With the growth in the use of health economic evaluation to inform healthcare resource allocation decisions, the challenges in applying standard methods to child health have become apparent. A unique limitation is the paucity of child-specific preference-based measures. A single, valid, preference-based measure of utility that can be used in children of all ages does not exist. Thus, the ability to derive a QALY for use in cost-utility analysis (CUA) is compromised. This paper presents and discusses existing and novel options for deriving utilities for paediatric health states for use in CUAs. While a direct elicitation may be preferred, a child's ability to complete a standard gamble or time trade-off task is hampered by cognitive and age limitations. The abstract notions contained in indirect instruments such as the EQ-5D and Health Utilities Index may also pose challenges for young children. Novel approaches to overcome these challenges include the development of age-appropriate instruments such as the EQ-5D-Y, the development of new child-specific utility instruments such as the Child Health Utility-9D and the re-calibration of existing adult instruments to derive preference weights for health states from children themselves. For children aged <6 years, researchers have little choice but to use a proxy reporter such as parents. While parents may be reliable reporters for physical activity limitations and externally manifest symptoms, their ability to accurately report on subjective outcomes such as emotion is questionable. Catalogues of utility weights for a range of conditions are increasingly becoming available but retain many of the same limitations as valuing health states from children or from proxies. Given the dynamic relationship in quality of life (QOL) between family members when a child is ill, it seems appropriate to consider a 'family perspective' rather than an individual perspective in child health state valuation. In a collective approach, health state utilities derived from multiple family members may be combined mathematically. Alternatively, in a unitary approach, a single utility estimate may be determined to represent the family's perspective. This may include deriving utilities through parent-child dyad estimation or by using a household model that combines the utility weights of the patient and family members, incorporating reciprocal QOL effects. While these various approaches to child health state valuation represent novel research developments, the measurement challenges and threats to validity persist. Given the importance of non-health benefits to child health, especially in the domains of education and public policy, it may be worthwhile to consider an approach that allows incorporation of externalities to produce a cost-benefit analysis. The use of discrete-choice methods to assess willingness to pay for novel child health interventions holds promise as a means to produce meaningful economic evidence. Regardless of the approach taken, the highest degree of methodological rigour is essential. The increasing attention being paid by health economic researchers to the measurement challenges of paediatric health state valuation can only increase the value of child health economic evidence for decision making.