Nationwide and regional trends in distance and near visual acuities during 2000–2017 in Finland

To evaluate time trends in distance and near vision at the national and regional levels during 2000–2017 in Finland.


| I N T RODUC T ION
Vision plays an important role in many everyday activities and the overall quality of life.As such, vision loss is a significant burden both on an individual and the society.At the individual level, decreasing vision is associated with detrimental development of one's quality of life and mental health by limiting daily and social activities (Purola et al., 2021;Purola, Koskinen, & Uusitalo, 2023;Taipale et al., 2019;Wu et al., 2021).Vision loss is also associated with an increased number of fall-and injury-related hospitalizations and premature retirement (Bramley et al., 2008;Kulmala et al., 2008Kulmala et al., , 2009;;Marques et al., 2021;Rein et al., 2006Rein et al., , 2022)).At the societal level, vision loss manifests itself as an increased use of health care resources and loss of productivity.Globally, over 1 billion people are estimated to have impairments in distance or near vision, and this number is expected to rise along with the number of vision-threatening diseases due to ageing of the population and increasing life expectancy (Bourne et al., 2021).Hence, it is vital to follow time trends in visual function at the population level and the impact of interventions performed to prevent vision loss.
Majority of the previous publications on vision have focused on visual impairment and specific populations.However, we lack information about the trends in visual acuities at the national level, in different regions of a country and different population subgroups.Therefore, the aim of the current study was to analyse the time trends in distance and near vision in Finland during 2000-2017 based on three nationwide population-based surveys.
The prevalence of different vision levels was estimated both at the national and regional levels.A comparison between urban and rural regions was also included.

| Study design and population
This study is based on a nationally representative sample of Finnish adults from three cross-sectional nationwide health examination surveys conducted by the Finnish Institute for Health and Welfare: Health 2000, Health 2011 and FinHealth 2017 (Aromaa & Koskinen, 2004;Koponen et al., 2018;Koskinen et al., 2012).The aim of the surveys was to collect comprehensive, up-to-date information on health, functional capacity and welfare.All surveys included comprehensive and comparable self-reported assessments as well as a health examination conducted at a nearby screening centre.If the invited participants did not attend the examination, an abridged examination was included in the Health 2000 and Health 2011 Surveys that were conducted at home or in an institution.
The first survey, Health 2000, was carried out in Finland in 2000-2001.A representative sample of the Finnish adult population was selected by utilizing a probabilityclustered sampling and weighting scheme.The sample included 9873 adults aged 18 years or older living in Finland.Those aged 30 years or older were invited to a health examination while younger subjects were interviewed and filled in a questionnaire.The participation rate for the health examination was 85%.The sample was drawn by a two-stage stratified cluster sampling, and persons aged 80 or older were oversampled by doubling the sampling fraction.The sample weights were calibrated by poststratification, defined by age, sex, region and native language to account for non-response and missing data.The second survey, Health 2011, was carried out in 2011-2012 and it included all living members of the Health 2000 Survey sample who had not refused further contact as well as a new sample of young adults.Altogether, 8550 subjects were invited to a health examination, and the participation rate was 59%.The third survey, FinHealth 2017, was carried out in 2017.It included a new sample of 10 247 adults aged 18 years or older, and the participation rate for the health examination was 58%.The details of the design and sampling of each survey have been described previously (Borodulin & Sääksjärvi, 2019;Heistaro, 2008;Lundqvist & Mäki-Opas, 2016).
For the present study, we included participants aged 30 years or older representing the Finnish adult population at the three time points: 7977 in 2000, 7964 in 2011 and 9288 in 2017.Separate weights were applied for the surveys to produce results representing the Finnish population at each time point.In the Health 2011 and Fin-Health 2017 surveys, specific population weights were created for the subjects who participated in the health examination.In the Health 2000 Survey, overall weights were created for subjects who participated in any part of the survey.These weights were used to extrapolate the number of persons in each vision group to the total population at each time point.
Information on the region of residence was available for all survey participants.It was classified according to the five university hospital districts that represent Southern, Western, Central, Eastern and Northern Finland, respectively.The total number of residents by age and sex at each district in 2000, 2011 and 2017 was obtained from the Digital and Population Data Services Agency.The region of residence was also classified into urban (cities), sub-urban (local centres in rural regions and rural regions close to cities) and rural (countryside) based on data available from 7778 (97.5%), 7798 (97.9%) and 5428 (58.4%) participants aged 30 years or older in 2000, 2011 and 2017, respectively.

