Assessment of health-related quality of life (HQoL) has been widely applied in medical and eye care services, supporting conventional clinical findings with supplemental outcome measures. Questionnaires, also known as instruments in outcome research literature, have been developed to allow assessment of latent traits in patients, such as self-perception of HQoL and functional vision (FV).1–4 In children with normal development, visual abnormalities impact negatively on HQoL.5–9 The impact of visual impairment on HQoL in adults with one form of intellectual disability (ID), Down syndrome, has been investigated by at least one study and is reported in abstract form.10–12 In one recent study, FV was assessed by questionnaire before and after surgical correction of high ametropia in children with developmental delay and other neurobehavioral disorders.13 To our knowledge, there are no instruments specifically designed for FV assessment in children with ID, which is surprising in view of the relatively high incidence of ocular abnormalities in children with ID.14–19 Such instruments may be useful to measure the extent of the impact of visual abnormalities on HQoL in children with ID and as outcome measures after interventions.
In recent years, people with ID are increasingly integrated into the general community.20 The high incidence of ocular abnormalities in the intellectually disabled populations raises questions on the significance of these abnormalities in terms of their impact on life quality and participation in work and community activities. Self-report instruments may prove valuable clinically as indicators of the patient's level of visual function-related difficulty. It seems likely that visual abnormality might have a different impact on the population of children with ID than in children with normal development, because children with ID also commonly have a range of impairments that may impact on their quality of life. For example, communication difficulties, hyperactivity, and gaze avoidance have a significant impact on social interaction, success in education, and future career prospects.21,22 In addition, people with Down syndrome commonly have defects of the heart and gastrointestinal tract, among other abnormalities, and in fragile X syndrome, orthopedic problems are common.15–17 Visual dysfunction is, therefore, one of a number of factors that may have a negative impact on quality of life in people with ID. It should be noted, however, that visual abnormality differs from many of the difficulties encountered in ID populations, in that its major causes (e.g., refractive errors and strabismus) can be treated, at least during childhood.17,18,23 Thus, it is possible to improve visual function to normal or near normal levels in many visually impaired children with ID, and it is important to evaluate the effect of intervention with valid instruments to document the changes with intervention on self-reported FV and HQoL.
The questionnaire-based survey, widely used in HQoL assessment, offers an efficient means by which to gather individual perspectives. Assessment of HQoL often involves patient-based assessments, in which subjects judge their own functional status and life quality perspectives.24,25 For the assessment of HQoL in children, proxy perspectives may be sought from parents or guardians. However, the children's perspectives may differ substantially from those of the proxy respondents. In addition, children with ID may have different perspectives of HQoL than their peers without ID.26,27 The availability of an appropriate instrument would allow exploration of HQoL in children with ID.28 This study aimed to refine questionnaires for this purpose, so that FV and HQoL may be assessed from the child's perspective. Instruments of this kind will add to clinical measures of vision, such as visual acuity, which indicate the functional status of the visual system, but do not directly indicate how well the system functions from the patient's perspective. After the evaluation of psychometric properties, the instruments are expected to be of use in outcome research in children with ID.
This study was approved by the Human Research Ethics Committee of the University of New South Wales and was formally ratified by the Principal of the Lu Jia Zhui School for Special Education, Shanghai, China, where data were collected. A parent or guardian of each child provided a signed declaration of informed content before the child's participation.
