Background
Patient-reported outcomes (PROs), together with Clinician-reported outcomes (CROs) and laboratory tests (or device measurements) are three types of endpoints of diseases. The major representative of formal PROs is reliable and validated multi-item questionnaires [
1]. There are many vision-related functional questionnaires [
2‐
4], such as Visual Functioning 14 (VF-14) [
5], NEI-visual functioning questionnaire 25 (NEI-VFQ 25) [
6] and Catquest nine-item short-form (Catquest-9SF) [
7,
8]. Catquest-9SF has been adopted by the International Consortium for Health Outcomes Measurement (ICHOM) to specifically measure the risk factors for and outcomes of cataracts, which are the main global cause of blindness and vision impairment [
9].
Originally, Catquest-9SF contained 19 questions, was available in Swedish and was used by the National Swedish Cataract Register to evaluate the visual disability of cataract patients [
10]. However, its nine-item short-form Rasch-scaled version (Catquest-9SF) was shown to be more reliable and valid in measuring the visual disability outcomes of cataract surgery [
7].
Currently, Catquest-9SF has been translated and culturally adapted, as well as validated, in Australia [
11], Germany and Austria [
12], Italy [
13], and the Netherlands [
14], among other countries. Recently, this questionnaire has been assessed by using Rasch analysis in Chinese populations [
15,
16].
However, unsolved problems issues arise on the Chinese Catquest-9SF. On the one hand, the results of Chinese Catquest-9SF is still controversial, one study claimed that all nine questions of Chinese Catquest-9SF are valid and reliable [
15], while the other study suggested that it is better remove item 7, for item 7 is misfitting [
16].
On the other hand, the aim of the study was also to find the direct evidence that the Catquest-9SF would be used as a routine clinical tool in community screening. Previous studies have mainly focused on the validation of cataract surgery candidates in a hospital setting, but a questionnaire valid in a hospital setting serves only as an indirect evidence it is also valid community screening. There is a usually overlooked difference between the two groups: for patients who were diagnosed as cataract in a community screening could have no awareness of their disease while cataract patients who go to a hospital to ask for clinical help usually are fully aware of the troubles which cataract bring to them. And to prove that Catquest-9SF could be used as a routine clinical tool in community screening, the direct evidence on the validity of Catquest-9SF in a community based population are needed. And the function of Catquest-9SF would be extended if it could be used in community screening.
Furthermore, as an indicator of CROs, Lens Opacities Classification System (LOCS) III cataract grading has been widely used to assess lens opacities [
5]. Thus, by determining the correlation between Catquest-9SF scores and LOCS III cataract grading, we can further discover whether Catquest-9SF could reflex the opacity of lens.
Discussion
The subjects of previous studies on Catquest-9SF validation were mostly hospitalized patients expecting cataract surgery (patients awaiting cataract operation [
7,
12‐
16]) because they were not satisfied with their present vision and had visual disability. The results of our study showed that the Chinese Catquest-9SF was also valid and reliable for assessing cataract patients in community screening and that Catquest-9SF scores have a statistically significant correlation with LOCS III cataract grading. Both of these outcomes suggest that the Chinese Catquest-9SF partly reflects the severity of cataracts in Chinese population-based community screening.
Based on the MNSQ values and PCA, the Chinese Catquest-9SF has demonstrated good unidimensionality. Similar results were found in a study by Lin et al.
.... [
15] and for versions in other languages [
7,
11‐
14]. However, in a study by Wang et al, the question about “Seeing to do delicate work (Q8)” was removed from the questionnaire because it was deemed “ambiguous” and failed to demonstrate a good fit (outfit value > 1.3) [
16]. Wang et al attributed the misfit to the ambiguity of the word “delicate”. In our study, we elaborated on “delicate work” as “needlework, handiwork, carpentry, etc.” to ensure that patients thoroughly and accurately understood the meaning.
Our study demonstrated an ordered threshold in the category probability curves, which means that patients who responded that they had more visual disability for a certain item indeed had more visual disability for that item than people who claimedthat they had less disability. The combination of a good PSI and good PR suggested that the measurement precision of Catquest-9SF was good, which means that the instrument could accurately distinguish between low and high performers. In the case of Catquest-9SF, this finding specifically means that the measure could accurately distinguish between people with and without cataract-related visual disturbances. Our results were consistent with previous studies, regardless of the Catquest-9SF version used [
7,
11‐
16].
The person-item map showed significant mistargeting (1.39-logit difference in means) of persons and items, suggesting that the specified tasks were relatively easy for the cataract patients, even with decreased visual abilities. Better targeting was found in the study by Wang et al.
.. [
16]. This discrepancy may be partially due to the fact that we assessed cataract patients within a community-based population, who tend to have more satisfaction with their present vision and less visual disability in general than hospitalized patients expecting cataract surgery. Meanwhile, mistargeting (1.61 logits) was also found in the study by Lin et al [
15] and in studies of a Swedish version (1.95-logit difference in means) [
20], Italian version (2.04-logit difference in means) [
13], and Dutch version (1.64-logit difference in means) [
14]. Thus, the inclusion of additional items that could facilitate better targeting of items for visual abilities should be considered in future studies.
Significant correlation was found between some questions on Catquest-9SF and LOCS III grading, while a moderate correlation was observed for Q4 and posterior subcapsular grading. In addition, fair correlations were observed between “Seeing to walk on uneven ground (Q6)”, “Reading text on TV, in movie or on advertising board (Q7)” and “Seeing to do delicate work (needlework, handiwork, carpentry, etc.) (Q8)” and cortical grading for the better eye. Fair correlations were observed between “Vision difficulty in everyday life (Q1)” and nuclear colour and nuclear opalescence grading for the better eye. In a previous study by Skiadaresi et al.
...., no correlation was found between the general score on the Italian Catquest-9SF and LOCS III grading [
13]. However, in another previous study by Pan et al
, a significant and moderate correlation was found between the general score on another vision-related functional questionnaire (VF-14) and LOCS III grading (especially nuclear opalescence grading) [
5]. Instead of calculating the general score for all questions, we specifically calculated the score for each question. In addition, the sample size of the study by Skiadaresi et al was too small (only 24 patients with nuclear cataracts, 3 with cortical cataracts, and 25 with posterior subcapsular cataracts) to obtain significant results. Thus, Catquest-9SF can reflect lens opacities to some extent.
This study has a few limitations. First, all patients were recruited only from a single community, so larger and more representative samples are needed in future studies. Second, we only investigated the relationship between LOCS III grading and Catquest-9SF, other objective methods, such as lens density and objective scatter index measurement, could be used in future studies. Third, to determine whether Catquest-9SF could be useful as a screening tool, future studies between patients with and without cataract-related visual disturbances might be still needed.
Conclusion
In conclusion, the Chinese Catquest-9SF is a concise, valid and reliable questionnaire that is easy to understand and quick to complete for Chinese-speaking patients in Chinese community. Moreover, Catquest-9SF scores had a certain correlation with LOCS III grading. Thus, the Chinese Catquest-9SF is expected to be an auxiliary tool for preliminary cataract screening use.