Abstract
Background
We investigated the impact of social distancing policies (SDPs) on ophthalmic severity in children who underwent emergency ophthalmic referrals during the coronavirus disease 2019 pandemic period.
Methods
We reviewed all children with ophthalmic referrals in a single academic hospital emergency department during the period from February 2017 to December 2019 (prepandemic) or February 2020 to December 2022 (pandemic). Baseline features, diagnosis‐based severity, and outcomes were compared between the two periods. The Government Response Stringency Index (GRSI), which ranges from 0 to 100, was used as a surrogate for the intensity of SDPs during the pandemic. Poisson regression was used to quantify the association of the GRSI with the severity.
Results
Among 1074 children with ophthalmic referrals, 437 (40.7%) visited during the pandemic. This was 31.4% lower than that during the prepandemic period. In numbers, pandemic‐related declines were more modest in high severity than in medium‐to‐low severity (35.1% vs. 55.0%), and for injury than for illness (28.5% vs. 36.1%). In percentages, high severity increased from 63.3% to 71.3% (p = 0.016). The hospitalization rate increased from 1.7% to 3.9% (p = 0.029). For every 10‐point increase in GRSI, there was a 20.0% decrease in high severity (95% confidence interval, 5%–30%).
Conclusions
This study shows an inverse association of SDPs with ophthalmic severity and an increase in severe cases along with consistent flow of injury cases, amid the overall decline in eye‐related visits to the emergency department during the pandemic period.
Keywords: COVID‐19, eye diseases, eye injuries, ophthalmology, physical distancing
INTRODUCTION
The coronavirus disease 2019 (COVID‐19) pandemic strengthened compulsory viral testing and isolation processes in emergency departments (EDs), social distancing policies (SDPs), and guardians' concerns for nosocomial infection, leading to a 56.7% to 63.8% decrease in overall pediatric ED visits. 1 , 2 , 3 , 4 This downtrend would have been reproduced in children's eye‐related ED visits in developed countries. Specifically, pediatric eye‐related visits to or ophthalmic referrals in EDs decreased by 51.0%–89.5% in the USA, 5 , 6 51.6% in Israel, 7 47.1%–50.6% in Türkiye, 8 , 9 46.1% in France, 10 and 42.8% in Spain. 11 Previous studies reflect government policies in a qualitative fashion—for example, lockdown versus post‐lockdown. Furthermore, they used diagnoses based on the International Classification of Diseases codes or broad terms that were not confirmed by pediatric ophthalmologists.
Although the COVID‐19 pandemic is over, for future pandemics, pediatricians and emergency physicians need to allocate pediatric emergency resources properly for specific needs, such as ophthalmologist practices. Hence, we analyzed the pandemic‐related changes in pediatric ophthalmic referrals, and tested a possible association of the governmental stringency of SDPs, such as school closure, with ophthalmic severity (hereafter, “severity”) in children who visited an ED in Korea. For this purpose, we used quantitative data on the intensity of SDPs during the pandemic and detailed diagnoses provided by pediatric ophthalmologists.
METHODS
Study design, setting, and population
This retrospective study is a planned secondary analysis of a single academic hospital ED‐based study on pediatric ophthalmic referrals. 12 The hospital is situated in Gyeonggi‐do Province, Korea, which has an under‐18 population of approximately 2,200,000. Before the COVID‐19 pandemic, the ED attended to roughly 21,000 children annually, which adjusted to 12,000–19,000 during the pandemic. The ED is staffed by board‐certified emergency physicians or pediatricians, or emergency medicine residents in postgraduate years 4–5 (ie, senior residents in Korea), and maintains a 24/7 on‐call ophthalmologist practice. Children presenting with ocular symptoms are initially evaluated by the professionals and, if it is deemed necessary, are referred to the ophthalmologists. The criteria for referral did not change throughout the pandemic. Routine ophthalmic examinations are conducted in order of visual acuity, intraocular pressure, slit lamp, pupil, ocular movement, and fundus, using age‐appropriate methods. 13 , 14
We included children (<18 years) with ocular injury or illness, who visited the ED from February 1, 2017 through December 31, 2019 (prepandemic) or from February 1, 2020 through December 31, 2022 (pandemic). In the case of multiple visits, the index visit was analyzed to ensure the independence of each visit. Exclusion criteria were visits for procedural complications and discharges before thorough ophthalmic examinations. This study was approved by the institutional review board with a waiver for informed consent (IRB no. AJOUIRB‐DB‐2024‐066).