| Vision measurement
The binocular distance and near visual acuity (VA) were measured in the health examination by a trained study nurse with current visual correction.Illumination was set to ≥350 lx on the modified logMAR letter chart (Ferris et al., 1982).Distance VA was measured at 4 m, and for the examination of near VA, the subjects held the chart at the distance where they could see it best.All measurements were standardized.For both the distance and near vision, the result was entered as the lowest line on which the subject correctly identified at least four letters.All VA values are presented as Snellen decimal equivalents.Low VA values outside the modified logMAR letter chart that could not be determined were reported as 0.01 (20/200).The methodology of recording VA was identical in all three time points.Based on our previous studies (Purola et al., 2021;Taipale et al., 2019), we classified distance and near VA values into the following groups: good vision (VA ≥ 1.0; VA ≥ 20/20), adequate vision (VA 0.63-0.8;VA 20/32-20/25), weak vision (VA 0.32-0.5;VA 20/63-20/40) and impaired or worse vision (VA ≤ 0.25; VA ≤ 20/80).In addition, we combined the last two groups into a weak or worse vision group (VA ≤ 0.5).Distance vision was missing from 4 participants in 2000, 2 participants in 2011 and 1 participant in 2017, who were excluded from respective analyses.

| Statistical analyses
All analyses were carried out using r software (v.4.2.1, R Core Team, R Foundation for Statistical Computing).The sampling design of the survey was accounted for using Survey Package 3.37 for R (Lumley, 2004) and weighting scheme calculated by the Finnish Institute for Health and Welfare.Population estimates were calculated using the function svytotal in the Survey package.Prevalence rates were estimated as the percentage of all survey participants with measured vision at each time point using function svyratio in the Survey package.Age groups ≥30 and ≥85 years were adjusted for age and sex in the regional investigation due to their heterogenic nature.Pearson's chi-squared test was used to compare proportions.All p-values were two-sided and were considered statistically significant when the values were <0.05.

| R E SU LT S
Eligible number of participants in the Health 2000, Health 2011 and FinHealth 2017 Surveys was 7977, 7964 and 9288, respectively.The mean age in each eligible sample was 54, 55 and 57 years, respectively, with a standard deviation ranging between 15 and 16 years.Vision data were available from 83.6%, 58.0% and 58.3% of the eligible participants in each respective survey, and the study sample consisted of these participants.The study sample demographics are summarized in Table 1.Overall, the study samples from the three time points shared similarities according to age and sex between each other and the original eligible samples.
The prevalence of different near vision levels in Finland is shown in Table S2 and visualized in Figure 2. Between 2000 and 2017, the prevalence of different near vision levels remained relatively the same among the entire adult population, but among the population aged 85 years and older improvements in similar to distance vision were observed: the prevalence of good near vision increased from 7.3% (95% CI 4.0%-10.7%)to 16.7% (95% CI 8.6%-24.8%)while impaired or worse near vision decreased from 30.5% (95% CI 23.8%-37.2%) to 7.5% (95% CI 0.8%-14.2%).These changes were larger from 2000 to 2011 than from 2011 to 2017.
The distribution of good and weak or worse distance vision in the five university hospital districts is illustrated in Figure S1 and the respective distribution of near vision in Figure S2.Overall, the improvements in vision seen at the national level were visible throughout different regions as well.Although there was more variance in older age groups than in younger ones between regions, particularly on lower vision levels, no outstanding trends were observed in any of the regions regarding different vision levels.
Time trends in the prevalence of good and weak or worse distance and near vision in urban, sub-urban and rural regions are shown in Figure 3.The positive time trends in vision seen at the national and regional levels were also visible in both urban and rural regions.Overall vision level was somewhat lower in rural regions compared to urban regions at all three time points: this difference was significant (p < 0.05) at all three time points regarding good near vision and in 2000 regarding weak or worse near vision.Only significance for distance vision was observed in 2000 regarding good distance vision.

