| The importance of the theoretical underpinning of PROMs |
PROMs assess a variety of constructs (eg, quality of life, wellbeing, health status, functional status) so the purpose of measurement should be clearly defined at the outset as not to conflate the underlying constructs.8
|
Vision-related outcomes of interests (eg, vision-related quality of life vs. visual ability) need to be clearly distinguished and measured with appropriate PROMs. |
| Child PROMs need to be developmentally appropriate |
Child PROMs need to be developmentally appropriate but because of variability in children's development and abilities, there is no fixed age-related criterion for judging when children can reliably complete a PROM.8, 18 Matza et al18 recommend 4 key age groups as a starting point for making decisions about age-appropriate PROM administration (1. below 5 years, 5 to 7 years: child-report is possible, but reliability and validity often questionable, 2. 8–11 years: reliability and validity of child-report improves, 3. 12–18 years: self-report is preferred). However, it is recommended that specific age cut-offs should be determined individually for each PROM (developed and validated with adequate sample size at the upper and lower bounds of the target age range) and tested with cognitive interviews in each new target population.18
|
With available PROMs, age-related boundaries may need to be treated flexibly because of varying degrees of a delay in acquisition of key developmental milestones associated with significant visual impairment from infancy (eg, consider if a form intended for 5–7 year old children may or may not be more appropriate for a visually impaired 8 year old).
If existing PROMs with set age-appropriate cut offs are used, it should be reported if these were used flexibly to account for developmental variation in visually impaired children and this should be considered in interpretation of scores/findings. |
| Age-appropriate formats and administration methods |
Child-centred PROMs should be designed and formatted appropriately for the target age group,8, 18 including considerations of health-related vocabulary and reading level, response scale, recall period, instrument length, pictorial representations, formatting, methods or administration and electronic data collection.18
|
Flexible formats and administration approaches need to be considered and/or developed for children with differing levels of visual impairment of different ages to enable self-reporting whenever possible. |
| A child-targeted PROM should be grounded in children's voices and be psychometrically robust |
Content validity of a child PROM should be established with children. Children should be included in the early qualitative research stages (through interviews and focus groups) conducted to determine that the content of the PROM is relevant and comprehensible to children.18
A PROM also needs to be psychometrically robust, demonstrating reliability, validity, responsiveness, precision, interpretability, acceptability, and feasibility.8
|
The reality and implications of small sample sizes when developing and applying PROMs for visually impaired children, due to the rarity of the population, need to be recognised and considered in interpreting the findings.
The sources of potential bias (eg, lower response rates by families from more socio-economically deprived subgroups) should be recognised and reported. |
| Self-report vs. proxy report. If proxy is used – when, by whom and why? |
Children's own self-report should be encouraged and collected whenever possible8, 18
Proxy-reports (eg, by parents, teachers or clinicians) can be used if children are unable to self-report (due to age or cognitive limitations), but attention should be given to considering ‘who' is the best proxy and ‘why' in a given context.18 If proxy-reports are used these must not be aggregated with self-reports.8
Ideally, where both child and parent versions of a PROM are available, both should be collected to help interpret results when children's self-reports are unavailable.8
|
‘Flexibility' should be allowed for different levels of self-reporting ability in children with different levels of visual impairment who may require different levels of adult input to complete a PROM (eg, reading and scribing for blind children).
Appropriate instructions should be provided for the adults (parents or professionals) to allow them to help, where required, the child to ‘self-report', without influencing the child's response.
Information on whether and what kind of help was needed should be recorded systematically and its impact on the child's responses should be assessed. |
| Cross-cultural issues |
Content validity and measurement properties of a paediatric PROM may not transfer to a different cultural setting and will need to be re-examined within each new culture where it is being used.18
|
|