Coronavirus disease 2019 situation in Korea
The pandemic period was defined based on the World Health Organization's declaration of “public health emergency of international concern,” which was valid from January 30, 2020 through May 5, 2023. In Korea, the first adult case of COVID‐19 was confirmed on January 20, 2020, and the epidemic started in the following month. During the period from March 2020 to April 2022, Korean government implemented a set of SDPs, such as mask mandates, without issuing a stay‐at‐home order. The school closure phase was defined as February to May 2020 (cf., school attendance, June 2020 to December 2022). The decision was based on the fact that after the winter break (late December to early February), school reopening was postponed until mid‐February, and the schools reopened gradually from May 13 through June 1, 2020. 15
Government Response Stringency Index
The Government Response Stringency Index (GRSI) is a quantitative, country‐based, composite measure based on nine indicators of SDPs, including school closure, workplace closure, cancellation of public events, restrictions on gatherings, public transportation, stay‐at‐home order, restrictions on internal movement, international travel controls, and public information campaigns. 16 The index, which ranges from 0 to 100, was reported daily during 2020 and 2022 on the Oxford COVID‐19 Government Response Tracker. 16 The GRSI quantified how Korean governmetal SDPs were stringent. 3 , 16 Among the six developed countries cited in the introduction, the mean GRSI was lowest in Korea (Supporting Information, Table S1).
Data collection
The following data were collected: age, age groups (0–3, 4–6, 7–12, and 13–17 years), sex, injury, transfer form another hospital, at night, and on weekend/holiday, ED‐assigned high acuity (Korean Triage and Acuity Scale level 1–2), 12 , 17 arrival‐referral time (minutes from arrival at the ED to the first referral), referral‐answer time (minutes from the first referral to posting ophthalmologist's diagnosis), diagnosis‐based severity (see details in below section), emergency surgery or hospitalization to Department of Ophthalmology. We analyzed the mechanism and place of injury, and chief illness complaint. 12
Diagnosis‐based severity
Diagnoses entered by the ophthalmologists in the medical records were rated by the following levels of severity: “high,” requiring emergency surgery or hospitalization (eg, corneal penetrating laceration), “medium,” requiring only on‐call ophthalmologist practice (eg, uveitis), and “low,” requiring follow ups or medication (eg, conjunctivitis). This rating was primarily based on a table from a 2006–2011 US Nationwide ED sample‐based study, which classified ophthalmic diagnoses as “likely emergent,” “unlikely emergent,” and “could not determine,” corresponding to the “high,” “medium,” and “low” severities, respectively. 18 Unrated diagnoses were evaluated by a pediatric ophthalmologist (SA Chung). In the case of multiple diagnoses, a representative one was chosen based on the severity and clinical context. Diagnoses that did not require ophthalmic referrals, such as eyelid injury repaired by an emergency physician, were not considered.
Statistical analysis
Data were presented as medians with interquartile ranges or means ± standard deviations and as percentages and numbers for continuous and categorical variables, respectively. They were compared using the Mann–Whitney U‐tests, analyses of variance, χ2 tests, or Fisher's exact tests. Given the expected nature of the pandemic‐related decrease in number of visits, we primarily compared percentages, unless specified as “in number (s).” To quantify a possible association of the stringency of SDPs with severity as a dependent variable, Poisson regression models were constructed with month, year, and GRSI as fixed effects. Regression discontinuity in time was used to estimate a pandemic effect on high severity. Statistical significance was defined as p < 0.05. We used R Statistical Software, version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria) and IBM SPSS Statistics for Windows, version 25.0 (IBM Corp, Armonk, NY).
RESULTS
Study population
During the prepandemic period, 84,401 children visited the ED. Of these, 1695 had ocular symptoms, of whom 660 were referred to the ophthalmologists. During the pandemic, 40,649 children (down by 51.8% in numbers from the prepandemic period) visited the ED. Of these, 918 had ocular symptoms, of whom 446 were referred to the ophthalmologists. After excluding 32 children, we analyzed 1074, which consisted of 637 prepandemic cases and 437 cases during the pandemic; down by 31.4% in number, from the prepandemic period; Figure 1).
FIGURE 1.
Flowchart for selection of the study population. “High,” “Medium,” and “Low” denote OPH severity. OPH, ophthalmic; Sx, symptom.
Baseline characteristics
From the pre‐pandemic to pandemic periods, increases were noted in the age of 7–12 years, high severity, and hospitalization, while decreases were noted in the age of 13–17 years, visits via transfers, and medium severity (Table 1). In terms of numbers, high severity cases were down by 35.1% (from 345 to 224), whereas medium‐to‐low severity cases were down by 55.0% (from 200 to 90). Injury showed no change in percentages, and a smaller decrease in numbers compared to illness (injury, down by 28.5% from 393 to 281 vs. illness, down by 36.1% from 244 to 156). As for the mechanism or place of injury, or chief complaint of illness, both in percentages and numbers, cases of falls, home, and foreign body sensation/mass increased, whereas motor vehicle accidents and swelling/erythema decreased (Table 2).
TABLE 1.