| DI SC US SION
Several population studies have been conducted to collect information about vision in European countries (Cedrone et al., 2006;Klaver et al., 1998), USA (Killeen et al., 2023;Vitale et al., 2006), Asia (Xu et al., 2020;Yamada et al., 2010), Australia (Attebo et al., 1996) and Latin America and the Caribbean (Leasher et al., 2019).However, all these studies either represent regional data or a specific population or solely focus on visual impairment caused by a particular condition.Previous studies have reported that even mild vision loss can impact one's quality of life and mental health (Purola, Koskinen, & To fill this gap in knowledge, in the current study, we have analysed population-level time trends in distance and near vision in Finland from 2000 to 2017 based on three nationwide surveys.Overall, the level of distance and near vision has improved throughout the country with a larger improvement between 2000 and 2011 than between 2011 and 2017.The most noticeable improvements have occurred among the elder population with a positive shift from weak or impaired vision to good or adequate vision levels.Furthermore, no noticeable differences in vision levels were observed between the five main regions of Finland, and the small differences between urban and rural regions have remained fairly similar. There are many potential explanations why the overall vision level has improved after the year 2000.First of all, based on our recent studies, the number of patients suffering visual impairment caused by age-related macular degeneration, glaucoma or diabetic retinopathy has significantly decreased in the past decades (Purola et al., 2021;Purola, Kaarniranta, et al., 2023;Purola, Nättinen, et al., 2022;Purola, Ojamo, et al., 2022;Taipale et al., 2019;Vaajanen et al., 2022).The incidence of visual impairment due to these diseases has decreased or at least shifted to older age groups between the 1980s and the 2010s.This positive development could be explained by the improved awareness of the risks of these diseases, improved eye care, new aids, strengthened screening programmes in diabetic retinopathy and better and novel therapies.Second, the use of more up-to-date spectacles and other aids may have become more common, particularly after the economic depression that occurred in Finland during the 1990s.Third, a new Finnish legislation was introduced in 2005 (National Guaranteed Access to Healthcare), obligating medical procedures, for example, cataract operations, to be performed within 6 months after the notification.This has likely decreased cataract-related vision loss.
The decrease in the prevalence of weak or worse distance vision as shown in this study carries significant merits.VA of 0.5 (20/40) is the criterion used to determine eligibility for a driver's licence in many countries, and it has been associated with the point when vision starts to significantly reduce person's quality of life and limit daily functioning (Purola, Koskinen, & Uusitalo, 2023, Taipale et al., 2019).Therefore, this favourable shift from lower to higher vision levels is an important factor in retaining a good quality of life in the population.Furthermore, this change is also important for society by reducing the use of health care services and costs that are strongly associated with vision loss (Marques et al., 2021).
The prevalence vision loss is strongly associated with age, particularly from the age of 65 upward (Laitinen et al., 2005;Purola et al., 2021).In the current study, VA improved already among the middle-aged population, but the relative improvement was larger among those aged 65 or over, and the most prominent improvements in vision were observed among the population aged 85 years and older: the prevalence of weak or worse distance vision decreased by 61% and near vision by 57% between 2000 and 2017.As it is the most rapidly growing age group in Finland, sustaining this positive trend will be crucial.
Based on our results, near vision showed similar improvement during the 17-year follow-up as distance vision, although on a smaller scale.A similar shift from lower to higher near vision levels was observed among the population aged 85 years and older.Near vision loss is strongly associated with ageing due to the increasing risk of presbyopia.Like distance vision loss, near vision loss is associated with decreased quality of life (Taipale et al., 2019), socio-economic disparity (Wang et al., 2020) and loss of productivity (Frick et al., 2015).Despite this, there is a sparsity of near vision data worldwide, even more than of distance vision (Bourne et al., 2021).The analysis of near vision loss is more problematic due to the lack of definition of near vision impairment.For example, the World Health Organization did not have a definition of near vision impairment until 2019.In future, more nationwide or global health surveys need to incorporate the measurement of near vision to allow the estimation of levels and time trends in other nations.
Older age and socio-economic factors such as low education and low income are associated with lower distance and near vision in both low-and high-income countries (Killeen et al., 2023;Wang et al., 2017Wang et al., , 2020)).However, in the current study, the positive time trends in vision were quite similar throughout different regions of Finland.While previous studies have reported lower vision levels in rural populations compared to urban populations (Nakamura et al., 2010;VanNewkirk et al., 2001;Xu et al., 2006), our results show that the differences between urban and rural regions are quite small.This underlines the importance of equal and accessible health care throughout a country and improving the awareness of various eye diseases and their impact on the entire population.Vision interventions are reported to provide some of the largest returns on investment due to their lower costs and relatively high potential for cost recovery in areas such as cataract surgery compared with other health interventions (Brown et al., 2014).
The strength of our study was the use of data from three nationwide surveys with high participation rates allowing a relatively long follow-up period of 17 years.The decline in participation rates from 2000 to 2017 was relatively small and was further corrected by applying the weights.The survey samples represent the Finnish adult population at the three time points, and the surveys addressed health issues more comprehensively than national health surveys do on average.Furthermore, participants with mobility restrictions, which may have limited their participation in the health examination, were offset by conducting home visits in 2000 and 2011 that included the measurement of vision.All analyses were conducted on the largest possible number of participants.In contrast to many earlier surveys, our study population did not consist of specific patient groups collected from health care units allowing for better generalization of the results.
Our study also has limitations.Visual field and contrast sensitivity were not measured in the health examination, and as such, we were unable to include them in the determination of impaired vision among survey participants.As binocular vision was studied, the possible effect of a lack of optical correction in one eye or diseases such as cataracts may be overlooked.Furthermore, the number of persons with impaired vision was rather low in the FinHealth 2017 Survey, and the lack of an abridged health examination at home may have caused selection bias.Still, most of the improvements in vision occurred between 2000 and 2011, and, therefore, the bias cannot have a significant effect on the results.The variation in the age between different vision groups was large, but we corrected this by age stratification and age adjustment.All three surveys included predominantly Finnish participants, and therefore, the results may not be directly applicable to other countries.
In conclusion, the overall vision level has improved in Finland during the past two decades despite the ageing of the population and consequent increase in the risk of visionthreatening eye diseases.This positive trend is explained mostly by a shift from lower to higher vision levels among older age groups.As persons live longer with good or adequate vision, they retain their functional capacity, work ability and quality of life longer.This positive development in vision showed remarkable similarity between different regions in Finland, including urban and rural settings, which underlines the importance of equal and accessible eye care throughout the country.To sustain this optimal trend, the level of ophthalmological intervention must keep pace with the continuing ageing of the population.

