OBJECTIVES: The use of mapping algorithms has been suggested as a second-best solution for estimating health-state utility values when no generic preference-based measure is incorporated into a study. However, predictive performance of these algorithms may be variable and hence assessing their external validity in different settings before application is of utmost importance. This study assessed the external validity and generalizability of existing mapping algorithms for predicting Child Health Utility 9-Dimension (CHU9D) utility values from the Pediatric Quality of Life Inventory (PedsQL^TM) in a nationally representative sample of children from the Australian general population living with or without disabilities/medical conditions. METHODS: A validation cohort of 6,623 children (n=3,376 aged 10-11 years, n=3,247 aged 14-15 years) from the Longitudinal Study of Australian Children was utilised of which 294 had a disability/medical condition. Three published mapping algorithms estimated using robust MM, generalised linear and ordinary least squares regression models, respectively, were assessed. Predictive accuracy was evaluated using mean absolute error (MAE) and mean squared error (MSE) estimates. Absolute agreement between observed and predicted CHU9D utilities was tested using intraclass correlation coefficients (ICC). RESULTS: Values for the MAE (0.1528-0.3562), MSE (0.0365-0.0730) and ICC (0.434-0.491) for all validations were within the range of published estimates. The algorithms performed better amongst 10-11 year-olds (MAE: 0.1468-0.2194; MSE: 0.0365-0.0700) compared to 14-15 year-olds (MAE: 0.1588-0.2159; MSE: 0.0428-0.0730). Across all ages, predictive accuracy for the algorithms was stronger amongst those without medical conditions/disabilities relative to those with disabilities/medical conditions (MAE: 0.1468-0.1588 vs 0.2159-0.2194; MSE: 0.0365-0.0428 vs 0.0700-0.730). The MM algorithm performed best in all cohorts. CONCLUSIONS: : The mapping algorithms have acceptable predictive accuracy and appear to perform better in younger cohorts and in children and adolescents without disabilities. We recommend validating these and other PedsQL-to-CHU9D algorithms in cohorts with a larger sample of younger people with disabilities/medical conditions.