Background: To assess cross-cultural validity between Dutch and English versions of the FVQ_CYP, a patient-reported outcome measure developed in the United Kingdom (UK) for children and adolescents with (severe) visual impairment or blindness (VI for brevity) to measure functional vision. Methods: The 36-item FVQ_CYP was translated and adapted into Dutch using standard guidelines. The questionnaire was administered to Dutch children and adolescents aged 7–17 years (N = 253) with impaired vision (no restrictions regarding acuity). Data were compared to existing UK data of children and adolescents aged 10–15 years (N = 91) with VI (acuity LogMar worse than 0.48). As with the original UK FVQ_CYP validation, a rating scale model (RSM) was applied to the Dutch data. Results: Minor adaptations were needed in translation-rounds. Significant differences in item responses were found between the Dutch and UK data. Item response theory assumptions were met, but fit to the RSM was unsatisfactory. Therefore, psychometric properties of the Dutch FVQ_CYP were analysed irrespective of the original model and criteria used. A graded response model led to the removal of 12 items due to missing data, low information, overlapping content and limited relevance to Dutch children. Fit indices for the remaining 24 items were adequate. Conclusions: Differences in population characteristics, distribution of responses, non-invariance at the model level and small sample sizes challenged the cross-cultural validation process. However, the Dutch adapted FVQ_CYP showed high measurement precision and broad coverage of items measuring children's functional vision. The underlying reasons for differences between countries in instrument performance are discussed with implications for future studies.
Keywords: Visual impairment; Cross-cultural validation; Children; Functional vision; Item response theory; Differential item functioning
Ellen B. M. Elsman and Valerija Tadić contributed equally to this work. Jugnoo S. Rahi and Ruth M. A. van Nispen also contributed equally to this work.
In recent years, emphasis on patient-centred care has resulted in the development of generic and disease-specific patient-reported outcome measures (PROMs) [[
Availability and implementation of instruments to assess functional vision in paediatric ophthalmology would complement objective clinical measures of visual function, such as visual acuity and visual field. Furthermore, these instruments can be used to monitor and evaluate the effectiveness of low vision rehabilitation. Currently, three measures of functional vision are available. Both of the two versions of the LV Prasad-Functional Vision Questionnaire have been developed for children in the developing world [[
The Functional Vision Questionnaire for Children and Young People (FVQ_CYP) for 10 to 15 year olds was previously designed to capture self-reported level of difficulty in the performance of vision dependent activities and is intended for children and adolescents with visual impairment (VI), severe VI or blindness i.e. with acuity in their better seeing eye of logMAR worse than 0.48. It was developed for and validated in a nationally representative sample of UK children [[
No measure of functional vision is currently available in the Netherlands. Although progress is being made in the development of age-specific versions of the Participation and Activity Inventory for Children and Youth (PAI-CY) [[
The cross-cultural validation of the FVQ_CYP was conducted in two phases. The first phase consisted of translation of the FVQ_CYP into Dutch, in keeping with standardized guidelines [[
The FVQ-CYP was translated in Dutch using an established process for cross-cultural adaptation of PROMs [[
Forward translation from English (source language) into Dutch (target language) was carried out by two independent bilingual translators, having Dutch as native language but were also fluent in English. Both translators were researchers regularly working with visually impaired children and aware of the concept of functional vision, and were as such informed translators. The instructions, questionnaire items, and scale were translated independently without any discussion between the translators.
The two translations were compared and any discrepancies were resolved through discussion and/or consultation of a third researcher not involved in the forward translation. Working from the original FVQ_CYP, as well as from the two translations, a synthesis of these translations was produced, resulting in one common translation.
The translated version of the FVQ_CYP was then translated back to English by two bilingual translators who were native English speakers. The two back-translators were naïve to the original English version of the FVQ_CYP and lay to the concept of functional vision and VI.
An expert committee including the project leader and all four translators reviewed all translations and resolved discrepancies through discussion resulting in consensus on the final wording to be used for the final version of the Dutch FVQ_CYP (FVQ_CYP_NL).
