Impact of vision on generic health‐related quality of life – A systematic review

The aim of the study was to provide an overview on the relation between vision and generic health‐related quality of life (HRQoL) in general populations worldwide. We conducted a systematic review to identify articles that have evaluated the relation of binocularly measured distance visual acuity (VA) and HRQoL in adults at population level around the world. Used search terms were related to visual acuity, vision loss, visual impairment and HRQoL. We identified 780 articles, of which 12 satisfied the eligibility criteria. The studies were conducted in the USA, Europe, Asia and Australia. Generic HRQoL was measured using EQ‐5D, 15D, SF‐36, SF‐12 and/or WHOQoL. Nine studies reported a trend between decreasing VA and deteriorating HRQoL of at least some degree. Vision loss was mostly associated with dimensions related to physical health and capability, including self‐care, mobility and usual activities. Mental dimensions – vitality and social functioning in particular – were also affected, although the results were inconsistent. Six studies compared the impact of different medical conditions on HRQoL, all of which reported that the impact of visual impairment was equal or stronger compared with other major medical conditions. In conclusion, generic HRQoL instruments are sensitive to detect the relation between vision and HRQoL, as they cover dimensions influenced by vision. Declining vision is associated with reduced functioning, which leads to dependence and isolation – both major factors in an individual's well‐being. This phenomenon showed surprising similarity around the world and therefore emphasises the global burden of vision loss at both individual and societal level and encourages its prevention and treatment.


| I N T RODUC T ION
Declining vision has been shown to have a detrimental impact on an individual's quality of life (QoL) by limiting mobility, visually intensive tasks and independence (Fenwick et al., 2016;Kempen et al., 2012;Lamoureux et al., 2004;Purola et al., 2021;Taipale et al., 2019;West et al., 2002).The absence of mobility and self-care reduces social activities and activities of daily living and therefore can lead to social isolation (Alma et al., 2011;McLaughlin et al., 2010).In addition, visual impairment is associated with an increased risk of accidents, injuries and even mortality (Black & Wood, 2005;Bramley et al., 2008;McCarty et al., 2001;Ramulu et al., 2012).
However, there is a need to improve the understanding of causes that lead to reduced mental health among visually impaired.The determination of factors influencing QoL in vision loss is crucial for public health to prioritise and allocate healthcare resources, as well as for individuals to reduce and prevent the burden of visual impairment on one's life.Furthermore, addressing this issue in different countries and cultures is important in providing a global overview of the situation, because different social and economic conditions and values and norms of local populations are all factors playing a role in the impact of disease and health problems on an individual's QoL (Schalock et al., 2005;Smith, 2004).

Impact of vision on generic health-related quality of life -A systematic review
Petri Purola 1,2  | Seppo Koskinen 3 | Hannu Uusitalo 1,2,4 In previous decades, the evaluation of an individual's health problems had focused on functional measurements.However, for a comprehensive assessment of an individual's health burden, it is crucial to estimate one's mental and physical well-being from their perspective (Fayers & Machin, 2013).Hence, there is an increasing interest in the evaluation of vision and eye care based on health-related quality of life (HRQoL) and patientreported outcomes that can provide a comprehensive picture of the burden of vision loss beyond clinical evaluation.HRQoL provides an overview on an individual's health utilities by assessing physical, psychological and social function and general health.
When evaluating the relation between vision and HRQoL, specific visual-related QoL instruments have been commonly used (Jones et al., 2017;Khorrami-Nejad et al., 2016;Machado et al., 2019).However, they are known to overestimate the impact of vision on the overall physical condition (Elliott et al., 2007;Frick & Massof, 2009).Generic instruments assessing HRQoL have been used less frequently in ophthalmological studies.These instruments are considered generic because the evaluated dimensions are not specific to any health condition or patient group.They are commonly used in clinical and economic appraisal and population health surveys.While no single HRQoL is considered to be the gold standard, these instruments can produce utility scores that are remarkably consistent with each other (Hawthorne et al., 2001).
Although there is a paucity on studies that have evaluated the relation between vision and generic HRQoL, a few analyses have been conducted using population representative data.Considering the generalisable and comprehensive nature of the generic HRQoL instruments, these studies can provide comparable and robust results and interpretations despite the scarcity in their numbers.Therefore, we conducted a systematic review on the relation of vision and generic HRQoL based on population-based studies around the world, including the impact of declining vision on overall QoL and its dimensions.More specifically, this review aims to scope the sensitivity and comparability of different HRQoL instruments in detecting the relationship between deteriorating vision and HRQoL.The secondary aim is to compare this relationship in different populations based on ethnicity, sex and age, as well as to eye diseases and other medical conditions.Generic HRQoL instruments used in the studies were EQ-5D, 15D, SF-36, SF-12 and WHOQoL.We focused on bilateral, habitual distance visual acuity (VA), as distance VA has been reported to have a more substantial impact on QoL than near VA (Taipale et al., 2019).

