Indoor and outdoor human behavior and myopia: an objective and dynamic study

Abstract

Significance: Myopia holds significant public health concern given its social, ocular disease and economic burdens. Although environmental factors are primarily to blame for the rapid rise in prevalence, key risk factors remain unresolved.

Purpose: The aim of this study was to objectively characterize, using a wearable technology, the temporal indoor and outdoor behavioral patterns and associated environmental lighting characteristics of young myopic and nonmyopic University students.

Methods: Participants were recruited to continuously wear an Actiwatch for 3 weeks, during either or both academic and non-academic periods. The device allows continuous recording of activity and incident light. Recorded illuminance levels were used as a proxy for outdoors (>1,000 lux), with the dynamics (interval frequency and duration) of indoor and outdoor activities, as well as lighting characteristics derived. In addition, participant input regarding near work was obtained daily. Participants were classified by both myopia and axial length status (based on collected refractive error and biometry data) for the purpose of data analysis.

Result: A total of 55 students, aged 18 to 25 years of age, participated. Overall, the dosing of indoor and outdoor activities was similar across participants, regardless of myopia status, during the academic period. Nonetheless, an apparent difference in the timing of outdoor activities was noted with myopes going outdoors later in the day, particularly during the weekend (p = 0.03). While a trend was observed between increased lighting levels experienced outdoors and shorter axial lengths, there was no significant relationship with myopia status. Noteworthy, participants generally significantly overestimated time spent outdoors, compared to Actiwatch-derived estimates of the same.

Conclusion: While the findings from this cohort of young adult students did not reveal substantial myopia-related differences in behavior, the power of a more objective and dynamic approach to quantifying behavior cannot be understated, providing argument for general adoption of wearable technologies in future clinical myopia studies.

Keywords: myopia; near work; outdoor time; visual environment; wearable technologies.

Figure 1

Subjective reports vs. objective recordings of outdoor activity. Both myopes and non-myopes tended to over-estimate time spent outdoors on subjective questionnaires, although myopes to a slightly less extent on average [mean overestimation (SD): myopes: 116 (68.3) mins, non-myopes: 145 (74.2) mins].

Figure 2

Outdoor activity interval dosing in myopes and non-myopes. Refractive error [SER, (A,C)] and axial length [AL, (B,D)] plotted against mean frequency of daily outdoor intervals (A,B) and interval duration (C,D) for myopes (blue) and non-myopes (red); behaviors are similar for the two groups and neither parameter is significantly correlated with either SER (A,C) or AL (B,D).

Figure 3

Outdoor activity dynamics. Outdoor activity dynamics defined in terms of the time of day of outdoor activities, shown by myopia and axial length categories for both weekdays [(A,C), respectively] and weekends [(B,D), respectively]. On each violin plot is superimposed a box plot denoting the related median of the distribution and +/− 1SD whiskers.

Figure 4

Outdoor lighting characteristics in myopes and non-myopes. Refractive errors (SERs) and axial lengths (ALs), plotted against mean daily illuminance (lux, top panel) and spectral composition expressed as R:B ratios (bottom panel) to which participants were exposed. Daily outdoor illuminance was slightly lower on average for myopes (blue) compared to non-myopes (red) (A). Superimposed solid lines represent results of correlation analyzes. Outdoor illuminance and axial length are weakly correlated (B), while the spectral composition of outdoor lighting shows no correlation in either SER (C) or AL (D).

Figure 5

“Near” behavior of myopes and non-myopes. Refractive errors [SERs, (A)] and axial lengths [ALs, (B)] plotted against mean daily reported “near” activity for myopes (blue) and non-myopes (red); the two groups showed similar and quite variable behavior, which was not significantly correlated with either SER or AL.

Figure 6

Indoor activity interval dosing in myopes and non-myopes. Refractive errors [SERs, (A)] and axial lengths [ALs, (B)] plotted against mean daily indoor interval frequency (A,B) and interval duration (C,D); trends were both similar for both myopes (blue) and non-myopes (red) and neither variable was significantly correlated with either SER (A,C) or AL (B,D).

Figure 7

Indoor activity lighting characteristics in myopes and non-myopes. Refractive errors (SERs) and axial lengths (ALs) plotted against mean indoor daily illuminance (lux) (A,B), and spectral composition, as expressed as the R:B ratio (C,D). Compared to non-myopes (red), myopes (blue) experienced slightly higher mean indoor daily illuminance although no significant correlation with SER (A) or AL (B) was observed. The spectral composition of indoor lighting was similar for non-myopes and myopes, and for myopes, R:B ratios were significantly correlated with both SER [p = 0.04, (C)] and AL [p = 0.03, (D)].