Children and adolescents aged 7–17 years enrolled for care at two Dutch low vision rehabilitation centres at the time of the study or in the past were invited to participate in the study. Children were required to have adequate knowledge and understanding of the Dutch language to participate in the study. Children with registered extensive (cognitive) impairment were excluded from the sample to be invited to participate by the low vision rehabilitation centres. Children with low vision from any cause were eligible and there was no restriction regarding visual acuity. As such, the inclusion criteria were more liberal with respect to both age and visual acuity than for the original instrument development and validation in the UK, which was intentionally limited to children and adolescents aged 10–15 years old with VI/severe VI or blindness i.e. visual acuity in the better eye of logMAR ≥0.48 [[
Although the original FVQ_CYP is intended for self-administration, Dutch children and adolescents who participated in the study were visited at their home by a researcher in order to administer the FVQ_CYP_NL using an interview format, providing an extra check on ability to participate. Besides, using an interview format was in line with the mode of administration applied for testing the PAI-CY. Parents provided information on socio-demographic and clinical characteristics of their child, such as age, gender, siblings, cause of low vision, level of VI, and other impairments, using a web-based survey questionnaire. Decimal visual acuity, visual field and ophthalmic diagnoses were retrieved from patient files at the low vision rehabilitation organisations. Missing values in patient files were supplemented by self-reported data of parents (n = 8). Visual acuity was converted into logMAR, and put into 5 levels based on the better seeing eye, according to World Health Organisation (WHO) taxonomy of VI [[
The study protocol was approved by the Medical Ethical Committee of the Amsterdam UMC, location VUmc, the Netherlands. The study adhered to the tenets of the Declaration of Helsinki and its later amendments. Written informed consent was obtained from all Dutch participants, i.e. from parents of all children, and from children and adolescents aged 13 years and older. Secondary analysis of the existing anonymised UK FVQ_CYP dataset did not require ethics approval. The data were drawn from the original development and psychometric study which involved individual consent to participation and was approved by the National Health Service Research Ethics Committee for UCL Institute of Child Health and Great Ormond Street Hospital, United Kingdom, and followed the tenets of the Declaration of Helsinki.
All statistical analyses related to IRT were conducted in R [[
Participants with > 25% missing responses were removed from the analyses. Sociodemographic and clinical characteristics for the Dutch and UK sample were investigated separately.
Item analysis, comprising descriptive statistics of each of the individual items, were conducted for the Dutch and UK samples. Differences in the distribution of responses over the response categories were investigated using chi-square tests.
Following the cut-off criteria used in the validation of the original FVQ_CYP [[
Then, IRT was applied on the Dutch sample. IRT comprises a collection of modelling techniques from modern measurement theory. It provides a powerful context to develop instruments which are more efficient, reliable and valid [[
-
Unidimensionality , which assumes that a single latent trait explains the covariance of items [[
24 ]]. Unidimensionality was assessed by performing an eigen value decomposition on the matrix of robust (Spearman) correlations between the items. A difference approximation to the second-order derivatives along the eigenvalue curve (scree plot) was calculated. This acceleration-approximation indicates points of abrupt change along the eigenvalue curve and the number of eigenvalues before the point with the most abrupt change (the point with the maximum acceleration value) represents the number of latent dimensions that dominate the information content [[26 ]]. Subsequently a principal component analyses (PCA) was performed to proxy if all items load on a single component (where the component is taken as a proxy for the latent trait). -
Local independence of items , which requires that item responses are independent given their relationship to the latent trait. Local dependence was assessed by inspection of possible excess covariation (> 0.25) among items in the residual matrix resulting from PCA. Local dependence could occur in items that are similar in content, refer to a similar condition (similar stem) or are presented successively [[
24 ]]. Item pairs with excess covariation were flagged. -
Monotonicity , which states that the probability of endorsing a higher item response category should not decrease with increasing levels of the underlying latent trait [[
27 ]]. Monotonicity was evaluated using Mokken scale analysis. The graphs were visually inspected, and the Loevinger H coefficient was calculated to assess scalability [[28 ]] (see also [[29 ]]). A Loevinger H coefficient < 0.30 was considered unsatisfactory.
