See related article on page 744 Prior to the publication of the Gross Motor Function Classification System (GMFCS) in 1997,1 functional limitations in children with cerebral palsy (CP) were often graded using terms such as 'mild', 'moderate', and 'severe'. Indeed, within our gait laboratory we used the Hoffer descriptors of 'community ambulators', 'household ambulators', 'non-functional ambulators', and 'non-ambulators'; described for patients with myelomeningocele, basically because of a lack of any valid alternative system. The development of the original GMFCS 5-point ordinal scale was, therefore, very much welcomed and has changed considerably our means of communicating information on trunk and lower limb function in children with CP. It has provided clinicians with a simple yet clinically useful tool that, among many other things, can assist with clinical decision-making and set more realistic goals within many aspects of care. I was therefore delighted to read the article in this issue by Palisano and colleagues on the expanded and revised GMFCS (GMFCS-E&R). The development of a 12- to 18-year age band adds to the four previously described age bands and thus provides a complete picture of functional ability throughout childhood in this population. The main strength of this and the original GMFCS publication1 is that the descriptors for gross motor function have been drawn up using prospective longitudinal data. The comprehensive methodology also lends further weight to the validity of the GMFCS-E&R. It is noteworthy that the authors have incorporated aspects of the conceptual framework of the International Classification of Functioning, Disability and Health (ICF) into the new age band and have also taken the opportunity to revise the original descriptors for the 6- to 12- year age band. Environmental and personal factors have been incorporated into the new descriptors and, within the user's instructions, emphasis is yet again placed on usual performance (what a child or young person does). Phrases such as 'may participate' or 'may use' (particularly within the descriptors for levels I and II) intrinsically suggest personal choice and, greater variation in mobility methods, whether for environmental or personal reasons (particularly within levels II and III) is realistic. While feedback from the Delphi survey raised the concern that the GMFCS-E&R may affect the ability to compare future research with previous findings, the extent to which environmental and personal factors can have an impact on methods of mobility and recreational activity is such that they needed to be included in any revised system. The new descriptors for the 6- to 12-year age band are no doubt more comprehensive and, for clinicians who may have deliberated between adjacent levels in the past, make classification that little bit more straightforward for certain children. From my own experience, they are more likely to shift these same children into a higher level (i.e. IV to III and III to II) and thus present some potential for bias in future comparisons with previous work. The additional descriptors in the GMFCS-E&R that may have this influence include: for children in level II '… in the community children may walk with physical assistance, a hand-held mobility device, or use wheeled mobility when travelling long distances…' and for children in level III '… sit-to-stand and floor-to-stand transfers require physical assistance of a person …'. Furthermore, with regards to the descriptors for level IV, the change in phrase from adult supervision during walking, used in the original GMFCS, to physical assistance in the GMFCS-E&R more clearly defines 'true' ambulators as children classified in levels I, II, and III only. Within our own laboratory, we now routinely collect GMFCS data using the GMFCS family report questionnaires and more recently have incorporated the Manual Ability Classification System (MACS) and Functional Mobility Scale (FMS) into our subjective assessment. As an adjunct to our standard examination procedures they provide a more global appraisal of the child at home and in the community. In particular, the FMS builds upon the need to assess the level of assistance required within different environments and provides greater potential for detecting change post intervention. Many clinicians, including myself, will welcome the development and publication of the GMFCS-E&R. Reliability testing and further validation will be required, and comparative study between the old and new descriptors for the 6- to 12-year age band may alleviate any concerns regards a shift in classification. The GMFCS family report questionnaire for this latter age band may also need revision and development of a family report questionnaire for the new 12- to 18-year age band would be extremely useful. On a more general note, the ongoing development of classification systems and measurement tools in the field of CP continues to improve our means of assessment, and thus provides us with the greater potential to scrutinize our current practises. This presents a challenge to clinicians and service providers alike, and one that must be embraced if we are to provide best possible care, improved participation, and a better lifestyle for this population of children and young people.