| Search and selection process
The entire workflow of this systematic review is illustrated in Figure 1, including used search terms and number of included and excluded articles.We performed this systematic review following the guidelines according to the Preferred Reposting Items for Systematic Reviews and Meta-Analysis (PRISMA 2020) (Page et al., 2021).
One investigator (PP) performed a systematic search of medical literature published on PubMed, PubMed Central and Web of Science before 11 January 2022.Reference lists of the included articles were also searched.Search terms included alternatives in relation to visual impairment and HRQoL, and the search model was following: ("visual impairment" OR "vision loss" OR "low vision" OR "impaired vision" OR "blindness" OR "visual acuity" OR "vision" OR "functional visual ability" OR "visual function") AND ("health related quality of life" OR "quality of life").All types of study design were included.Reviews and meta-analyses were not analysed but were used to find more relevant articles.
After the initial title-based search, duplicates were removed, and two levels of screening were performed based on the abstract and full-texts.Abstracts that provided sufficient detail for exclusion were removed, and the remaining full-text articles were retrieved to be fully analysed.The abstract-level screening was performed by one investigator (PP), and the full-text articles were read independently by two investigators (PP and HU) to determine inclusion for the final assessment.
We had the following inclusion criteria: original research reported in English; studies with randomised population-based sample; studies addressing adults aged 18 years or older; studies that have assessed the relation between generic HRQoL and binocularly measured habitual distance VA.We excluded studies with only vision-related quality of life, uncommonly used HRQoL instrument (QoL instrument developed for specific country or condition/situation, e.g.marriage), self-reported vision, studies with comparisons between less than three vision groups (e.g.blindness vs. normal vision) and/or study population with specific condition/disease/treatment (e.g.visually impaired population or glaucoma patients).

| Generic health-related quality of life instruments
The generic HRQoL instruments included in this review are all commonly used instruments which have been applied to various medical conditions and are generally considered valid and reliable within different health states.Details of the evaluated instruments are listed in Table 1.
The EQ-5D (EuroQol-5 Dimension) is a standardised measure of HRQoL developed by the EuroQol Group (Brooks, 1996).It assesses health status in terms of five dimensions of health and independence.A three-level response form of the instrument (EQ-5D-3L, later referred as EQ-5D) was used in all evaluated studies.A similar, 15-dimensional 15D instrument has been developed in Finland, which includes five dimensions that overlap with EQ-5D: mobility, usual activities, discomfort/symptoms, depression and distress (Sintonen, 2001).Although EQ-5D is more commonly used, 15D is considered more comprehensive due to its wider coverage of dimensions (Hawthorne et al., 2001).Despite the overlapping dimensions, correlation between the instruments has been reported to vary between different diseases and conditions (Kontodimopoulos et al., 2012;Saarni et al., 2006;Vartiainen et al., 2017).However, both instruments are known to correlate well with VA (Taipale et al., 2019).
The SF-36 (36-Item Short-Form Health Survey) is one of the most commonly used HRQoL instruments in the world (Busija et al., 2020).It contains 36 items measuring eight domains, with physical component summary score (PCS) and mental component summary score (MCS) (Ware & Sherbourne, 1992).The SF-12 (12-Item Short Form Health Survey) derives summary scores from specific items from the eight domains of the SF-36 into PCS and MCS (Ware et al., 1996).The SF-36 and its shorter derivative SF-12 are reported to correlate well with each other in various medical conditions (Lacson et al., 2010;Lam et al., 2013), although the domains are better represented by the more comprehensive SF-36.Both instruments have been reported to be sensitive to vision (Cypel et al., 2017;Davidov et al., 2009).
The WHOQoL (World Health Organization QoL) instrument measures six domains varying from physical to psychological (The Whoqol Group, 1998).It was used only in one study in this review, and its abbreviated form, WHOQoL-BREF, is generally more used.The WHOQoL has shown moderate correlation with EQ-5D and SF-36 (Chan et al., 2014;Chokotho et al., 2017).

