The impact of smartphone use on accommodative functions: pilot study

ABSTRACT

Evidence about the effect of smartphone use on accommodation is limited and inconclusive. Several studies have investigated either symptoms or near triad measures following smartphone use. These suggest that, certainly for the short term, smartphones have a negative impact on the near triad and produce symptoms. In addition, there is a body of recent work reporting cases of acute acquired concomitant esotropia (AACE) that may be caused by the accommodation-vergence demand of excessive smartphone use. A pilot study was undertaken to investigate accommodative measures before and after 30 minutes of smartphone use. Participants aged 16-40 years were invited to participate. Accommodative facility (AF), near point of accommodation (NPA), and near point of convergence (NPC) before and after 30 minutes of habitual smartphone use were assessed. NPA and AF were assessed with both eyes open (BEO), right eye (RE) and left eye (LE). Accommodative facility was assessed using ±2DS flipper lenses and measured in cycles per minute (cpm). NPA and NPC were assessed using the RAF rule and measured in centimeters. Data were analyzed in StatsDirect using non-parametric statistical tests. Eighteen participants were recruited, with a mean age of 24 years (SD: 7.6yrs). AF improved by 3cpm (p= .015) for BEO, 2.25cpm for RE (p = .004) and 1.5 cpm for the LE (p =. 278) after smartphone use. NPA with BEO became worse by 2 cm (p =.0474), with the RE worse by 0.5cms (p = .0474) and the LE, worse by 0.125 cms (p = .047). Convergence worsened by 0.75 cms (p = .018). Although these appeared to represent a change in measures following smartphone use, post-hoc analysis with Bonferroni correction demonstrated that these were not statistically significant at the 0.07 significance level. This pilot study found that there was no difference in accommodative and convergence measures before and after 30 minutes of smartphone use. These results suggest evidence contrary to the existing literature. There are several limitations to this pilot study and previous work, which are discussed. Suggestions on future work to explore the effect of smartphone use on the near triad to address limitations and further knowledge, in this area, are provided.

KEYWORDS:

• Accommodation
• smartphone
• convergence

Introduction

Daily internet use has increased from 35% to 89% from 2006 to 2020,¹ and 78% of the adults in Great Britain access the internet via their smartphone,² and there is a need to understand the effects of this relatively new technology on the eyes. When reading from a book, computer, or smartphone, the eyes must undergo a triad of ocular reflexes in order to maintain comfortable vision. The eyes must converge, which means that they point in toward the object of regard, the pupils must constrict, and accommodation occurs. These actions contribute to the clarity of vision for near tasks.

Smartphone use and Acute Acquired Concomitant Esotropia (AACE)

Literature reporting the effect of smartphone use on the eyes falls into three main categories, which are accommodation and vergence issues, eye surface, and blue light.^3–5 Only a small number of studies have undertaken a prospective investigation into the effect of smartphone use on measurable functions.^6–12 Of note and relevance is the growing body of the literature describing cases of AACE due to excessive smartphone use,^13–20 which may result from changes in accommodative and vergence measures.^10–12 Lee et al.¹³ is cited as the initial observer of this scenario where 16 cases of AACE are reported to have improved by 10 prism diopters following reduced smartphone use. Subsequently, perhaps due to the COVID-19 pandemic, further descriptions of over 100 cases are given in the literature of AACE caused by excessive smartphone use.^14–20 Excessive smartphone use could be considered more than 4 hours use per day,²⁰ but it was not always stated. The authors are finding similar patient characteristics that demonstrate an increased risk of AACE due to excessive smartphone use. These characteristics include: existing esophoria,¹⁶ shorter working distance,¹⁹ teenage years,¹⁷ and myopia.^15,17 The authors of the above studies postulate that abnormal accommodation and vergence responses result in some changes in the medial rectus muscle activation or anatomy.^17,18 Interestingly, some authors believe it is the spasm of accommodation that causes the esotropia,^14,15 however Van Hoolst et al.¹⁹ demonstrated no spasm of accommodation within their AACE case series following refraction and AC/A ratio measurements. Therefore, despite some postulation over the cause of AACE due to smartphone use, there remains further work to identify the effects of smartphone use on the near triad and the preexisting risk factors that may be exacerbated. This pilot study explores the relationship between these functions and smartphone use.

