Article Text

Download PDFPDF

How does spending time outdoors protect against myopia? A review
  1. Gareth Lingham1,
  2. David A Mackey1,
  3. Robyn Lucas1,2,
  4. Seyhan Yazar1,3
  1. 1 Centre for Ophthalmology and Visual Science, Lions Eye Institute, University of Western Australia, Perth, Western Australia, Australia
  2. 2 National Centre for Epidemiology and Population Health, Research School of Population Health, Australian National University, Canberra, Australian Capital Territory, Australia
  3. 3 Single Cell and Computational Genomics, Garvan Institute of Medical Research, Sydney, New South Wales, Australia
  1. Correspondence to Seyhan Yazar, Centre for Ophthalmology and Visual Science, Lions Eye Institute, University of Western Australia, Perth, WA 6009, Australia; seyhanyazar{at}lei.org.au

Abstract

Myopia is an increasingly common condition that is associated with significant costs to individuals and society. Moreover, myopia is associated with increased risk of glaucoma, retinal detachment and myopic maculopathy, which in turn can lead to blindness. It is now well established that spending more time outdoors during childhood lowers the risk of developing myopia and may delay progression of myopia. There has been great interest in further exploring this relationship and exploiting it as a public health intervention aimed at preventing myopia in children. However, spending more time outdoors can have detrimental effects, such as increased risk of melanoma, cataract and pterygium. Understanding how spending more time outdoors prevents myopia could advance development of more targeted interventions for myopia. We reviewed the evidence for and against eight facets of spending time outdoors that may protect against myopia: brighter light, reduced peripheral defocus, higher vitamin D levels, differing chromatic spectrum of light, higher physical activity, entrained circadian rhythms, less near work and greater high spatial frequency (SF) energies. There is solid evidence that exposure to brighter light can reduce risk of myopia. Peripheral defocus is able to regulate eye growth but whether spending time outdoors substantially changes peripheral defocus patterns and how this could affect myopia risk is unclear. Spectrum of light, circadian rhythms and SF characteristics are plausible factors, but there is a lack of solid evidence from human studies. Vitamin D, physical activity and near work appear unlikely to mediate the relationship between time spent outdoors and myopia.

  • optics and refraction
  • public health
  • epidemiology
  • experimental—animal models

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Myopia is a common eye condition that is becoming increasingly prevalent globally.1 Not only does the rising prevalence have both direct and indirect associated costs,2 but also current data suggest it will add to the burden of blindness and visual impairment due to myopia-associated conditions such as myopic maculopathy, glaucoma and retinal detachment,3 the risk of which increases with increasing severity of myopia.3 4 Thus, identifying modifiable risk factors for myopia onset and progression is a research priority.

In a seminal 2008 publication, spending more time outdoors, regardless of whether engaged in physical activity or not, was associated with lower prevalence of myopia in children.5 Longitudinal studies and randomised controlled trials have since confirmed that spending more time outdoors reduces the risk of myopia onset and possibly progression in children and adolescents.6–8 Spending more time outdoors may reduce the risk of myopia by a number of pathways (figure 1). Identifying the factors behind this relationship might allow interventions to be targeted, thereby avoiding potentially detrimental effects associated with spending time outdoors, such as sunburn and an increased risk of skin cancers.9 The means by which spending time outdoors might lower myopia risk have been briefly discussed in broader reviews, as well as comprehensively in reviews on individual elements (online supplementary table 1). However, the rapid expansion of studies warrants an updated and focused review comparing the evidence for different causal factors linking time spent outdoors and myopia. We conducted a comprehensive narrative review of eight leading hypotheses on how time spent outdoors prevents myopia. The online supplementary material contains a description of the literature search.

Figure 1

Spending more time outdoors may reduce risk of myopia via multiple means.