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; 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.
The current study was conducted in line with the tenets of the Helsinki Declaration.All procedures in the Health 2000, Health 2011 and FinHealth 2017 Surveys involving human participants were performed 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.The Health 2000 Survey was approved by the Ethical Committee for Research in Epidemiology and Public Health at the Hospital District of Helsinki and Uusimaa, and the Health 2011 and FinHealth 2017 Surveys by the Coordinating Ethics Committee at the Hospital District of Helsinki and Uusimaa.All participants received an information letter regarding the study beforehand.Written informed consent was obtained from all participants.The ethical approval process details are discussed in previous publications (Borodulin & Sääksjärvi, 2019; Heistaro, 2008; Lundqvist & Mäki-Opas, 2016).

F
Prevalence of good (a), adequate (b), weak (c) and impaired or worse (d) distance visual acuity (VA) in Finland by age group and study year with 95% confidence intervals.Lines indicate a statistically significant difference (p < 0.05) between two time points.

F
Prevalence of good (a), adequate (b), weak (c) and impaired or worse (d) near visual acuity (VA) in Finland by age group and study year with 95% confidence intervals.Lines indicate a statistically significant difference (p < 0.05) between two time points.

F
Prevalence of good (a) and weak or worse (b) distance visual acuity (VA) and good (c) and weak or worse (d) near VA in Finland by study year based on the type of region of residence with 95% confidence intervals.Lines indicate a statistically significant difference (p < 0.05) between two time points.
Full study protocol, contact details, publications and the process for collaborating and data requests can be found on the websites (thl.fi/health2000 and thl.fi/ finhealth).ORC I DPetri Purola https://orcid.org/0000-0001-5308-8622RE F E R E NC E S Summary of study populations of the Health 2000, Health 2011 and FinHealth 2017 Surveys with vision measured.