Because on the original FVQ_CYP the rating scale model (RSM) was used [[
Some items might not fit the GRM model, and therefore individual item fit was assessed using the X
Differential item functioning (DIF) analyses were used to assess whether participants from different groups (i.e. age and gender) with the same level of disability have different probabilities of selecting a certain response to an item [[
Known-group validity was assessed to ensure the FVQ_CYP_NL is able to discriminate between groups [[
Some minor differences in wording of two items related to activities at school, i.e. "taking part in science classes" and "seeing the board in the class" were found and were resolved by discussion: examples (i.e. physics and biology) were added to science classes, and board was translated to schoolboard or digital board, as most schools in the Netherlands nowadays use a digital board. During the first questionnaire administration to participants, it was noted that the response option 'not applicable' was warranted, because in the Netherlands young children usually do not have homework for which they need the computer, and not all classes (e.g. science and geography) are obligatory for all ages to which the questionnaire was administered. It is worth noting that the 'not applicable' option was also included in the original FVQ_CYP UK study, but was subsequently removed due to high endorsement of this category, resulting in a high proportion of 'missing' data. Furthermore, it was noted that different wording might be necessary for younger children vs. the older children. For example, math classes were translated into the Dutch word "wiskunde" (i.e. mathematics), but only children in high school have "wiskunde". In primary school, this class is called "rekenen" (i.e. to calculate). Therefore, two age-appropriate versions (7–12 years and 13–17 years respectively) of the FVQ_CYP_NL were created with minor differences in the wording of five items related to activities at school.
A total of 263 Dutch children and adolescents were recruited in this study. Ten participants were excluded from the analyses because they had an excessive number of missing responses. In the UK dataset, this was the case for three participants, resulting in a dataset containing responses of 91 children and adolescents. The demographic and clinical characteristics of the Dutch sample and the UK sample are summarised in Table 1. As expected there were differences in age, level of VI and occurrence of other impairments, due to more liberal inclusion criteria with respect to these variables in the Dutch sample.
Socio-demographic and clinical characteristics of the Dutch and the UK sample
Dutch sample UK sample Age in years, mean ± SD (range) 11.06 ± 2.87 (7–18) 12.09 ± 1.84 (9–15) Gender, n (%) Male 150 (59.3) 52 (57.1) Female 103 (40.7) 39 (42.9) Level of VI VI0: logMAR ≤0.47 126 (49.8) – VI1: logMAR 0.48–0.70 56 (22.1) 42 (46.2) VI2: logMAR 0.71–1.00 35 (13.8) 31 (34.1) SVI: logMAR 1.01–1.30 4 (1.6) 10 (11.0) Blind: logMAR ≥1.31 24 (9.5) 8 (8.8) Unknown 8 (3.2) – Nationality Ethnic majority 228 (90.1) 77 (84.6) Ethnic minortiy 25 (9.9) 14 (15.4) Other impairment, n (%) Yes 117 (46.2) 28 (30.8) No 124 (49.0) 62 (68.1) Unknown 12 (4.7) 1 (1.1) Siblings, n (%) No 29 (11.5) 6 (6.6) One 119 (47.0) 49 (53.8) Two or more 93 (36.8) 33 (36.3) Unknown 12 (4.7) 3 (3.3) Siblings with VI, n (%) Yes 41 (16.2) 21 (23.1) No 171 (67.6) 61 (67.0) N/A (no/unknown siblings) 41 (16.2) 9 (9.9)
Table 2 presents the distribution of responses over the response categories for the Dutch sample and the UK sample. The response option 'not applicable' was treated as a missing value. As such, four items in the Dutch sample had missing scores > 20% ("using the computer for homework", "taking part in science classes", "taking part in geography classes", and "watching plays and shows in the theatre") and these items were removed. None of the items had floor or ceiling effects, and in all items all four response categories were endorsed. However, infrequent endorsement of the response option 'very difficult or impossible' in almost all items in the Dutch sample motivated the collapsing of response options 'very difficult or impossible' and 'difficult'. There were no item pairs displaying high inter-item correlations (> 0.7). There were significant differences in the distribution of responses between the Dutch and the UK sample for all but five items. In general, the Dutch sample was more likely to opt for the response options 1 or 2 ('very easy' or 'easy') and less likely to opt for the response options 3 or 4 ('difficult' or 'very difficult/impossible') than the UK sample. Matching the Dutch sample on UK inclusion criteria (i.e. age 10–15 years and VI logMAR ≥0.48; n = 63 for Dutch sample and n = 85 for UK sample) did not influence these results.