Effect of sustained near work on accommodative functions

Although accommodative lag has been found to increase following a period of reading, and similarly following computer (visual display unit) use,^21–23 this has only recently been explored in relation to smartphone use. No difference was found in accommodative response when measured using an infrared optometer at minute intervals when reading from an iPod, Kindle, or hard copy text.²⁴ Also, accommodative results were the same when smartphone, tablet, and hard copy text stimuli were compared.⁸ This may suggest that sustained effects on accommodation following smartphone use are similar to those found when undertaking any form of near work. However, these studies do not account for the day-to-day (habitual) use of smartphones, which can include; playing games, watching videos, messaging, and other activities. Smartphones are also known to be used at shorter working distances,²⁵ which over a longer period of use may affect near triad functions.

Effect of smartphone use on accommodative functions

Several studies have reported on the effects specifically of smartphone use either upon symptoms, near triad measures, or both.^6–12 Most studies used a 10-minutes to 1-hour window of smartphone use and measured outcomes pre- and post-smartphone use;^8–12 however, one study used a self-reported amount of smartphone use and its effect on symptoms.⁶ Three of these studies measured only symptoms,^6,7,9 three studies measured only functions of the near triad,^8,10,11 and one studied both symptoms and near triad functions.¹² These studies reported an increase in symptoms following smartphone use and a negative effect on near triad functions, such as receding near point of accommodation (NPA) and near point convergence (NPC) ranging from 0.2 to 1.9 cm difference, increased accommodative lag of 0.8DS, and reduced accommodation facility (AF) of 2.29 cycles per minute following smartphone use.^10–12 This suggests that smartphone use could fatigue the accommodative and vergence system and with that produce symptoms.

Although the existing literature does suggest some effect of smartphone use on near triad functions, there are limitations to this. Of all the studies looking into symptoms, two used validated questionnaires but these were not validated for use with smartphone users, as none exists. Kang et al.¹² used the Ocular Discomfort Analog Scale and Ram et al.⁹ used the Convergence Insufficiency Symptom Survey. The study by Ram et al.⁹ also targeted the smartphone use to bedtime and it was not clear how such methods were controlled to ensure standardized length of time on the smartphone and approach to symptom survey. Apart from one study examining the effect following self-reported smartphone use,⁶ others have measured either symptoms or near triad measures pre- and post-smartphone use.^8,10–12 There is some evidence to suggest a negative effect on near triad with reduction in the NPC and NPA and increase in accommodative lag.

Those studies that undertook multiple tests did not appear to use a post-hoc correction in their analysis; therefore, the results found may be open to type 1 error.^10–12 One study that did multiple analyses testing differences between vergence stimuli to different devices did apply a post hoc Tukey’s honest test, and this study was the only one not finding the effect of smartphone use on vergence response.⁸ A review of the literature suggested that it may be applicable to use a post-hoc correction following analysis of such results as accommodative measures are related to each other.²⁶

Rationale

Some literature suggests that smartphones, at least during short-term use, result in symptoms possibly caused by a fatigued near response. However, these results may be limited by the methods and analysis used in the studies. In addition to this, reading on most platforms (paper, VDU, tablet, and smartphone) showed a similar response in accommodative function. This means that, experimenting with the habitual use of smartphones (that differ from reading) may build on the existing research. The aim of this pilot was to find out if there is a difference in accommodative and vergence measures following habitual smartphone use. With a view to continue into a larger scale study, collecting data about smartphone usage habits and applying the learning from this pilot.

Materials and methods

Ethics: Ethical approval was granted by the University of Liverpool Research ethics committee (ref no. 5286)

Participants were recruited from the University of Liverpool Staff and Student population. Participants were included if they had “good” visual acuity (better than 0.3 logMAR) in either eye and no known ophthalmological conditions. Uniocular visual acuity was measured using the EDTRS chart to confirm good vision. The participants were informed that they would be required to undertake a series of eye measurements, then use their smartphone in their “usual way” for 30 minutes (the activity and working distance was not monitored though the participant remained in the same room as the assessor) and again they would undertake the same series of measurements (Table 1). All assessments were undertaken in the same room under the same lighting conditions though the room illumination was not measured in lux.