Questionnaire Selection and Adaptation
With the aim of generating a set of questions that children with ID would be able to interpret and answer, a questionnaire package was developed, consisting of two sections for completion by children. One questionnaire was based on an illustrated children's self-questionnaire [the Autoquestionnaire Enfant Image (AUQUEI)],29,30 which addresses children's perceived satisfaction (using a Likert scale from “not happy at all” to “very happy”) with 23 items in HQoL domains that included school activities, health, social life, leisure activities, and family life. Another questionnaire (including 22 items and using a dichotomous scale with “yes” and “no” response options) was adapted from the LV Prasad-Functional Vision Questionnaire (LVP-FVQ),7 which covers aspects of FV in children, such as academic performance, gross motor activities, fine motor activities, object discrimination, and distance vision. Validation of these instruments for use in children with ID would provide useful tools for the assessment of HQoL and FV in this population. Our previous study reports on the feasibility of HQoL and FV assessment in Chinese children and adolescents with ID using these instruments after modification in format and wording.31
Questionnaire Survey Administration
The questionnaires were used in children with ID at the Lu Jia Zhui School for Special Education, Shanghai, China. Subjects completed the questionnaires in classrooms, with guidance from a teacher, strictly following instructions provided with the questionnaire. Participants were allowed as much time as needed to complete the questionnaire set. To maximize test validity and reliability, the questionnaires were administered in a consistent manner (environment, time allowed for completion, and administration protocol were the same for all subjects). Requirements for individual guidance were likely to vary across this group because of, in part, the different types and severity of ID. With this in mind, groups of five to six subjects were allocated to a teacher who was familiar with the academic performance and behavior of the members of the group. The teacher read each question to the subjects and guided the subjects through the questionnaire, by explaining the meaning of each question if necessary. All teachers involved in the survey went through a group orientation (conducted by author Y.C.) to clarify appropriate questionnaire instruction, including the importance of not guiding subjects toward a response. Although questionnaires were being completed, an investigator (Y.C.) observed the procedure as a check of teacher and subject compliance.31
An ophthalmic examination was conducted on 107 subjects whose parents gave consent for this part of the study. This examination allowed us to determine whether subjects had visual abnormalitiesa and provided an indication of the extent to which the LVP-FVQ, intended for children with low vision, would apply to these subjects. The screening procedure included monocular visual acuity testing (Lea Symbols; Precision Vision, Villa Park, IL), static retinoscopy, a color vision test (Waggoner HRR Pseudoisochromatic Plates; Home Vision Care, Gulf Breeze, FL), a stereoacuity test (Randot Preschool Stereoacuity Test; Stereo Optical, Chicago, IL), cover test, and direct ophthalmoscopy. Subjects found to have significant visual abnormalities were referred for further assessment as appropriate.
For each item, the distribution of responses was studied. Skewed distribution (e.g., most respondents selecting the lowest level on the response scale) would indicate a ceiling or floor effect. Data normality and appropriateness were investigated using a commercially available software package (SPSS 15.0 for Windows; SPSS Science, Chicago, IL). Data within the range of ±2.00 in skew and kurtosis were considered compliant with normality.4 Appropriateness was assessed by determining the degree of missing data and ceiling/floor effect. Missing data were defined as any item to which more than 10% of respondents did not respond. Ceiling/floor effect was defined as more than 90% (on the dichotomous LVP-FVQ scale) or 50% (on the 4-option AUQUEI scale) of respondents selecting one response category at the high or low end of the scale. After examination of skew, kurtosis, and normality, items not compliant with normality were assessed in terms of the item context and its match or mismatch with adjacent items on the scale.
Rasch analysis was performed using Winsteps 3.3532 with the joint maximum likelihood estimation (Wright and Masters'33 rating scale model). Ordinal data were transformed to an interval scale, using a unit known as the logit (log-odds unit). Rasch modeling stems from the theory:
in which probability of success/probability of failure is called “success odds.” According to this theory, the relationship of person ability and item difficulty can be described on a common interval scale using logit units. When the data are fitted to the model, a common linear scale is established, and on this scale, the person-to-person, item-to-item, or item-person differences are represented using logits on the same scale.34 As assumed by Rasch analysis, the logit is a calibrated unit in measuring item difficulty and person ability. The mean item difficulty/ability is marked as zero on the common scale of both person and item. In this study, the person-logit refers to the difference between each person's ability and the group-mean item difficulty. If the person-logit has a positive score, the person's ability is higher than the average level of difficulty for items.7 A similar approach applies to item-logits. In the AUQUEI, the person-logit describes the child's life satisfaction level in health related daily-life scenarios; in the LVP-FVQ, this value indicates the child's FV performance in vision-related activities. Construct validity of the instrument is specified by the data fit and the form of relationship between respondents and the items. Furthermore, construct validity is indicated by the order of items determined by Rasch analysis, i.e., whether the ranking of the items is consistent with the construct theory.