Differences in distribution of responses over the response categories for the Dutch sample (n = 253) and the UK sample (n = 91)
Item Item content Distribution of responding population Dutch sample (%) over the response optionsa Missing responses Dutch sample (%) Distribution of responding population UK sample (%) over the response optionsa Missing responses UK sample (%) 1 2 3 4 1 2 3 4 FV_1 Watching TV 43.6 50.0 4.8 1.6 1.2 28.1 42.7 25.8 3.4 2.2 < 0.001 FV_2 Playing video and computer games 50.4 42.8 5.9 0.8 6.7 23.9 47.7 20.5 8.0 3.3 < 0.001 FV_3 Playing other games, e.g. board games or card games 35.4 53.6 10.5 0.4 6.3 21.3 34.8 38.2 5.6 2.2 < 0.001 FV_4 Using the computer for homework 35.8 56.2 7.3 0.7 45.8 26.1 52.3 17.0 4.5 3.3 0.020 FV_5 Reading food packets, labels or recipes 10.2 31.3 36.6 22.0 2.8 6.0 14.3 40.5 39.3 7.7 0.002 FV_6 Doing household chores, e.g. washing up 33.6 57.3 7.7 1.4 13.0 20.0 52.5 18.8 8.8 12.1 < 0.001 FV_7 Telling the time on a wrist watch 34.1 37.4 23.8 4.7 15.4 13.8 34.5 32.2 19.5 4.4 < 0.001 FV_8 Telling the time on a wall clock 26.8 45.2 17.6 10.5 5.5 11.2 28.1 30.3 30.3 2.2 < 0.001 FV_9 Using the computer for lessons 42.1 52.3 5.1 0.5 14.6 23.6 56.2 15.7 4.5 2.2 < 0.001 FV_10 Reading small print text books, worksheets and exam papers 11.6 28.9 35.5 24.0 4.3 3.4 13.6 38.6 44.3 3.3 < 0.001 FV_11 Reading enlarged text books, worksheets and exam papers 47.9 43.8 2.5 5.8 5.1 42.0 42.0 10.2 5.7 3.3 0.030 FV_12 Drawing or painting 32.5 56.3 8.3 2.9 5.1 30.3 33.7 31.5 4.5 2.2 < 0.001 FV_13 Reading hand writing 10.4 44.6 30.7 14.3 0.8 13.6 12.5 53.4 20.5 3.3 < 0.001 FV_14 Seeing the board in the class 25.0 46.0 19.4 9.7 2.0 8.2 23.5 35.3 32.9 6.6 < 0.001 FV_15 Recognizing people, e.g. in school corridors 32.4 48.6 15.0 4.0 0.0 15.9 33.0 33.0 18.2 3.3 < 0.001 FV_16 Recognizing other people's facial expressions 23.4 45.6 20.6 10.3 0.4 21.8 28.7 28.7 20.7 4.4 0.008 FV_17 Finding friends in the playground 21.6 43.6 30.8 4.0 1.2 8.0 28.4 36.4 27.3 3.3 < 0.001 FV_18 Taking part in science classes 23.7 60.4 13.7 2.2 45.1 23.3 51.1 20.0 5.6 1.1 0.261 FV_19 Taking part in geography classes 24.1 56.0 16.3 3.6 34.4 18.2 44.2 28.6 9.1 15.4 0.027 FV_20 Taking part in math classes 27.2 49.6 21.2 2.0 1.2 22.2 52.2 21.1 4.4 1.1 0.528 FV_21 Taking part in PE 39.3 50.4 9.4 0.8 3.6 20.7 35.6 35.6 8.0 4.4 < 0.001 FV_22 Taking part in English/Dutch classes 30.5 55.0 12.4 2.0 1.6 20.2 57.3 20.2 2.2 2.2 0.150 FV_23 Keeping up with the teacher in lessons 21.8 53.2 22.6 2.4 0.4 21.1 37.8 35.6 5.6 1.1 0.023 FV_24 Keeping up with other students in class 23.9 52.6 22.3 1.2 0.8 23.1 33.0 39.6 4.4 0.0 0.001 FV_25 Getting around the school by yourself 43.3 50.4 6.0 0.4 0.4 42.9 41.8 13.2 2.2 0.0 0.047 FV_26 Getting around outdoors by yourself 36.5 52.0 10.7 0.8 0.4 17.2 41.4 34.5 6.9 4.4 < 0.001 FV_27 Reading signs and posters at stations or shops 19.2 40.8 30.8 9.2 5.1 13.6 26.1 30.7 29.5 3.3 < 0.001 FV_28 Getting around in crowds by yourself 12.1 32.0 44.1 11.7 2.4 11.4 19.0 39.2 30.4 13.2 0.001 FV_29 Seeing small moving objects, e.g. balls 14.6 40.3 30.4 14.6 0.0 10.3 11.5 40.2 37.9 4.4 < 0.001 FV_30 Seeing large moving objects, e.g. cars passing 39.1 45.1 10.7 5.1 0.0 