PURPOSE: To examine parent-child agreement regarding a child's health-related quality of life (HRQOL) among three health status groups. METHODS: Parent-child agreement was evaluated for three health status groups of a population-based sample: (1) children with mental health problems (N = 461), (2) children with physical health problems (N = 281), and (3) healthy controls (N = 699). The KIDSCREEN-27 was used to assess HRQOL. The children were 9-14 years of age. RESULTS: Intraclass correlation coefficients were mostly good across all HRQOL scores and health status groups. This relatively high level of agreement was also reflected by the following findings: first, the AGREE group was the largest in three out of five HRQOL subscales in all health status groups; second, when disagreement occurred, it was often minor in magnitude. Despite this relatively high level of agreement, the means of self-ratings were significantly higher for all HRQOL scores and health status groups than the means of proxy ratings. These higher self-ratings were especially pronounced among children with mental health problems in certain HRQOL domains. CONCLUSIONS: Even though the level of parent-child agreement regarding a child's HRQOL is relatively high, it should be considered that children (especially those with mental health problems) often report better HRQOL than their parents. It is, therefore, highly recommended that both proxy- and self-ratings are used to evaluate a child's HRQOL comprehensively.