The process of analysis used here began with the evaluation of the compliance of rating categories to a continuum of lower to higher difficulty/ability. Thus, a respondent selecting a lower category response would be rated lower in the respective latent trait. If categories were disordered, this would indicate either that the range of categories or that the names of categories were not applicable to the respondents. For example, a polychotomous scale may have been used, whereas a dichotomous scale would have been more applicable, and category names such as “happy” may be interpreted differently by the respondents than by the researchers.31 Disordered categories provided an indication that categories should be combined.35 The frequency distributions of categories were examined to identify categories that were redundant. Fit statistics were used to investigate the degree to which the data from items and people conformed to the expectations of the Rasch model (i.e., those easy items are easy for all people).36 Persons with poor-fit statistics (e.g., providing contradictory responses) were identified, response patterns were examined, and the data relating to that person were excluded from analysis if appropriate (e.g., if the person clearly misunderstood a question). Poorly fitting items were also identified as overfit (mean square value of item <0.8 Rasch index), indicating low variation in response pattern, probably because of redundancy, or as underfit (mean square value >1.3), which suggested that the response pattern is unlikely to conform with the Rasch model.4 Moreover, the unidimensionality of each instrument was tested using fit statistics and Rasch residual-based principal component analysis (PCAR) to determine whether items were consistent with a single underlying construct. PCAR provides an indication of the dimensionality of instruments (an indication of the range of factors addressed by the instrument).
Of the 200 children invited to participate, 168 completed the questionnaires, with signed informed caregiver consent obtained. On the basis of school records, subjects were classified by severity of ID according to the Wechsler Intelligence Scale for Children. Fifty-three subjects (31.5%) with IQ of 55 to 70 (mean = 60.8, SD = 4.1) were classified as having mild ID, and 115 (68.5%) with ID of 40 to 54 (mean = 44.5, SD = 3.5) were classified as having moderate ID. Subjects with IQ <40 (severe or profound ID) and/or otherwise unable to respond to the questionnaires were not included in this study. The mean subject age was 14.3 years (SD = 3.0), and 57% of the respondents were male. Of the 107 subjects who underwent vision screening, 35 (32.7%) had some form of visual abnormality as defined here (see Methods). All 35 were referred for further assessment. Of these, 11 achieved corrected acuity in the poorer eye of at least 20/30, indicating that the visual abnormality was uncorrected refractive error in these cases. The remaining 24 (22.4%) had corrected acuity poorer than 20/30 in at least one eye.
After initial analysis for appropriateness, four items of the AUQUEI and six items of the LVP-FVQ, which did not provide normally distributed data, were identified (items with “superscript a” in Tables 1 and 2).
Fig. 1 shows children's FV (degree of activity difficulty) for the original 22-item LVP-FVQ. The right side of the logit scale indicates the number of subjects with a given logit value on FV. Person and item appear in descending order of ability and difficulty on a common scale. Item-person misalignment was found in the LVP-FVQ, shown by lack of appropriate items for participants at the top of the scale. Thus, items in this questionnaire described tasks that were well within the ability of the respondents. A more even spread of items was obtained for the 17-item AUQUEI scale (degree of activity enjoyment) after item reduction (Fig. 2), indicating that AUQUEI items align better with respondents than LVP-FVQ items. The overall fit results may be summarized as follows (Tables 1 and 2): infit range, 0.7 to 1.5; outfit range, 0.4 to 1.9.
Eight of 23 items in the AUQUEI questionnaire were found to have disordered thresholds. This indicates that the rating scale of this questionnaire needs to be redesigned for use in this population. In its original form, the AUQUEI rating scale consists of four categories: “not happy at all,” “not happy,” “happy,” and “very happy.” The least frequently selected category in this questionnaire was “not happy,” being not used by 132 of 168 participants. Subsequent to initial Rasch analysis, this underutilization was overcome by combining response 1 “not happy at all” and response 2 “not happy.” Consequently, the scores for all items were recoded by collapsing four categories to three categories. The category collapse improved targeting of items to subjects, as indicated by a reduction in the difference between the mean value for the children and the mean value of the item from 1.30 to 0.5. In addition, category reduction also succeeded in improving person separation from 2.08 to 2.41. This improvement is also reflected in item and person reliability, as given in Table 3. With further item modifications, all items showed ordered thresholds. The LVP-FVQ dichotomous rating scales yielded no disordered thresholds.