Table 1. Showing the order of procedures. NPA = Near point of accommodation, NPC = Near point convergence, AF= Accommodative facility, BEO= both eyes open, RE= right eye, LE= left eye.

Timeline	Assessments	Conditions
Pre-smartphone	NPA: BEO, RE, LE x 3	Lighting controlled N5 print used Wearing habitual glasses
	NPC: BEO x 3
	AF: BEO, RE, LE x 3
30 mins habitual use	Participant located within the testing lab	Lighting controlled in lab
Post-smartphone	NPA: BEO, RE, LE x 3	Lighting controlled N5 print used Wearing habitual glasses
	NPC: BEO x 3
	AF: BEO, RE, LE x 3

Near point accommodation

This was measured three times with the right eye, the left eye, and with both eyes open. The RAF rule was placed on the bridge of the participants’ nose and they were asked to fixate on one word from the N5 print line. The participants were asked to inform the examiner when the letters became blurred, still readable but no longer clear. At this point, the carrier was pulled back slowly until the letters became clear again, and the distance from the RAF rule was recorded.

Near point convergence

The RAF rule was placed on the bridge of the participant’s nose, they were asked to fixate on the dot stimulus on the carrier. The participants were asked to inform the examiner when the target became double. It was then pulled back until it was single and the distance from the RAF rule recorded. This test was undertaken BEO only.

Accommodative facility

A brief view through each lens was provided to the participant. The reduced ETDRS chart was used at 40cms distance from the patient using the attached string.+2.00DS/-2.00DS flipper lenses were used to measure the facility. The timer was set to 1 minute, and when the participant was ready, they were asked to first clear the −2.00DS lenses. Without delay, the lenses were flipped to the+2.00 (twisted), each time the participants reported it was clear the lenses were flipped. A note of the number of flips was recorded, and a full cycle was considered clearance of both −2.00DS and+2.00DS.

Smartphone activity

Participants were instructed to use their smartphones in their habitual way for 30 minutes within the testing room. They were then asked to perform the ocular measures again as previously done.

Results

Eighteen participants were recruited (mean 24 yrs SD 7.7 yrs).

The NPA for all conditions (BEO, RE, and LE) and NPC appeared to recede following smartphone use and AF appeared to increase (Table 2). The difference in the median of the three measures was calculated, and Statsdirect was used to ascertain whether there was a statistically significant difference. A Wilcoxon signed rank test with two-sided p-value was performed for each difference pre- and post-smartphone use.

Table 2. Showing measures of accommodation and vergence pre-post smartphone use. Note: Median of three measures reported.

Measure	Pre Smartphone Median (range)	Post Smartphone Median (range)	P value
AF BEO (cpm)	8 (4–12.5)	11 (0–13)	0.015
AF RE (cpm)	8 (2–11.5)	10.25 (0–15.5)	0.037
AF LE (cpm)	9.5 (1–12.5)	11 (0–17)	0.278
NPA BEO (cms)	12.25 (8.5–23)	14.25 (8–24)	0.047
NPA RE (cms)	13.25 (8.25–22)	13.75 (8.13–21.5)	0.047
NPA LE (cms)	13.5 (8.12–21.75)	13.63 (8.19–21.6)	0.047
Convergence (cms)	9 (6–15.5)	9.75 (6–23.5)	0.018

It was found that the NPA receded by 2 cm with BEO (p = .047), by 0.5 cm with the RE (p = .047) and 0.13 cm with the LE (p = .047). NPC receded by 0.75 cm (p = .018). The AF increased by 3 cpm with BEO (p = .015), increased by 2.25 cpm with the RE (p = .037), and increased by 1.5 cpm with the LE (p = .278). P values are reported prior to Bonferroni correction in Table 2, differences and overall significance levels with Bonferroni correction are reported in Table 3. The differences were not found to be statistically significant following post-hoc Bonferroni analysis due to a newly accepted P-value of <0.07 for each test (Table 3).

Table 3. Overview of the significance of the differences found, including post-hoc correction and clinical significance.