| Summary of the search and selection
We identified 780 articles, of which we selected 12 articles that were relevant for the study.The selected studies were conducted in the USA, Europe, Asia and Australia.Seven studies were cross-sectional, two longitudinal and three included both designs.Eight studies analysed a general adult population, one study examined an older

| Relation between vision and overall health-related quality of life
We evaluated the sensitivity of the used instruments in detecting the impact of VA on HRQoL, and the results are summarised in Table 3. Overall, a monotonic association between VA and HRQoL was reported for each instrument.Only one study out of nine reported no significant association between HRQoL and VA.
Based on an eye study in the Australian population, Chia et al. (2004) reported an association between increasing visual impairment (VA 0.25-0.40,VA 0.10-0.20,and VA < 0.10) and lowering MCS of SF-36 after adjusting for age and sex.No significant trend was observed with PCS of SF-36.
Among the Latino population in the USA, Varma et al. (2006) did not observe significant differences in SF-12 between persons without visual impairment (VA > 0.50), persons with mild visual impairment (VA 0.32-0.50)and persons with moderate/severe visual impairment (VA ≤ 0.25) after adjusting for age, sex and comorbidities.Still, a small effect size was associated with poorer MCS of SF-12 and a moderate effect with poorer PCS of SF-12 when persons with any severity of visual impairment were compared with those without visual impairment.Esteban et al. (2008) reported that the mean PCS of SF-12 decreased in Spanish older women with declining VA (VA ≥0.32, VA 0.10-0.25,and VA ≤ 0.063) even after adjusting for age and eye diseases.No significant differences in MCS were observed as vision worsened.
In a study based on ambulatory community-dwelling Chinese population, Leung et al. (2012) reported a significant trend between decreasing VA (VA ≥ 0.63, VA 0.32-0.5, and VA ≤ 0.25) and both PCS and MCS of SF-12.However, after adjusting for age, sex, education and co-morbidities, better VA was significantly associated with better PCS only when the best vision (VA ≥ 0.63) and the worst vision (VA ≤ 0.25) groups were compared.Moreover, no association between decreasing VA and MCS of SF-12 was found after adjusting for confounders.Wu et al. (2021) observed a significant correlation between decreasing VA (VA ≥ 0.50, VA 0.10-0.40,and VA < 0.10) and decreasing EQ-5D index score in both rural and urban populations in China.The impact of visual impairment on EQ-5D was more severe in urban population, although they also showed lower EQ-5D in general in comparison to rural population.
Based on a national health survey in South Korea, Park et al. (2015) reported that persons with moderate (VA 0.125-0.25)and severe visual impairment (VA ≤ 0.10) showed significantly poorer EQ-5D index scores than those with normal vision (VA ≥ 0.80) after adjusting for age, sex, socio-demographic status, lifestyle and comorbidities.However, persons with mild visual impairment (VA 0.32-0.63)did not show significant difference from those with normal vision.
T A B L E 1 Generic health-related quality of life (HRQoL) instruments used in the reviewed studies.