Consistent with previous work on questionnaire development,36 Rasch indices, such as overall model fit, item and item/person separation reliability, and difference between the item difficulty and person ability were examined. Given the iterative process of item reduction, a number of criteria (Table 3) were followed during the procedure. In the AUQUEI, four items with high skew, kurtosis, and ceiling effect were indentified based on data distribution. Removal of these items from the current instrument resulted in improved targeting of items to the subjects (mean value difference from 0.50 to 0.21). Item 2 (“When I go to bed at night I feel …”) and item 13 (“When I watch television I feel …”) with poor-fit statistics (infit >1.5) and disordered categories were also eliminated, thus, a total of 6 items were excluded from the AUQUEI. Item removal could be continued to further improve item fit, but this would negatively affect person separation. For this reason, the modifications to this questionnaire were a compromise between item fit and person separation, with some items retained in the interests of maintaining a reasonable level of person separation. The original validation of the AUQUEI found internal consistency as indicated by Cronbach α of 0.88, which remained consistent with the shortened version (0.89); item reliability was also unchanged (0.99). In general, the levels of variance explained by Rasch measures >60% and variance not explained by Rasch measures but explained by first contrast <5% are accepted as an indication of the unidimensionality of items without redundancy in a construct.33 The variance of the data that can be explained by Rasch measures determined by PCAR was found to be 70.7% originally, increasing to 85.1% in the shortened version. The variance of data not explained by Rasch measures but explained by first contrast changed from 3.4 to 1.9%, respectively. These findings indicated that the Rasch model fit is improved after category and item reductions. Seventeen of 23 items remained in the revised AUQUEI. After a similar process, six items were removed from the LVP-FVQ. However, given the low person-separation reliability of 1.35, the instrument would not be appropriate for application to children with normal vision or mild visual abnormalities.
The hierarchy of items in AUQUEI as listed in Fig. 1 shows that the least enjoyable HQoL activities for children with ID are indicated by the items “When I am sick I feel …,” followed by “When I stay in hospital I feel …,” “When I take medicine I feel …,” and “When I go to the doctor I feel ….” In contrast, the most enjoyable HQoL activities are “When it is holiday I feel …,” “When I am playing outside I feel …,” “When I am at school I feel …,” and “When I am having dinner with family I feel ….” This pattern is consistent with expected degree of enjoyment of a diverse range of daily activities. No such pattern was derived from the LVP-FVQ.
Because the two questionnaires selected here were initially developed for use in children with normal development, our refinement process needed to take into account the cognitive and other characteristics of the population of children with ID. This study included children with Down syndrome, cerebral disorders, and disability caused by brain trauma. These groups differ considerably in their behavioral characteristics and in their cognitive capacity, and for this reason, the process of development and refinement of the questionnaires was conducted with this range in mind to maximize the likelihood that children in these groups would comprehend the meaning of items. This aspect of development is discussed more fully elsewhere.24,31 Despite this variation within our subject group, Rasch analysis identified misalignments between some items and subjects' lifestyles and capabilities, suggesting that these items were not appropriate for this population. Further consideration and perhaps rewording are needed for these items.
The LVP-FVQ was initially developed and validated in a population of children with low vision and normal development. However, it was unknown whether this instrument was valid in application to children with ID. In this study, the psychometric properties of the LVP-FVQ, such as person separation, did not reach acceptable levels, even though item reduction was attempted to better target the population of children with ID. Low item precision and misalignment might be attributed to the fact that the original items were used for children with low vision. Thus, many items may have been relatively easy for the population tested here. This is not surprising since just over one-third of the children tested here had a form of visual abnormality, but most did not have low vision. Further study in children with ID who have a wide range of visual abilities is required to gain a better understanding of the contribution of vision to HQoL. Given the limitations of LVP-FVQ revealed in this study, self-perceived FV in children with ID and the extent to which vision impacts on HQoL in this group remain unclear.
Rasch analysis indicated disordered categories in the AUQUEI, and modifications allowed correct ordering. For this reason, the instrument was modified with the intention of ensuring that thresholds were correctly ordered. The disordered threshold on the four-category scale suggested that simplification of the rating scale was required when children were asked to describe attitudes towards HQoL. Level of enjoyment of daily activities revealed by Rasch analysis demonstrated that children with ID are able to indicate the way in which activities affect their HQoL.31 Thus, the AUQUEI seems to be a useful tool for assessment of HQoL in children with ID.
In this study, the relationship between self-reported FV and HQoL could not be explored, given the limited person separation of the LVP-FVQ. In addition, criterion validity of the instruments in terms of the extent to which the instruments are related to other measures, such as visual function,7 is an unknown factor. Further study on the relationship between HQoL, FV, and measures of visual function such as acuity and contrast sensitivity in children with ID would help to determine the extent to which visual abnormalities impact on quality of life in this group.