30.7 50.0 12.5 6.8 3.3 0.546 FV_31 Using the escalators 39.6 49.4 9.4 1.6 3.2 40.4 38.2 18.0 3.4 2.2 0.077 FV_32 Playing team sports, e.g. football, without adaptations 27.6 50.7 18.7 3.0 19.8 18.4 21.8 33.3 26.4 4.4 < 0.001 FV_33 Watching films in the cinema 40.5 51.4 7.3 0.9 13.0 34.8 40.4 19.1 5.6 2.2 0.001 FV_34 Watching plays and shows in the theatre 26.2 55.9 14.9 3.0 20.2 15.0 38.8 31.3 15.0 12.1 < 0.001 FV_35 Reading price tags 16.5 49.6 24.2 9.7 2.0 12.0 30.1 33.7 24.1 8.8 < 0.001 FV_36 Finding correct money to pay 22.6 55.7 20.0 1.7 9.1 22.4 40.0 25.9 11.8 6.6 < 0.001
The acceleration factor suggested a one-factor solution for the Dutch data. Principal components of the one-factor solution were all positive and moderate to large. Inspection of item and factor content gave no reason for multidimensional solutions. The first factor accounted for 33% of the variance, whereas the second factor accounted for 5% of the variance; thus, the ratio of explained variance by the first and second factor is 6.6, which is higher than the required minimum of 4 [[
Five items were removed after the first application of the GRM: "reading food packets, labels or recipes", "doing household chores, e.g. washing up", "telling the time on a wrist watch", "drawing or painting", and "keeping up with the teacher in lessons". These items were removed because they provided very little information (i.e. little precision/discrimination) and/or because they covered the same area on the disability trait as another item, but provided less information and/or provided information over a smaller range of the disability trait. Content validity, item relevance and similarities with other items were also considered. Three additional items were removed after the second fit of the GRM ("taking part in math classes", "taking part in physical education", and "taking part in Dutch language classes"), mainly because they still provided very little information.
The Likelihood Ratio test showed that the full GRM outperformed the polytomous Rasch model for the 24 items (LRT = 40.0, p = 0.015). The fit indices reflected adequate overall model fit of the 24 items to the GRM: RMSEA = 0.061, SRMR = 0.062, TLI =0.965, and CFI = 0.968. Table 3 summarizes GRM item parameters, information and fit statistics of the FVQ_CYP_NL. Item discrimination ranged from 1.11 to 2.27. The item with the lowest discrimination was "reading small print text books, worksheets and exam papers", and the item with the highest discrimination was "reading enlarged text books, worksheets and exam papers". Item threshold parameters ranged from − 2.26 to 2.60. Item information ranged from 1.76 to 4.26, and total information of the 24 items was 65.32. All items fitted the GRM at the p < 0.01 level. Despite the fact that some items still provided little information, further item removal was considered unfavourable given the location of these items on the disability trait and for reasons of content validity. The item-person map shows that items are distributed almost entirely across the disability trait. The thetas of respondents adequately match the item thresholds, although there are no items for persons with low levels of disability (Fig. 1).
