PROMs for Kids
PedEyeQ offers outcome measures for assessing impact of ophthalmic conditions in childhood. Howard Larkin reports.
Published: Wednesday, June 30, 2021
How do you make health care more patient-centred for children? Just as with any other group—ask them, and their parents. The recently developed PedEyeQ provides a structured assessment tool for doing just that for a wide range of paediatric ophthalmic conditions, said Jonathan M Holmes BM, BCh at the Association for Research in Vision and Ophthalmology 2021 Annual Meeting. “The PedEyeQ now allows assessment of functional vision and quality-of-life domains in paediatric eye conditions from infants to teenagers, and may thereby prove useful in the management of these conditions and as an outcome measure in research,” Dr Holmes, professor and chair of ophthalmology and vision science at the University of Arizona, Tucson, Arizona, USA. The key questions for developing effective patient-reported outcome measures (PROMs) are “what bothers you about your condition?” and “how does it affect your everyday life?” said Dr Holmes. Having patients answer such questions without clinician interpretation can give deep insight into issues ranging from functional vision, that is what the patient can or cannot do because of their vision (what the patient can and cannot do because of their vision), health-related quality-of-life (what bothers the patient), and symptoms. GO TO THE SOURCE "In developing the PedEyeQ questionnaire, we avoided the traditional process of experts coming up with PROM questions and testing them. Instead, we interviewed parents and children," Dr Holmes said. Semi-structured interviews captured three perspectives— the children’s; the parents’ perception of the child’s experience, known as a proxy assessment; and the effect on the parents themselves. Altogether, 204 children aged 5 to 17 years and 328 parents were interviewed across 10 diagnostic groups ranging from amblyopia to retina/optic nerve. The open-ended interviews were transcribed and coded, resulting in 614 candidate questions for child and proxy, and 589 for parents. A literature review to ensure completeness added no questions. For the PedEyeQ, questions pertaining to symptoms were removed, focusing on functional vision and quality-of-life. The process included development of parallel questions, reflecting functional vision, for example, “are you unable to play with Buzz Lightyear because of your eyes?” and quality-of-life related to that aspect of functional vision, for example, “does it bother you that you are unable to play with Buzz Lightyear because of your eyes?” The next step was grouping similar questions and coming up with general questions to cover them. For example, “are you unable to do things that you want to do because of your eyes?” rather than a specific Buzz Lightyear question. The resulting 63 items were then combined into master child and proxy questionnaires, which were tested on a new cohort of children and parents, with separate questionnaires for children aged 5 to 11 years and 12 to 17 years; proxy questionnaires for children 0 to 4 years, 5 to 11 years, and 12 to 17 years; and a parent questionnaire. WINNOWING QUESTIONS Factor analysis revealed distinct groups of questions, and each group was assessed for unidimensionality to ensure that unlike concepts were not combined in scoring. Each group of questions was then labelled as a domain, with a descriptive title such as “functional vision.” Another child domain was labelled “social,” with questions such as “do you get teased?”, and a further domain was labelled “bothered by eyes and vision” with questions such as “does it bother you that you have a hard time seeing the board at school?” Frustration and worry was the fourth and final child domain. Proxy questions were analysed separately by age, 0 to 4, 5 to 11, and 12 to 17 year-olds, and the same domains as the child questionnaire emerged, with the addition of an eyecare domain. The parent questionnaire was found to have four separate domains—impact on parent and family; worry regarding child’s eye condition; worry regarding child’s self-perception and interactions; and worry regarding child’s visual function. For scoring, Rasch analysis was used to convert nonlinear question scores to linear measures. For each domain, a Rasch map was created to separate questions according to severity. For example, in the functional vision domain, the question “do you have a hard time seeing?” is answered affirmatively by more children than “do you run into things because of your eyes?” indicating a higher severity score for the second question. Rasch analysis was also used to eliminate items that did not fit or resulted in poorer questionnaire performance. The aim was to have 10 or fewer questions per domain. Dr Holmes highlighted recent studies using the PedEyeQ that have found children with a wide range of ocular conditions, including amblyopia, cataract, refractive error, and strabismus, score lower than controls across functional vision and quality-of-life domains. (Hatt SR et al. AJO 2020. Leske DA et al. JAAPOS 2021.) The development of the PedEyeQ and subsequent validation studies have been a joint collaboration between Dr Holmes’ team at Mayo Clinic, Rochester, Minnesota, and Dr Eileen Birch’s team at the Retina Foundation of the Southwest, Dallas, Texas. The questionnaires and scoring sheets are available for download at the Paediatric Eye Disease Investigator Website www.pedig.net. and the development of the questionnaire has been published as Hatt SR et al, AJO 2019. Jonathan Holmes: firstname.lastname@example.org