Function	Difference	Statistically Significant	Bonferroni Correction	Clinically Significant
AF (BEO)	+3cpm	Y	N	N
AF (RE)	+2.25cpm	Y	N	N
AF (LE)	+1.5cpm	N	N	N
NPA(BEO)	−2cms	Y	N	N
NPA (RE)	−0.5 cm	Y	N	N
NPA (LE)	−0.13cms	Y	N	N
NPC	−0.75cms	Y	N	N

Discussion

This pilot study found no statistical difference in near triad measures following 30 minutes of smartphone use. The trends found in this study agree with previous work to a certain extent except for accommodative facility which increased following smartphone use. This difference may be due to the methods used to test AF where a practice session may be advisable to avoid practice effects being interpreted as change. A practice effect may show a simulated improvement in AF measurements. Despite these trends, these pilot data cannot be accepted as showing any difference following statistical analysis.

This is contrary to much of the existing literature supporting that smartphones negatively affect the accommodative vergence system,^6–12 producing symptoms of tired, uncomfortable eyes, blurred vision, redness, watering, and AACE^6,7,13–20 Reports suggest that accommodative lag increases, accommodative facility decreases, and near point of accommodation and convergence recede.^11,12 It is certain that excessive smartphone use can result in AACE^13–20though an RCT would be required to ascertain the parameters that lead to this. The mechanism of such a response to smartphone use alongside risk factors is changes in accommodation that result in either spasm as suggested by Kaur et al.¹⁴ and Hayashi et al.¹⁵ or lag as measured by several others.^10–12 As there are two possible mechanisms (spasm vs lag) postulated, it may be the underlying patient characteristics that determine whether accommodative lag or accommodative spasm may result from smartphone use and potential to drive an acute onset concomitant esotropia. Both this pilot and existing works on pre- and post-smartphone measures are limited by their methods, sample sizes, and statistical analysis.

While some studies had larger sample sizes than this pilot,^6,9 many did not.^7,8,10 All four studies reporting on symptoms caused by smartphones used either a non-validated tool or one not originally designed specifically for smartphone-related symptoms. Many of the studies used the “reading text” stimulus, which may have resulted in different findings to this pilot study, which asked participants to use their phone in a “habitual” way. One study used a video,⁹ and another related symptoms to self-reported hours of smartphone use.⁶ In support of the study by Ram et al.⁹ the smartphone uses of today are changing to include activities such as videos, socializing, and shopping, however, the study did not control this activity as they asked participants to do this 1 hour before bed. It was unclear when and how the symptoms were recorded and whether the timing of smartphone use caused the effects reported in that study.

A substantial difference found between this pilot and the existing research was that many other studies supporting the negative effect of smartphone use on near triad measures did not adjust their statistical testing (p-value) to account for multiple tests of difference. The one study that found no apparent difference in vergence between no task, tablet, and smartphone had used post-hoc analysis to account for multiple tests.⁸ This suggests that had other studies undertaken post-hoc adjustments to their accepted significance level, they would not find a difference.

Further work is required to account for the limitations in the current literature and this pilot study. A median of the three results were taken for this study, but there was no randomization in the order of testing or the eye tested. This may result in differences found in repeated measures; therefore, any difference found must account for this, and randomization may go some distance to prevent the effects of fatigue. Within this pilot, there was no record of what the participants carried out on their smartphones during the 30 minutes which if collected could provide an opportunity to understand the results found. There was also no control over the device's type and settings, which could have prevented a difference being found.

While these differences are not statistically or clinically significant, the question of the effect of smartphone use on the near triad remains inconclusive and open to further study. A larger sample size reaching higher statistical power or longer use of the smartphone may reveal effects on the near triad. Alternatively, smartphone use may be monitored or reported and then compared to near triad measures. This is similar to the study by Kim et al.⁶ who did find a significant relationship between the duration of smartphone use and symptoms in adolescents. Those with higher levels of smartphone use were 2.26 times more likely to fall into the group with 5–7 out of 7 symptoms compared to the reference group with short-term use. This study looked only at symptoms, but a similar design could be applied with the addition of using ocular measures to compare also to smartphone use.

In conclusion, a future study is planned where smartphone data, a modified symptom questionnaire, a diary of smartphone use, and accommodative measures will be used to further investigate the effects of smartphone use on the near triad. Learning points from this pilot that will be included are the effect of long-term usage could not be captured through this pilot methodology, some way to capture and analyze smartphone features was required, ordering of testing procedures requires randomization.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The author(s) reported that there is no funding associated with the work featured in this article.