GRM item characteristics for the 24 item FVQ_CYP_NL (n = 253)
Item Item content Discrimination α Threshold β1 Threshold β2 Item information X2 FV_1 Watching TV 1.32 −0.26 2.49 2.45 8.17 0.52 FV_2 Playing video and computer games 1.19 − 0.02 2.56 2.12 13.04 0.22 FV_3 Playing other games, e.g. board games or card games 1.41 −0.57 1.89 2.58 12.41 0.26 FV_8 Telling the time on a wall clock 1.74 −0.84 0.85 3.06 13.45 0.20 FV_9 Using the computer for lessons 1.29 −0.36 2.60 2.41 11.51 0.18 FV_10 Reading small print text books, worksheets and exam papers 1.11 −2.26 −0.43 1.76 11.69 0.39 FV_11 Reading enlarged text books, worksheets and exam papers 2.27 −0.004 1.70 4.26 8.95 0.18 FV_13 Reading hand writing 1.22 −2.24 0.24 2.17 9.03 0.53 FV_14 Seeing the board in the class 1.42 −1.05 0.88 2.46 19.18 0.12 FV_15 Recognizing people, e.g. in school corridors 1.50 −0.67 1.33 2.66 11.86 0.46 FV_16 Recognizing other people's facial expressions 1.67 −1.03 0.74 2.96 19.43 0.05 FV_17 Finding friends in the playground 1.36 −1.24 0.67 2.33 15.44 0.22 FV_24 Keeping up with other students in class 1.28 −1.19 1.20 2.27 14.36 0.35 FV_25 Getting around the school by yourself 1.72 −0.23 2.11 3.24 14.04 0.05 FV_26 Getting around outdoors by yourself 1.80 −0.44 1.63 3.34 9.00 0.44 FV_27 Reading signs and posters at stations or shops 1.64 −1.28 0.40 2.84 13.22 0.21 FV_28 Getting around in crowds by yourself 1.25 −2.05 −0.23 2.05 15.07 0.18 FV_29 Seeing small moving objects, e.g. balls 1.26 −1.80 0.23 2.15 14.90 0.19 FV_30 Seeing large moving objects, e.g. cars passing 1.82 −0.34 1.36 3.25 14.89 0.19 FV_31 Using the escalators 1.58 −0.34 1.80 2.89 13.38 0.15 FV_32 Playing team sports, e.g. football, without adaptations 1.58 −0.86 1.16 2.85 12.22 0.27 FV_33 Watching films in the cinema 1.90 −0.37 1.77 3.59 8.50 0.29 FV_35 Reading price tags 1.61 −1.45 0.64 2.93 15.53 0.11 FV_36 Finding correct money to pay 1.48 −1.16 1.17 2.70 9.81 0.46
Graph: Fig. 1 Item-person map of the 24 item FVQ_CYP_NL
After two iterations, analysis of DIF for age indicated three items with some level of DIF, which was all uniform (Fig. 2a). However, change in McFadden's R
Graph: Fig. 2 Item response functions, McFadden's pseudo R2 and p-values, and IRT parameters for items displaying DIF for age (a) and gender (b) (n = 253)
Figure 3a shows the total impact of DIF for age on the test characteristic curve (TCC), and Fig. 3b the total impact of DIF for gender. The TCC shows the relation between the total scores (y-axis) and thetas (x-axis). The left graphs show the impact on the test score for all items, whereas the right graphs show the impact of only those items with DIF. The curves show that the total score is the same for both age groups and genders, indicating minimal impact of DIF by age and gender.
Graph: Fig. 3 Total impact of DIF on the test characteristic curve (TCC) for age (a) and gender (b) (n = 253)
Known-group validity was established for groups that differ on level of VI and gender. Those with severe VI/blindness had significantly higher thetas than those with moderate VI and mild VI (p= 0.002 and p<0.001 respectively), indicating that they experienced more disability and the FVQ_CYP_NL was able to discriminate them. Females had significantly higher thetas than males (p=0.008), and no significant differences were found in thetas between those with and without other impairments.