HRQoL instrument Reference Dimensions Scores
European Quality of Life-5 Dimensions (EQ-5D)  Three studies were conducted based on a nationwide health examination survey in Finland.Taipale et al. (2019) reported that when compared with persons with good vision (VA ≥ 1.00), all four declined vision groups (VA 0.63-0.80,VA 0.32-0.50,VA 0.10-0.25,and VA < 0.10) showed statistically significant and clinically meaningful decrease in EQ-5D and 15D index scores.Furthermore, the detrimental impact on HRQoL increased as the vision worsened.Purola et al. (2021Purola et al. ( , 2022) ) reported a trend between decreasing EQ-5D and 15D index scores and decreasing VA independent of glaucoma, unoperated cataract and retinal degeneration.

| Relation between vision and dimensions of health-related quality of life
We evaluated the impact of VA on the dimensions of the EQ-5D, 15D and SF-36 based on six studies (Table 3).The dimensions of SF-12 and WHOQoL were not investigated in any of the studies.
Two studies conducted in Australia investigated the dimensions of SF-36.Chia et al. (2004) reported that persons with mild visual impairment (VA 0.25-0.40)had significantly lower scores in the physical functioning dimension of PCS and vitality and mental health dimensions of MCS .Persons with moderate-to-severe visual impairment (VA < 0.25) had lower scores in role limitation due to physical problems dimension of PCS, and vitality, social functioning and mental health dimensions of MCS.Vu et al. (2005) reported using partial SF-36 that visual loss (VA < 0.50) was associated with dependency and emotional well-being after adjusting for age, sex and co-morbidities.
Four studies investigated the dimensions of EQ-5D, two of which also included 15D.Based on the Finnish health examination data, Taipale et al. (2019) reported that an increasing proportion of respondents reported difficulties in every EQ-5D dimension except for pain/discomfort with decreasing VA even after adjusting for age, sex and co-morbidities.Based on the same data, Purola et al. (2021) reported that the most affected dimensions in individuals with visual impairment (VA ≤ 0.25) were usual activities and mobility in EQ-5D, and vision, usual activities and vitality in 15D.In addition, these dimensions were affected by decreasing VA regardless of whether the person had glaucoma or not (Purola et al., 2022).Wu et al. (2021) reported differences in Chinese rural and urban population: among rural people, visual impairment had a detrimental impact on self-care, mobility and usual activities of EQ-5D, whereas among urban people also anxiety/depression and pain/discomfort were affected.

| Longitudinal impact of vision on healthrelated quality of life
Among the Latino population in the USA, McKean-Cowdin et al. (2010) reported that participants with a ≥ 2line improvement in VA showed small improvement in MCS of SF-12 during a 4-year follow-up.No measurable differences in SF-12 scores were found in participants with declined VA or no change in VA during the follow-up.An 11-year longitudinal data set representative of the Finnish population between 2000 and 2011 was utilised in two studies (Purola et al., 2021;Taipale et al., 2019).Taipale et al. (2019) reported that during the 11-year follow-up, a decline in VA was statistically significantly associated with decreased EQ-5D and 15D index scores even after adjusting for sex and incidence of common co-morbidities.The decline in VA was also associated with clinically meaningful decrease in 15D.However, Purola et al. (2021) reported a statistically and clinically significant improvement in the mean HRQoL among visually impaired (VA ≤ 0.25) between 2000 and 2011.In relation to this, increasing severity of vision loss showed a broader impact on the EQ-5D dimensions in 2000 with a significant effect in all dimensions except pain/discomfort, whereas in 2011 only one dimension (usual activities) was significantly affected (Taipale et al., 2019).

| Relation between health-related quality of life and vision in different populations
We observed the relationship between VA and HRQoL in different populations based on ethnicity, age and gender, and the results are summarised in Table 4.
Regarding ethnicity, all HRQoL instruments were observed sensitive in detecting VA, with very little difference between nations and populations.Lowest sensitivity was observed in the adult Latino population living in the USA using SF-12 (McKean-Cowdin et al., 2010;Varma et al., 2006).EQ-5D showed somewhat lower sensitivity in Korean population (Park et al., 2015) regarding visual impairment in comparison with Finnish and Chinese populations.
When age was investigated, three studies out of four reported that HRQoL decreased as age increased among both visually impaired and population with normal vision.Esteban reported that in older men, age showed stronger detrimental impact on PCS of SF-12 than VA (Esteban et al., 2008).On the other hand, based on EQ-5D, Park et al. (2015) reported that HRQoL was worse among visually impaired (VA ≤ 0.63) independent of age and Purola et al. (2021) reported that the impact of impaired vision (VA ≤ 0.25) on HRQoL was stronger than that of age.
Three out of four studies reported significant gender differences in the relationship between HRQoL and vision.Nutheti et al. (2006) reported that women showed lower HRQoL than men independent of visual impairment (VA ≤ 0.25) based on WHOQoL .Esteban et al. (2008) reported that both PCS and MCS of SF-12 were worse in women than in men for all categories of T A B L E 3 Sensitivity of evaluated generic health-related quality of life (HRQoL) instruments on visual acuity (VA).