Background

Onset of myopia is typically during childhood and adolescence. After onset, myopia progresses most quickly at younger ages, before stabilising in early adulthood.10 Axial eye growth (referred to as ‘eye growth’) is the predominant driver of the development and progression of childhood/adolescent myopia and is therefore a key biomarker.10

Insights into the causes of myopia and myopia progression have largely been derived from animal models, most commonly chick, tree shrew, guinea pig or rhesus monkey. There are differences between eyes in each of these species and humans, such as the number of cone photoreceptor types and mechanism of accommodation. The results of primate studies are most applicable to humans.11

There are two common models of induced myopia in animals (figure 2): form-deprivation myopia (FDM) and lens-induced myopia (LIM).11 Both FDM and LIM can be largely negated by 2–4 hours of normal visual input per day.12 13 It is unclear which model is most applicable to humans; human myopia is progressive, similar to FDM, but human eyes are generally not form-deprived in the normal environment. Alternatively, hypermetropic defocus could occur in children but would need to be almost constant to induce myopia, which also seems unlikely.

Figure 2

Illustrations of (top) LIM using a concave lens and (bottom) FDM using a translucent filter. In LIM, a concave lens induces hypermetropic defocus. The retina detects the sign of defocus and adjusts axial length to compensate for the amount of defocus (closed loop). In FDM, a translucent or opaque filter deprives the retina of a clear image, causing progressive, severe axial eye growth with no defined endpoint (open-loop). FDM, form-deprivation myopia; LIM, lens-induced myopia.

The biochemical pathways that lead to myopia-inducing eye growth are not fully understood. The current understanding is of a signalling cascade, initiated within the retina in response to visual input that leads to changes in the retinal pigment epithelium and choroid and ultimately results in scleral remodelling and eye growth.14 Biochemicals implicated in this signalling include dopamine, acetylcholine, insulin, melanopsin, retinoic acid, nitric oxide and gamma-aminobutyric acid.14

Possible causal pathways from time outdoors to myopia

Higher illuminance

The light illuminance (brightness) outdoors is typically 10–1000 times greater than that of indoor lighting7 15 and may link time spent outdoors and myopia.5 Chicks raised under low illuminance (50 lux) conditions develop myopia16 and brighter ambient lighting (10 000–40 000 lux) prevents FDM in chicks and rhesus monkeys in a dose-dependent manner.17–19 Inhibition of dopamine-2 receptors completely blocks the protective effect of brighter light on FDM in chicks, demonstrating the key role of dopamine in FDM.18 Illuminance appears to play a lesser role in LIM. Brighter lighting did not reduce the amount of LIM in chicks and rhesus monkeys but, in chicks, did slow the rate of development of LIM.20 21 Dopamine receptors also appear to play a smaller role in LIM, suggesting regulation by other pathways.22

In humans, light exposure can be directly measured using personal light meters worn for a 1–2 week period. Using light metres, longitudinal,7 23 but not cross-sectional,15 24 studies found that higher average daily light exposure was associated with slower eye growth, with 40 min a day of bright light (>3000 lux) exposure being associated with a significantly slower rate of eye growth.23 Artificially increasing light exposure may also lower myopia risk. In 22 young adults, 30 min of light therapy glasses (506 lux) used for 7 days resulted in a slightly thicker choroid,25 a measure associated with slower eye growth in children.26 In a 1-year intervention study (n=317), increasing classroom illuminance from approximately 100 to 500 lux reduced the risk of myopia onset by 6%; however, this study was open to selection bias.27 It is noteworthy that if brighter light is able to prevent myopia in humans, this may indicate that human myopia is FDM-like.

In summary, there is direct evidence of a relationship between brighter light exposure and lower myopia risk in children and adolescents. In animals, exposure to brighter light prevents FDM onset and progression; however, it may play a lesser role in LIM. Increasing ocular light exposure is a promising intervention for myopia. Further interventional studies into whether higher light illuminance independent of spending time outdoors can prevent myopia in humans are required.

Reduced peripheral defocus

Animal studies have demonstrated that hypermetropic defocus, as in LIM, causes eye growth. In rhesus monkeys and guinea pigs, peripheral hypermetropic defocus without central defocus induces both central and peripheral myopia, demonstrating that the peripheral retina can regulate eye growth.28 29 Hypermetropic defocus of the peripheral retina is therefore a potential driver of myopia and can be induced by peripheral objects being nearer than the object of central focus (environmental) or by a prolate eye shape.

The myopic eye is more prolate and has more relative peripheral hypermetropia (RPH) than emmetropic eyes.30 31 However, these characteristics are likely secondary to myopia onset as large longitudinal studies found no link between baseline RPH (up to 30° eccentricity) and risk of myopia onset or progression.32 33 Nevertheless, contact lenses designed to induce relative peripheral myopia (RPM) and orthokeratology lenses, which also induce RPM, seem to slow progression of myopia in children.34–36 Therefore, eye shape-induced peripheral defocus may be involved in myopia progression but not myopia onset.