This study reports the cross-cultural adaptation of the original UK version of the FVQ_CYP into Dutch and its important psychometric properties. The FVQ_CYP is a PROM which measures functional vision of children and adolescents with VI [[
We originally planned to perform 'strict' cross-cultural validation of the FVQ_CY_NL by applying the same criteria for item analyses as used in the validation study of the UK questionnaire [[
Interestingly, we found a number of differences in the psychometric performance of the instrument versions of the two countries. There were differences in the distribution of missing responses and response patterns between the Dutch data and the UK data. Some items had high missing responses in the Netherlands, but not in the UK, and Dutch children were less likely to opt the response category 'very difficult/impossible'. There are a number of possible reasons for these differences. Firstly, the difference in instrument performance between countries may have been driven by differences in the population due to the broader age range and less restrictions in degree of vision impairment in the Dutch sample. There were also differences in the presence of comorbidity between the samples. Matching the samples did not improve the results. Secondly, differences might have been influenced by different modes of administration. Data in the original UK study had been collected as self-report and self-completion with questionnaires returned by post [[
Besides the differences in psychometric performance, there was non-invariance at the model level; the RSM did not fit the Dutch data, whereas fit for the UK data was satisfactory. The RSM assumes that the discrimination parameter (i.e. the slope) is equal across all items (and therefore this model belongs to the Rasch family), and that the thresholds for each category response are also equal across items [[
Measurement invariance implies that the association between test scores and latent traits of persons is unconditional on group affiliation or time of measurement [[
Application of the GRM resulted in the identification of eight items that contributed very little information or covered the same area on the disability trait as another item while providing less information, and therefore these were removed. Together with the four items that had too many missing responses, this resulted in an instrument containing 24 items. Some of the items which were deleted might have been less relevant for younger children (i.e. "using the computer for homework", "doing household chores, e.g. washing up", and "reading food packets, labels or recipes"). The item "telling the time on a wrist watch" might have been superseded by modern technology, e.g. the use of smartphones. This might even be true for younger children, as 68% of the 10-year olds in the Netherlands had a smartphone in 2017 [[
The item thresholds of the final instrument reflected a good coverage across the disability trait. The FVQ_CYP_NL seemed better targeted to children and adolescents with higher thetas at the disability trait, and there might be a need for more difficult items. However, this was also expected, because the FVQ_CYP was originally developed with and for children with more severe visual impairment than the Dutch sample. This study shows that, with appropriate modification, it is also possible to administer the questionnaire to children and adolescents outside the original 10–15 years age interval, and to children and adolescents with less severe visual impairment. This was already anticipated by the authors of the original FVQ_CYP, who are currently completing development and an additional assessment of psychometric properties of age-appropriate versions applicable to a wider age range [[
The DIF analyses identified three items with uniform DIF for age and three items (one overlapping) with non-uniform DIF for gender (although results of χ
Infrequent endorsement of the fourth response category motivated collapsing the third and fourth category. However, adding the 'not applicable' response option might have caused attrition in the data, because children might have opted for the 'not applicable' category where they also could have opted for 'very difficult/impossible'. This was also speculated to be the case in the validation of the original instrument in the UK, and with frequent endorsement of the 'not applicable' option resulting in a high proportion of missing data, the authors subsequently decided to remove 'not applicable' as a response option from the instrument [[
In conclusion, non-invariance at the model level, small sample sizes, and differences in population characteristics and distribution of responses posed challenges to the standard cross-cultural validation process. However, although this imposes limitations to direct comparability of the FVQ-CYP between the Netherlands and UK, by using a GRM, we have established validity of the FVQ_CYP_NL as a stand-alone instrument for use in the Netherlands (thus the FVQ_CYP UK version served as the building block). The Dutch adapted FVQ_CYP – the FVQ_CYP_NL – is a unidimensional scale with high measurement precision and broad coverage of items measuring children's functional vision. Deletion of items ensured that only those items most applicable to the Dutch setting and providing high information were included in the final questionnaire. This study provides detailed information on item parameters, and shows that the FVQ_CYP_NL is targeted adequately to the abilities of children and adolescents aged 7–17 with different levels of VI. In its current form the FVQ_CYP_NL is a short, easy to administer instrument, with sound psychometric properties, which can be used to assess the self-reported level of difficulty in performing vision-dependent activities in children and adolescents with visual impairment. However, further evaluation of psychometric properties such as the application and functioning of the recommended response categories, construct validity, test-retest reliability, and responsiveness is necessary.