Studies (and study numbers) Association between HRQoL and VA Association between HRQoL dimensions and VA
European Quality of Life-5 Dimensions (EQ-5D) vision loss, those with normal vision.Park et al. (2015) reported decreased HRQoL among visually impaired (VA ≤ 0.63) independent of gender based on EQ-5D.Purola et al. (2021) reported that women were associated with decreased HRQoL independent of visual impairment (VA ≤ 0.25) based on EQ-5D, but not on 15D.

| Relation between vision and eye diseases or other major medical conditions
The impact of VA on HRQoL was compared with different eye diseases and other medical conditions, and the results are summarised in Table 5.In all six studies that investigated this relationship, the impact of VA on HRQoL was found equal or stronger than that of major medical conditions using any of the evaluated instruments.Chia et al. (2004) observed in Australian population that the impact of visual impairment (VA < 0.50) on MCS of SF-36 was much greater than that of other medical conditions, which included arthritis, stroke, heart attack, asthma, diabetes and angina.The impact of visual impairment on PCS was smaller than that of diabetes, stroke, angina and heart attack.Esteban et al. (2008) reported that in older Spanish women, the impact of VA on PCS was stronger than that of glaucoma, cataract, diabetic retinopathy and late age-related maculopathy.
Based on an eye disease study in India, Nutheti et al. (2006) reported that visual impairment (VA 0.10-0.25),blindness (VA < 0.1), hypertension, diabetes, major physical illness, corneal disease and glaucoma were significantly associated with decreased WHOQoL score.Cataract or retinal disease were associated with deteriorated HRQoL as a function of VA, whereas glaucoma and corneal disease were independent of VA.Visual impairment from uncorrected refractive errors was not associated with decreased WHOQoL score.Overall, blindness showed the most severe impact on WHOQoL score in comparison with all other factors.
Based on the Finnish health examination data, Purola et al. (2021) reported that the impact of visual impairment (VA ≤ 0.25) on EQ-5D and 15D was stronger than that of glaucoma, unoperated cataract, retinal degeneration, heart disease, pulmonary disease, vascular disease, musculoskeletal condition, hypertension, diabetes and cancer.Only psychiatric disorders and Parkinson's disease showed stronger impact on HRQoL.Furthermore, visual impairment associated with glaucoma was the major determinant of reduced HRQoL among glaucoma patients regardless of the awareness or the treatment of the disease (Purola et al., 2022).Based on the 11-year follow-up of the same data set, Taipale et al. (2019) reported that declining VA had stronger impact on both EQ-5D and 15D than any of the previously mentioned medical conditions based on their incidence.