However, it is not known whether environmentally-induced peripheral defocus is involved in myopia onset or progression. In the outdoor environment, objects are typically far from the eye and the dioptric environment is generally uniform (figure 3). Thus, peripheral objects are minimally defocused when outdoors. In contrast, when indoors, objects are typically nearer, and variation in the dioptric distance of objects is greater. In this dioptrically varied environment, there is potential for objects to produce greater RPH or RPM.3 It has been suggested that spending less time outdoors and more time indoors exposes children to more peripheral defocus, potentially leading to eye growth. Conversely, the dioptric uniformity of the outdoor environment and consequent minimisation of peripheral defocus may somehow protect against myopia.3 However, this hypothesis is complicated by RPM, which is also present in the indoor environment3 and is a strong inhibitor of eye growth.29 37 Additionally, nearly constant hypermetropic defocus is required to induce myopia in animals, which is an unlikely environmental situation in humans.12 13

Figure 3

(A) Dioptric distance is the reciprocal of distance and increases rapidly at distances less than 2 m. (B,C) Examples of indoor and outdoor environments, respectively. The green asterisk represents the point of fixation. The blue arrows and numbers correspond to the dioptric distance of the object from the observer. Objects nearer than the point of fixation will be hypermetropically defocused.

In summary, animal studies show that RPH can induce myopia. Human studies show that peripheral defocus is probably involved in myopia progression but potentially not onset. There is scarce evidence assessing the role of environmentally induced peripheral defocus in myopia. Whether the indoor environment could produce enough constant RPH to stimulate eye growth and whether peripheral defocus is actually substantially different in the outdoor environment are unknown. Wearable devices that measure the distances of objects will enable a better understanding of defocus patterns in the visual environment and may shed light on the role of environmentally induced peripheral defocus in human myopia.38

Vitamin D

Vitamin D is produced when the skin is exposed to ultraviolet radiation and could link time spent outdoors and myopia, although the possible molecular pathways are unclear.39 Alternatively, serum 25-hydroxyvitamin D (25(OH)D) levels (the usual marker of vitamin D adequacy) may act only as a biomarker of recent time spent outdoors (weeks–months) without being causally related to myopia. Table 1 summarises epidemiological studies assessing the association between 25(OH)D levels and myopia. Despite substantial heterogeneity between studies, a recent systematic review and meta-analysis found that, overall, those with myopia had lower 25(OH)D than those without myopia.40 The only longitudinal study found no association between baseline 25(OH)D levels and onset of ‘likely myopia’ in children after adjusting for previously reported time spent outdoors.41 Additionally, a Mendelian randomisation (MR) study based on four single-nucleotide polymorphisms (SNPs) found that a genetic predisposition to lower 25(OH)D concentration was not associated with increased risk of myopia.42

Table 1

Summary of epidemiological studies investigating the association between 25(OH)D concentrations and myopia

The evidence linking vitamin D and myopia is inconsistent. This may arise from residual confounding due to the inability to accurately measure time spent outdoors, or from reverse causation in cross-sectional studies. The MR study is able to largely ignore time spent outdoors as a covariate; however, this study may not have detected a threshold effect due to the small effect of SNPs on 25(OH)D concentration. Nevertheless, the longitudinal and MR studies provide the strongest evidence, and it seems unlikely that vitamin D is causally related to myopia.

Chromatic spectrum of light

Shorter wavelengths of light (eg, blue) are refracted more than long wavelengths (eg, red); thus, red light will be focused posterior to blue light in the eye. Eyes of chicks use such chromatic cues from light to regulate eye growth.43 44 Compared with tungsten and fluorescent lights, sunlight contains relatively more green, blue and ultraviolet lights.45 The different chromatic spectrum of artificial light could cause the retina to incorrectly interpret defocus signals and induce eye growth, thereby linking time spent outdoors and myopia.