Our study has implications for cross-cultural use of instruments in general. Given the scarcity of measures for children and adolescents in ophthalmology and the challenges in developing instruments de novo with heterogeneous and numerically small clinical populations, there is a value in using well developed instruments and adapting them cross-culturally. However, care must be taken that rigorous, standard cross-cultural validation processes are followed. Even when instruments are invariant at model or item level, it is possible to have language versions of an instrument that are reliable and valid for use in each country but differ extensively in wording or are even comprised of different items from item banks, that demonstrate identical response functions, facilitating cross-cultural use [[
This study was funded by a fellowship of Royal Dutch Visio. A travel grant for this study was provided by the Quality of Care programme of the Amsterdam Public Health research institute. The sponsors had no role in the design, conduct or outcomes of this study.
The initial study developing and validating the FVQ_CYP in the UK was funded by a Fight for Sight Project Grant (2014). Further support was received from the following sources: Guide Dogs for the Blind Association, National Institute of Health Research (NIHR) Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust and University College London Institute of Ophthalmology, and NIHR Central and East London Comprehensive Research Network. The study was undertaken at University College London Institute of Child Health/Great Ormond Street Hospital and Moorfields Eye Hospital/at University College London Institute of Ophthalmology, both of which receive a proportion of funding from the Department of Health's NIHR Biomedical Research Center's funding scheme. Members of the team are supported by the Ulverscroft Foundation. The FVQ_CYP is available on request from Professor Rahi. Copyright statement: The FVQ_CYP should not be reproduced or modified without Professor Rahi's permission. Copyright© 2013 University College London (UCL) Institute of Child Health.
We would like to thank all participating children.
EE, VT, RvN, GvR & JR were involved in funding acquisition; EE, VT, GvR, JR, & RvN conceptualized and designed the study; EE & VT collected the data; EE, VT & CP carried out the data analyses; EE, VT, CP & RvN were involved in interpretation of the data and results; EE & VT drafted the manuscript; JR, RvN, CP & GvR revised and approved the manuscript. All authors read and approved the final manuscript.
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
The study protocol was approved by the Medical Ethical Committee of the VU University Medical Centre, Amsterdam, the Netherlands. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from all Dutch participants, i.e. from parents of all children, and from children and adolescents aged 13 years and older. Secondary analysis of the existing anonymised UK FVQ_CYP dataset did not require ethics approval. The data were drawn from the original development and psychometric study which involved individual consent to participation and was approved by the National Health Service Research Ethics Committee for UCL Institute of Child Health and Great Ormond Street Hospital, United Kingdom, and followed the tenets of the Declaration of Helsinki.
Not applicable.
Dr. Peeters and Dr. van Nispen are editorial board members of BMC Medical Research Methodology. Professor Rahi is a National Institute for Health Research (NIHR) Senior Investigator. The views expressed in this article are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health. The remaining authors have no competing interests.
• CFI
- Comparative fit index
• CRC
- Category response curve
• DIF
- Differential item functioning
- FVQ_CYP
- Functional Vision Questionnaire for Children and Young People
- FVQ_CYP_NL
- Dutch version of the Functional Vision Questionnaire for Children and Young People
• GRM
- Graded response model
• IIC
- Item information curve
• IRT
- Item response theory
• PAI-CY
- Participation and Activity Inventory for Children and Youth
• PCA
- Principal component analysis
• PROM
- Patient-reported outcome measure
• RMSEA
- Root mean square error of approximation
• RSM
- Rating scale model
• SRMR
- Standardized root mean square residual
• TCC
- Test characteristic curve
• TLI
- Tucker-Lewis index
• UK
- United Kingdom
• VI
- Visual impairment
• WHO
- World Health Organisation
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
By Ellen B. M. Elsman; Valerija Tadić; Carel F. W. Peeters; Ger H. M. B. van Rens; Jugnoo S. Rahi and Ruth M. A. van Nispen
Reported by Author; Author; Author; Author; Author; Author