| DI SC US SION
In this study, we reviewed 12 articles that had evaluated the association between vision and generic HRQoL Women showed lower HRQoL than men after adjusting for impaired VA (≤0.25) and other medical conditions (10).HRQoL was worse among visually impaired (VA ≤0.63) regardless of gender (8).
HRQoL declined as age increased among those with impaired VA (≤0.25) (10).No significant differences in gender after adjusting for impaired VA (≤0.25) and other medical conditions (10).
Not investigated.
Not investigated.
Both PCS and MCS were worse in women than in men for all categories of deteriorated VA, excluding those with normal vision (5).
HRQoL declined as age increased among those with or without visual impairment (VA ≤0.25) (3).
Women showed lower HRQoL than men with or without visual impairment (VA ≤0.25) (3).
in general The studies were conducted four continents and therefore provide a good picture at global level.Nine studies reported a trend between worsening vision and deteriorating HRQoL, although it was not always linear.Vision loss was mostly associated with dimensions related to physical health and capability, including difficulties in independence and activities dependent on vision, whereas the association with mental dimensions remained somewhat uncertain.Six studies showed visual impairment or blindness to have an equal or stronger impact on HRQoL than major medical conditions.The detrimental impact of visual impairment on HRQoL was reported to increase with age, and women reported lower HRQoL than men regardless of vision level.All in all, the results showed surprising similarity in different countries and communities around the world.Differences in study methodologies, vision measurements, classification of visual impairment and population characteristics make direct comparisons between previous studies difficult.We identified three (systematic) reviews that had previously investigated the relation between vision and HRQoL at population level (Nayeni et al., 2021;Nispen et al., 2020;Nyman et al., 2010).However, these reviews also included populations with specific conditions, interventions, as well as visualrelated QoL and instruments focusing on mental health.Although our criteria were rather strict, leading to low number of included studies, our approach allowed for more robust and generalisable results.
Overall, majority of the reviewed studies reported a positive correlation between vision and HRQoL, particularly at lower VA levels.Taipale et al. (2019) reported that the threshold of VA 0.5 and below is associated with greater and progressive decrease in HRQoL.This threshold has a substantial role in independence and consequent well-being, as it serves as the requirement for standard driving licence in many states in the USA and the countries of the European Union.However, the linearity of the correlation varied between studies regardless of study population or used HRQoL instrument.A potential explanation for this is the arbitrary cut-offs used to define the severity of vision loss, which can be insufficient to reflect the gradual deteriorating trend between vision and HRQoL.In addition, Park et al. (2015) suggested that persons may adapt to their own health state, and therefore the decrease in HRQoL may not be linear between mild visual impairment of newer cases and more severe forms of those lived with the condition longer.
T A B L E 5 Impact of different medical conditions on generic health-related quality of life (HRQoL) in comparison with visual acuity (VA).

