Accommodation is unrelated to myopia progression in Chinese myopic children

Authors: Yunyun Chen (1,2); Bjorn Drobe (2,3); Chuanchuan Zhang (1,2); Nisha Singh (2,3); Daniel P. Spiegel (2,3); Hao Chen (1,2); Jinhua Bao (1,2,) and Fan Lu (1,2,)

  1. School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, China
  2. WEIRC, WMU-Essilor International Research Centre, Wenzhou, China
  3. R&D Vision Sciences AMERA, Essilor International, Singapore

Date: April 2020

Reference: Sci Rep 10, 12056 (2020) (LINK)


Although the relationship between accommodation and myopia has been widely investigated, disagreement remains on whether accommodative lag (under-accommodation) plays a part in myopia progression. The proposed mechanism being that the resultant hyperopic defocus stimulates axial length growth towards reduce the lag in focus. A contributing factor towards this conjecture is differences in adopted methodology, which the authors of the set out to address.

By implementing analysis across a continuous range of accommodation between 0 and 6.00D, and including distance accommodation facility, the authors found no association between both measurement conditions leading them to conclude that lag of accommodation has no influence on rate of myopia progression.

This study adds weight to the argument that accommodation lag has no influence on myopia progression, however, even though complex and thorough analysis was conducted, there remain many differences to real world viewing conditions, meaning the jury is still out on whether or not myopia progression is influenced by accommodation lag.

Clinical relevance

  • An interesting observation from this study was that a faster myopia progression rate at mean study cohort age (0.61D/yr) was found than previously reported by Donovan et al (0.48D/yr).1 Leading the authors to suggest that more attention should be given to younger myopes whose myopia progresses rapidly and thus more likely to become high myopes in later years.
  • Results from this study indicate that accommodation lag has no influence on rate of myopia progression, at least in a Chinese population. However, the results should not be considered as definitive due to differences between study measurement protocols and real world conditions.

Limitations and future research

  • Accommodation lag, while measured using an accurate system and carefully analysed, does not necessarily replicate actual accommodation lag under normal conditions:
    • Accommodation lag was measured under monocular conditions, meaning that blur was the only stimulus to accommodation – this does not match real world viewing where binocular fusion also acts as a stimulus to accommodation.
    • Accommodation was measured dynamically – the near fixation target focus demand was being continuously moved during the measurement process. This does not replicate static focus stimulus distance, i.e. when reading a book.
  • Refraction was subjective maximum plus and did not include cycloplegia, which is the current gold standard in a research setting.
  • This study involved Chinese children only due to the geographic setting so outcomes cannot be necessarily extrapolated to be representative of children from other ethnicities or countries.
  • Most myopia progression data were retrospective, meaning that any potential additional influential factors may not necessarily be similar across all participants. The authors argue that applying a Gompertz function to the data helps alleviate against this potential for error.
  • The authors were aware that different times spent concentrating at near wasn’t measured and therefore a conclusion on effect of accommodation-induced hyperopic defocus can’t be deduced.
  • Measurements were conducted in a clinical environment with controlled lighting and deliberately chosen working distances doesn’t necessarily translate to how children might read at home.

Full story


The aim of this study was to investigate the relationship between accommodation and myopic progression in Chinese myopic children.

Study design

This was a longitudinal study where retrospective and prospective data was collected from 113 myopic out-patients at the Primary Care Department of the Eye hospital affiliated with Wenzhou Medical University Eye Hospital, between August 2014 and September 2015.  The children had an age range from 8yrs to 15yrs old with a myopic spherical equivalent refraction (SER) of -0.75D myopia or more, and astigmatism no more than 2.00DC.  They also had no ocular or contact lens wearing history and any anisometropia was less than 1.00D.

Measurement procedure

Retrospective and prospective refractive error measurements were analysed, including at least four non-cycloplegic SER results at 6 month intervals (or more) prior to enrolling at the outpatient system.

The mean rate of myopia progression was -0.61 ± 0.31D per year at mean age 12.2 ± 1.61yrs. A Gompertz function was applied to refraction data to establish rate of progression at enrolment – this function follows an S shape, which when fit to the refraction measurements for each child can be used to establish the steepness of the curve (rate of progression) at the measurement visit when accommodation lag was measured.