Rasch analysis allowed us to modify instruments that had previously been validated for a different population and to test their applicability to the population tested here. The AUQUEI is useful in providing information on HQoL in children with ID, whereas LVP-FVQ was not found to be an appropriate instrument for the assessment of FV in children without visual impairment. A new instrument with appropriate items is required to address questions on the association between visual function and quality of life in children with ID.
We thank children, parents/caregivers, and staff of Da Ming School for Special Education in Ningbo, and Lu Jia Zui School for Special Education in Shanghai, China, for their participation in this project.
International Center for Eyecare Education
University of New South Wales
Sydney, New South Wales 2052
1.Quinn GE, Dobson V, Saigal S, Phelps DL, Hardy RJ, Tung B, Summers CG, Palmer EA. Health-related quality of life at age 10 years in very low-birth-weight children with and without threshold retinopathy of prematurity. Arch Ophthalmol 2004;122:1659–66.
2.Wolffsohn JS, Cochrane AL. Design of the low vision quality-of-life questionnaire (LVQOL) and measuring the outcome of low-vision rehabilitation. Am J Ophthalmol 2000;130:793–802.
3.Wolffsohn JS, Cochrane AL, Watt NA. Implementation methods for vision related quality of life questionnaires. Br J Ophthalmol 2000;84:1035–40.
4.Pesudovs K, Burr JM, Harley C, Elliott DB. The development, assessment, and selection of questionnaires. Optom Vis Sci 2007;84:663–74.
5.Felius J, Stager DR Sr, Berry PM, Fawcett SL, Stager DR Jr, Salomao SR, Berezovsky A, Birch EE. Development of an instrument to assess vision-related quality of life in young children. Am J Ophthalmol 2004;138:362–72.
6.Birch EE, Cheng CS, Felius J. Validity and reliability of the Children's Visual Function Questionnaire (CVFQ). J AAPOS 2007;11:473–9.
7.Gothwal VK, Lovie-Kitchin JE, Nutheti R. The development of the LV Prasad-Functional Vision Questionnaire: a measure of functional vision performance of visually impaired children. Invest Ophthalmol Vis Sci 2003;44:4131–9.
8.Nirmalan PK, John RK, Gothwal VK, Baskaran S, Vijayalakshmi P, Rahmathullah L. The impact of visual impairment on functional vision of children in rural South India: the Kariapatti Pediatric Eye Evaluation Project. Invest Ophthalmol Vis Sci 2004;45:3442–5.
9.Chak M, Rahi JS; British Congenital Cataract Interest Group. The health-related quality of life of children with congenital cataract: findings of the British Congenital Cataract Study. Br J Ophthalmol 2007;91:922–6.
10.Van der Steen-Kant SP, Arentz GM, Damen S. The impact of a visual impairment on the quality of life of persons with an intellectual disability. Part I: missed opportunities? J Intellect Disabil Res 2000;44:499.
11.Arentz GM, Damen S, Van der Steen-Kant SP. Impact of a visual impairment on the quality of life of people with intellectual disability. Part 3: detecting a visual handicap. J Intellect Disabil Res 2000;44:196.
12.Arentz GM, Damen S, Van der Steen-Kant SP. Impact of a visual impairment on the quality of life of people with intellectual disability. Part 2: understanding the causes of a visual handicap. J Intellect Disabil Res 2000;44:196–7.
13.Tychsen L, Hoekel J, Ghasia F, Yoon-Huang G. Phakic intraocular lens correction of high ametropia in children with neurobehavioral disorders. J AAPOS 2008;12:282–9.
14.Maino D. Diagnosis and Management of Special Populations. St. Louis: Mosby; 1995.
15.Woodhouse JM, Adler P, Duignan A. Vision in athletes with intellectual disabilities: the need for improved eyecare. J Intellect Disabil Res 2004;48:736–45.
16.Woodhouse JM, Adler PM, Duignan A. Ocular and visual defects amongst people with intellectual disabilities participating in Special Olympics. Ophthalmic Physiol Opt 2003;23:221–32.
17.Cregg M, Woodhouse JM, Pakeman VH, Saunders KJ, Gunter HL, Parker M, Fraser WI, Sastry P. Accommodation and refractive error in children with Down syndrome: cross-sectional and longitudinal studies. Invest Ophthalmol Vis Sci 2001;42:55–63.