Studies (and study numbers) Differences between eye diseases Differences between other medical conditions
European Quality of Life-5 Dimensions (EQ-5D) Dimensions that affected by vision were related to functioning and independence, such as mobility, care and usual activities.It has been suggested that this demonstrates the isolating effect of visual impairment, which has a detrimental impact on overall well-being (Varma et al., 2006).Mental well-being was also affected, although not as clearly as physical dimensions.Interestingly, Wu et al. (2021) reported that mental health was more affected among the urban population than the rural population, likely due to the major role of vision in education, occupations and everyday urban life.
All five generic HRQoL instruments included in the reviewed studies showed modest-to-strong sensitivity in detecting the impact of VA and visual impairment on HRQoL.SF-36 and its shorter variant, SF-12, were the most commonly used instruments.While SF-36 showed monotonic association between MCS and VA, there was more variation in SF-12, likely due to its smaller coverage of dimensions.Interestingly, while EQ-5D covered the least dimensions of the used instruments, in all five studies it was included it showed strong sensitivity comparable with that of 15D, despite it being a more comprehensive instrument.Considering that VA correlates strongly with mobility, self-care and usual activities that comprise most of the dimensions of EQ-5D, VA therefore likely impacts the index score of EQ-5D through these dimensions.WHOQoL was also reported to have a monotonic association with VA; however, as the instrument was used only in one study and one population, assessing the sensitivity of the instrument on this relationship requires more evidence.
In the two Finnish cross-sectional studies, the mean HRQoL among visually impaired was better in 2011 than in 2000 (Purola et al., 2021;Taipale et al., 2019), and it has been suggested that the impact of visual impairment and vision-threatening eye diseases on HRQoL has decreased.The overall well-being has also increased in Finland (Koskinen et al., 2012).Possible reasons for this development are the improvements and better availability of health services, aids and treatment.However, more longitudinal studies conducted in other countries are required to validate assess whether corresponding development has occurred at a global scale.
The HRQoL is known to be lower in patients with a chronic disease than in the general population (Saito et al., 2006).Based on six studies, the impact of visual impairment or blindness on HRQoL is comparable with that of major medical conditions (Chia et al., 2004;Esteban et al., 2008;Nutheti et al., 2006;Purola et al., 2021Purola et al., , 2022;;Taipale et al., 2019).Furthermore, the severity of visual impairment was associated with decreased HRQoL independent of underlying eye condition.Purola et al. (2021) suggested that the outcome of these diseases, visual impairment, is a more significant factor in deteriorated HRQoL than the awareness of the disease itself; hence, the information directed to the public on the risks and prevention of blindness should be encouraged and strengthened to prevent the deleterious effects of visual impairment.
Many other factors can also have an impact on HRQoL among visually impaired, such as social, economic and cultural status and demographic factors.For example, Park et al. (2015) reported that people of older age, low educational status, the unemployed and single or widowed persons were more likely to have visual impairment.Still, the relationship between HRQoL and vision was reported being surprisingly similar in different populations around the world.The sensitivity of the used instruments was relatively good regardless of ethnicity.Age was reported to have a decreasing impact on HRQoL; however, impaired vision showed significantly stronger impact on HRQoL than ageing (Purola et al., 2021).Gender inequalities regarding HRQoL are also well known, and women with visual impairment in particular show lower HRQoL than men (Esteban et al., 2008;Khorrami-Nejad et al., 2016;Leung et al., 2012;Nutheti et al., 2006).Interestingly, in Finland, the gender difference in EQ-5D among visually impaired was significant in 2000 but no longer in 2011, suggesting improved gender equality in the 2000s in this respect (Purola et al., 2021).
All in all, our review shows that generic HRQoL instruments are sensitive to detect the relation between vision and HRQoL at population level.This has important implications, because generic HRQoL instruments cover dimensions of well-being that are affected by impaired vision but are not necessarily measured by more visual-specific QoL instruments.Overall, this demonstrates the important role of vision in everyday life in different populations.Vision loss is associated with a significant detrimental influence on generic HRQoL around the world and therefore can be considered a global burden on the physical, social and mental health.
The main strength of our review is the populationbased data used in the studies, allowing for more robust and generalisable results and interpretations than clinical-based studies and interventions.In addition, seven studies reported high response rates.In all studies, VA was measured by healthcare professionals.The generic nature of the used HRQoL instruments allows the applicability of the results across countries and conditions.Although language limitations were observed during the screening phase, we were able to include studies with diverse ethnic population samples from different regions of the world; hence, we were able to provide a global overview on the subject.
There are limitations in our systematic review.Selfreported HRQoL can be subject to personal biases due to its subjective nature.Furthermore, there can be considerable variability on how the questionnaires were administered, and the use of different statistical methods can have an effect on the results (Saarni et al., 2006).The comparability of the articles was made difficult by differences in study designs, sample selections and inclusion criteria.There were also differences in the methodologies and classifications of vision and visual impairment.Clinical meaningfulness or effect sizes of the outcomes were evaluated only in six studies.
In conclusion, declining vision is associated with reduced functional status and well-being in different regions and communities around the world.As the prevalence of visual impairment increases globally, its burden on individual's QoL and public care is a growing global The chronic and often untreatable nature of visual impairment requires focusing on aims that improve vision, stop progression of eye diseases or at least postpone the onset of visual impairment or declining VA.Furthermore, therapeutic approaches should also put an emphasis on the subjective perception of the condition.In future, more population-based studies are needed to further assess the global burden of visual impairment and its development.There was a particular absence of studies from the Africa, Middle East and South America, despite their major part in global health.More longitudinal studies are also needed to obtain more evidence on the changes in visual impairment and its impact on HRQoL.Nevertheless, these results highlight the global burden of vision loss at both individual and societal level and therefore emphasise the importance of rehabilitation and social services in maintaining good HRQoL.

AC K NO W L E DGE M E N T S
This study was supported by Tampereen seudun Näkövammaisten tukisäätiö s.r, Tampere, Finland; Glaukooma Tukisäätiö Lux s.r, Helsinki, Finland; Elsemay Björn Fund, Helsinki, Finland; Finnish Federation of the Visually Impaired, Helsinki, Finland; Päivikki ja Sakari Sohlbergin Säätiö, Helsinki, Finland; and Juho Vainion Säätiö, Helsinki, Finland.Funding sources did not influence the study design, data collection, analysis, interpretation or writing of the publication.The decision of publishing the results was completely made by the authors.PP, SK and HU declare no conflict of interest.

12 articles selected for final assessment
Sensitivity of evaluated generic health-related quality of life (HRQoL) instruments in different populations.
T A B L E 4