Accommodation was measured in the right eye only (left eye occluded) using a Grand Seiko open field autorefractor adapted to continuously measure at 5x per second intervals. During measurements a near target was moved from a stimulus distance of -1.61D to 14.83D at a rate of 0.40D per second. A third order polynomial was then fit to individual participant measurements to allow calculation of lag at different accommodation stimuli and the area under the curve relative to the 1:1 perfect focus line between 0 and 6D of accommodation stimulus. Monocular distance accommodative facility (DAF) was measured in the right eye (left eye occluded) as number of cycles for clearing a 4.5m fixation target with and without a -2.00 lens.


  • In agreement with other studies, rate of myopia progression decreased with age and was faster in children with higher myopia at study enrolment:
    • Age 8: -1.10 ± 0.25D/yr
    • Age 15: -0.25 ± 0.08D/yr
  • Accommodation lag and mean value of accommodative lag area between 0 and 6D was not associated with rate of myopia progression.
  • Mean DAF was 18.29 ± 7.42 cycles per minute and was not associated with rate of myopia progression.

An interesting observation that the authors made from using a Gompertz function fit to longitudinal refraction data is that the function indicated a faster progression rate at mean study cohort age (0.61D/yr) than previously reported by Donovan et al who used a quadratic equation fit (0.48D/yr).1 Leading them to suggest that more attention should be given to younger myopes whose myopia progresses rapidly and thus more likely to become high myopes in later years.


A potential explanation for lack of association between lag of accommodation and myopia progression is that the mechanisms of myopia onset and myopia progression are different:

  • Myopia progression is not regulated by accommodation lag (current study)
  • Myopia develops rapidly (within 2 days) of negative lens wear in guinea pigs and more slowly thereafter2

Is accommodation lag a consequence of myopia? This could explain why mulitifocal lenses designed to reduce accommodation lag to slow myopia progression only have modest effect.

Lack of association between myopia progression and DAF is in agreement with other studies that have shown no effect of accommodation training towards slowing myopia progression.

The authors conclude by recognizing that myopization is a complex, multifactorial process, with their results suggesting that the myopization process involves factors other than accommodation.


Title: Accommodation is unrelated to myopia progression in Chinese myopic children

This study shows accommodative accuracy and distance accommodation facility in myopic children do not play a role in myopia progression. In 144 subjects, the monocular distance accommodative facility (DAF) and continuous accommodative stimulus–response curves (ASRCs) were measured at the enrolment. Retrospective and prospective refraction with regard to the enrolment visit were obtained from the outpatient database system based on noncycloplegic subjective spherical equivalent refraction (SER). The rate of myopic progression at enrolment was the first derivative of the Gompertz function, which was fitted with each subject's longitudinal refractive error data, including at least four records of SER with an interval of more than 6 months between each visit. A mixed linear model for multilevel repeated-measures data was used to explore the associations between the rate of myopia progression and accommodative parameters. The mean rate of myopia progression at enrolment was -0.61 ± 0.31 D/y with a mean age of 12.27 ± 1.61 years. By adjusting for age and SER, it was shown that the myopic progression rate was not associated with the accommodative lag (F = 0.269, P = 0.604), accommodative lag area (F = 0.086, P = 0.354), slope of ASRC (F = 0.711, P = 0.399), and DAF (F = 0.619, P = 0.432).

Abstract link is here


About Ailsa

Ailsa Lane is a contact lens optician based in Kent, England. She is currently completing her Advanced Diploma In Contact Lens Practice with Honours, which has ignited her interest and skills in understanding scientific research and finding its translations to clinical practice.


About Paul

Dr Paul Gifford is a research scientist and industry innovator based in Brisbane, Australia, and co-founder of Myopia Profile.


  1. Donovan L, Sankaridurg P, Ho A, Naduvilah T, Smith III EL, Holden BA. Myopia progression rates in urban children wearing single-vision spectacles. Optom Vis Sci. 2012;89:27-32.
  2. Lu F, Zhou X, Jiang L, Fu Y, Lai X, Xie R, Qu J. Axial myopia induced by hyperopic defocus in guinea pigs: A detailed assessment on susceptibility and recovery. Exp. Eye Res. 2009;89:101–8.

Leave a comment