18.Cregg M, Woodhouse JM, Stewart RE, Pakeman VH, Bromham NR, Gunter HL, Trojanowska L, Parker M, Fraser WI. Development of refractive error and strabismus in children with Down syndrome. Invest Ophthalmol Vis Sci 2003;44:1023–30.
19.Stewart RE, Woodhouse JM, Cregg M, Pakeman VH. Association between accommodative accuracy, hypermetropia, and strabismus in children with Down's syndrome. Optom Vis Sci 2007;84:149–55.
20.Cummins RA. Living with support in the community: predictors of satisfaction with life. Ment Retard Dev Disabil Res Rev 2001;7:99–104.
21.Koller H, Richardson SA, Katz M, McLaren J. Behavior disturbance since childhood among a 5-year birth cohort of all mentally retarded young adults in a city. Am J Ment Defic 1983;87:386–95.
22.Linna SL, Moilanen I, Ebeling H, Piha J, Kumpulainen K, Tamminen T, Almqvist F. Psychiatric symptoms in children with intellectual disability. Eur Child Adolesc Psychiatry 1999;8(Suppl 4):77–82.
23.Lanners J, Piccioni A, Fea F, Goergen E. Early intervention for children with cerebral visual impairment: preliminary results. J Intellect Disabil Res 1999;43(Pt 1):1–12.
24.Cummins RA. Self-rated quality of life scales for people with an intellectual disability: a review. J Appl Res Intellect Disabil 1997;10:199–216.
25.Cummins RA. Self-rated quality of life scales for people with an intellectual disability: a reply to Ager & Hatton. J Appl Res Intellect Disabil 2001;14:1–11.
26.White-Koning M, Arnaud C, Bourdet-Loubere S, Bazex H, Colver A, Grandjean H. Subjective quality of life in children with intellectual impairment—how can it be assessed? Dev Med Child Neurol 2005;47:281–5.
27.Vignes C, Coley N, Grandjean H, Godeau E, Arnaud C. Measuring children's attitudes towards peers with disabilities: a review of instruments. Dev Med Child Neurol 2008;50:182–9.
28.Hatton C, Ager A. Quality of life measurement and people with intellectual disabilities: a reply to Cummins. J Appl Res Intellect Disabil 2002;15:254–60.
29.Manificat S, Dazord A, Cochat P, Nicolas J. [Evaluation of the quality of life in pediatrics: how to collect the point of view of children.] Arch Pediatr 1997;4:1238–46.
30.Manificat S, Dazord A, Cochat P, Morin D, Plainguet F, Debray D. Quality of life of children and adolescents after kidney or liver transplantation: child, parents and caregiver's point of view. Pediatr Transplant 2003;7:228–35.
31.Cui Y, Stapleton F, Suttle C. Developing an instrument to assess vision-related and subjective quality of life in children with intellectual disability: data collection and preliminary analysis in a Chinese population. Ophthalmic Physiol Opt 2008;28:238–46.
32.Linacre JM. Winsteps, Version 3.67 (Computer Program). Beaverton, OR: Winsteps; 2008.
33.Linacre JM. A User's Guide to Winsteps: Rasch-Model Computer Program. Chicago: Mesa Press; 2002.
34.Wright BD, Masters GN. Rating Scale Analysis. Chicago: Mesa Press, 1982.
35.Lamoureux EL, Pallant JF, Pesudovs K, Hassell JB, Keeffe JE. The Impact of Vision Impairment Questionnaire: an evaluation of its measurement properties using Rasch analysis. Invest Ophthalmol Vis Sci 2006;47:4732–41.
36.Pesudovs K, Garamendi E, Keeves JP, Elliott DB. The Activities of Daily Vision Scale for cataract surgery outcomes: re-evaluating validity with Rasch analysis. Invest Ophthalmol Vis Sci 2003;44:2892–9.
aVisual abnormalities were defined as (1) uncorrected visual acuity >2 line interocular difference or better eye visual acuity worse than 20/30; (2) anisometropia >1.00 D (cylindrical or spherical); (3) manifest strabismus, any; (4) hyperopia >+3.50 D in any meridian; (5) myopia >−3.00 D in any meridian; (6) media opacity, any; (7) astigmatism >1.5 D at 90 or 180°, >1.00 D at oblique axis.
Keywords:© 2010 American Academy of Optometry
quality of life; questionnaire; intellectually disabled; children; Rasch analysis