for late effects, and the many decades of potential life ahead for most survivors means that it is critical to investigate the health-status of adults who have survived childhood cancer. Although many previous studies have investigated health-status by objective measures of health, relatively few studies have investigated self-reported health-status among survivors of childhood cancer. Investigating self-reported health-status is important because it may identify needs that may not be identified by objective measures of health such as, for example, a standard medical evaluation. Results from most studies 3-9 conducted so far indicate that the self-reported health-status of survivors is apparently broadly good. However, many earlier studies had design problems such as small sample size, lack of appropriate control group or use of a nonrepresentative sample of survivors. Because of these limitations the interpretation of much previous work may be questioned. We conducted an analysis using data from a British cohort study that was population-based and substantially la...