Rearing animals under narrow-band wavelengths of light produces changes in eye growth that differ among species and wavelengths of light (table 2). Likewise, monochromatic lighting has mixed effects on development of FDM or LIM. Red lighting prevented FDM or LIM in rhesus monkeys46 and LIM in guinea pigs47 but had no effect on FDM in chicks,48 whereas blue light slightly retarded FDM but not LIM in chicks.48 49 Although conducted in artificial conditions, these studies demonstrate that differing spectra of light can potentially induce refractive changes. In a more realistic study, rearing guinea pigs under a halogen lamp with emission spectrum close to sunlight or an indoor fluorescent lamp produced equal amounts of LIM.50

Table 2

Effect of raising various animals under monochromatic light on refraction, relative to controls raised under white light

Violet light (360–400 nm) may be important for myopia prevention. In chicks, the addition of a fluorescent violet light impeded FDM and LIM compared with controls, and in a retrospective review of medical records, eye growth in children who wore violet light-transmitting contact lenses (n=116) was slower by 0.05 mm/year compared with children who wore contact lenses that partially block violet light (n=31).51 However, it has been argued that the crystalline lens does not transmit enough violet light to have a relevant effect at the retina.48

The amount of medium-wavelength (green) or long-wavelength (red) absorbing opsin in the retina may also impact risk of myopia; a large case–control study found a lower prevalence of myopia among red/green colour vision-deficient individuals.52 A study in chicks found no association between the medium to long wavelength cone (M:L) ratio and severity of FDM, but a correlation between higher M:L ratio and more hypermetropic refractive error in control eyes.53

Differences in the chromatic spectrum of light can potentially alter eye growth, but the empirical evidence directly linking chromatic light and myopia onset is still relatively weak. Replication and new studies addressing whether the differences in chromatic spectrum of sunlight and artificial light are able to induce myopia are needed.

Physical activity

Early studies found that more physical activity protected against myopia perhaps via biochemical changes induced by exercise.54 55 However, these studies did not adjust for time spent outdoors or used a combined measure of physical and outdoor activity. In a cross-sectional study of children, more time spent in outdoor sport and activity was associated with lower myopia risk, but there was no similar effect for indoor sports and activity. This suggests that spending time outdoors, rather than physical activity, was protective.5 More recent longitudinal studies designed to adjust for time spent outdoors as a covariate have confirmed that physical activity itself is unrelated to myopia risk and hence does not link time spent outdoors and myopia.56 57

Circadian rhythms

There is significant overlap in signalling between regulation of circadian rhythms and eye growth.58 For example, dopamine is a key regulator of both circadian rhythms and eye growth.58 The eye’s axial length and choroidal thickness also undergo small (~10 to 20 µm) diurnal variations in approximate antiphase with one another.59 60 Induced hypermetropic defocus, as in LIM, disrupts these diurnal rhythms.60 61 Furthermore, in chicks, expression of genes regulating circadian rhythms was mildly reduced in form-deprived eyes compared with contralateral eyes and, in mice, retinal-specific conditional knockout of a key circadian rhythm gene, Bmal1, induced myopic eye growth.62 63 Thus, circadian rhythms may be involved in regulating refractive development; light exposure is important for entraining circadian rhythms, suggesting a possible pathway from time spent outdoors to myopia.

A study in young adults (n=42) found no difference in axial length or choroidal thickness diurnal rhythms, time spent outdoors, light exposure or sleep duration in those with and without myopia.64 These null findings may reflect the age of participants (19–30 years) in whom risk of myopia is generally lower.

Melatonin is strongly linked to circadian rhythms, being high during sleep and low during the day.64 In chicks, doses of melatonin of ≤1 mg injected intravitreally slightly enhanced FDM severity, while very high doses (2 mg) mildly retarded FDM.65 66 In humans, two cross-sectional studies (n=55, ages 21–64 years, and n=24, ages 17–40 years) found no direct relationship between melatonin concentration amplitude and phase and myopia.64 67 Conversely, in another study (n=54, ages 18–20 years), those with myopia had higher morning melatonin levels compared with those without myopia.68 These contrary findings may be due to the different ages of the participants or the small sample sizes. In two large cross-sectional studies of Chinese children and adolescents, myopia was associated with shorter self-reported sleep duration.69 70

On balance, it seems likely that eye growth can be regulated by circadian rhythms, possibly through dopaminergic pathways. However, evidence showing disrupted circadian rhythms in myopic humans is generally scarce and of limited quality. Melatonin itself is probably not directly related to myopia, but it may be a useful marker of circadian rhythm regulation.

Near work

Excessive near work has long been theorised to cause myopia. In a meta-analysis, near work was identified as a risk factor for myopia, despite inconsistent findings between the included studies and considerable differences in method of quantifying near work (eg, dioptre-hours, books read per week).71 Therefore, the relationship between time spent outdoors and myopia may simply arise from individuals performing less near work when outside. Several studies provide evidence that, in reality, time spent on near work and time spent outdoors are largely unrelated. The correlation between time spent outdoors and time on near work in Australian children aged 6–12 year was weak (r=0.20 and r=0.03, respectively).5 Other cross-sectional and longitudinal studies have found independent effects of time spent outdoors and near work on myopia risk, although this could arise from limitations in accurately measuring these variables.8 72

The possible causal pathways from near work to myopia are also unclear. Abolition of accommodation in chicks does not prevent LIM, suggesting that accommodation itself does not cause myopia.73 Accommodative lag, which occurs when less accommodation is exerted than theoretically required for near targets, imposes hypermetropic defocus on the retina and is greater in myopic individuals than emmetropic individuals.74 However, in studies of children32 or marmoset monkeys,75 accommodative lag was not associated with risk of myopia onset or progression32 75 and only increased after the onset of myopia, suggesting that myopia causes increased accommodative lag rather than the reverse. Overall, it seems unlikely that time spent on near work mediates the relationship between time spent outdoors and myopia.

Spatial frequency (SF) characteristics

It has recently been suggested that the SF characteristics of outdoor scenes may be beneficial in preventing myopia.76 In 191 human cone sensitivity-matched images of varied indoor and outdoor scenes, outdoor scenes, when compared with indoor scenes, contained greater amounts of mid or high SF, relative to low SF, details.76 Occluders or lenses used to induce FDM or LIM, respectively, preferentially reduce contrast at higher SFs.77 78 Moreover, environments with greater energy at mid79 and/or high77 80 SF are able to prevent myopic shifts in chicks. Temporal flicker, which is translated from movement of SF stimuli, also appears to modulate eye growth, with low, relative to high, frequency flicker environments inducing eye growth.81 This area deserves further investigation, particularly in human studies.

Conclusion

Spending time outdoors may prevent myopia via multiple means. Higher illuminance is the most well-established theory with support from both animal and human studies. Increasing light exposure is a promising intervention for myopia, but interventional studies that increase light exposure without altering time spent outdoors (eg, with artificial light) are needed before being adopted in a clinical setting. Peripheral defocus appears to be able to modulate eye growth; however, whether differences in peripheral defocus in the outdoor environment compared with the indoor environment mediate the relationship between time spent outdoors and myopia unknown. Wearable devices that measure distances of objects from the observer may help elucidate this relationship by characterising peripheral defocus in different environments. Circadian rhythms, chromatic spectrum of light and SF characteristics could link time spent outdoors and myopia, but there is limited human evidence to support this; verifications will be required before intervention studies are worthwhile. It seems unlikely that physical activity, vitamin D or near work are involved. Spending time outdoors may also prevent myopia via multiple means and some combination of these elements may be required for maximal efficacy. For example, in chicks, brighter light and induced RPM had an additive effect on reducing eye growth.82 Reassuringly, ultraviolet light exposure is likely not required for myopia protection; therefore, clinical recommendations or public health interventions to increase time spent outdoors are likely compatible with skin protection and sunglasses use, which do not reduce light intensity significantly.83 However, it should be noted that despite extensive sun protection campaigns in Australia, skin protection in adolescents remains suboptimal.84 Understanding how spending time outdoors prevents myopia will enable the design of targeted myopia interventions to combat the rising prevalence of myopia.

Acknowledgments

We thank Professor Ian Flitcroft for his advice regarding spatial frequency characteristics of indoor and outdoor scenes.

References

Footnotes

  • Contributors All authors contributed to the conception, design, drafting and revision of this manuscript.

  • Funding GL receives financial support through an Australian Government Research Training Program Scholarship. RL is supported by a National Health and Medical Research Senior Research Fellowship, SY by a CJ Martin Biomedical Fellowship and DAM by practitioner fellowship. This work is supported by a project grant from the National Health and Medical Research Council (1121979).

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.