Baseline characteristics of the study population.
| Variable | Total (N = 1074) | Prepandemic (N = 637) | Pandemic (N = 437) | p |
|---|---|---|---|---|
| Age, years | 8.0 (3.0–14.0) | 9.0 (3.0–14.0) | 8.0 (3.0–13.0) | 0.229 |
| Age group, years | ||||
| 0–3 | 277 (25.8) a | 162 (25.4) a | 115 (26.3) | 0.018 |
| 4–6 | 164 (15.3) a | 99 (15.5) a | 65 (14.9) | |
| 7–12 | 280 (26.1) a | 147 (23.1) a | 133 (30.4) | |
| 13–17 | 353 (32.9) a | 229 (35.9) a | 124 (28.4) | |
| Girls | 357 (33.2) | 206 (32.3) | 151 (34.6) | 0.449 |
| Injury | 674 (62.8) | 393 (61.7) | 281 (64.3) | 0.385 |
| Visit via transfer | 343 (31.9) | 219 (34.4) | 124 (28.4) | 0.038 |
| Visit at night | 536 (49.9) | 311 (48.8) | 225 (51.5) | 0.391 |
| Visit on weekend/holiday | 397 (37.0) | 225 (35.3) | 172 (39.4) | 0.178 |
| ED‐assigned high acuity b | 43 (4.0) | 29 (4.6) | 14 (3.2) | 0.268 |
| Arrival‐referral time, min | 59.0 (31.0–108.0) | 58.0 (29.0–108.0) | 61.0 (37.0–103.0) | 0.097 |
| Referral‐answer, min | 21.0 (6.0–56.3) | 15.0 (4.0–42.5) | 30.0 (11.0–80.0) | <0.001 |
| High severity c | 569 (66.2) | 345 (63.3) | 224 (71.3) a | 0.016 |
| Medium severity c | 165 (19.2) | 120 (22.0) | 45 (14.3) a | 0.006 |
| Low severity c | 125 (14.6) | 80 (14.7) | 45 (14.3) a | 0.889 |
| Emergency surgery | 30 (2.8) | 14 (2.2) | 16 (3.7) | 0.153 |
| Hospitalization c , d | 28 (2.6) | 11 (1.7) | 17 (3.9) | 0.029 |
Note: Values are expressed as medians (interquartile ranges) or numbers (with percentages in parentheses).
Abbreviation: ED, emergency department.
The sums of proportions are not equal to 100% due to rounding.
Korean Triage and Acuity Scale 1–2.
Given the absence of ophthalmic diagnoses, the denominators are 859, 545, and 314 in the order of column.
To Department of Ophthalmology.
TABLE 2.
Mechanism and place of injury, or chief complaint of illness.
| Variable | Total (N = 1074) | Prepandemic (N = 637) | Pandemic (N = 437) | p |
|---|---|---|---|---|
| Mechanism of injury | ||||
| Sports | 160 (14.9) | 102 (16.0) | 58 (13.3) | 0.215 |
| Impact/compression | 112 (10.4) | 74 (11.6) | 38 (8.7) | 0.124 |
| Poke | 67 (6.2) | 36 (5.7) | 31 (7.1) | 0.337 |
| Fall | 86 (8.0) | 39 (6.1) | 47 (10.8) | 0.006 |
| Assault | 74 (6.9) | 42 (6.6) | 32 (7.3) | 0.643 |
| Cut/scratch/pierce | 76 (7.1) | 40 (6.3) | 36 (8.2) | 0.219 |
| Chemical/burn | 57 (5.3) | 28 (4.4) | 29 (6.6) | 0.108 |
| Motor vehicle accident | 23 (2.1) | 19 (3.0) | 4 (0.9) | 0.021 |
| Foreign body | 19 (1.8) | 13 (2.0) | 6 (1.4) | 0.415 |
| Places of injury | ||||
| Home | 273 (25.4) | 136 (21.4) | 137 (31.4) | <0.001 |
| Playground | 186 (17.3) | 119 (18.7) | 67 (15.3) | 0.154 |
| Road | 109 (10.1) | 74 (11.6) | 35 (8.0) | 0.054 |
| Educational facility | 73 (6.8) | 43 (6.8) | 30 (6.9) | 0.942 |
| Public | 30 (2.8) | 19 (3.0) | 11 (2.5) | 0.649 |
| Others | 3 (0.3) | 2 (0.3) | 1 (0.2) | >0.999 |
| Chief complaint of illness | ||||
| Swelling/erythema | 86 (8.0) | 67 (10.5) | 19 (4.3) | < 0.001 |
| Red eye | 47 (4.4) | 31 (4.9) | 16 (3.7) | 0.343 |
| Eye pain | 36 (3.4) | 16 (2.5) | 20 (4.6) | 0.065 |
| Itching sensation | 29 (2.7) | 21 (3.3) | 8 (1.8) | 0.145 |
| Visual loss/blur | ||||
| Painless | 36 (3.4) | 16 (2.5) | 20 (4.6) | 0.065 |
| Painful | 29 (2.7) | 18 (2.8) | 11 (2.5) | 0.759 |
| Other visual | 29 (2.7) | 17 (2.7) | 12 (2.7) | 0.939 |
| Foreign body sense/mass | 24 (2.2) | 9 (1.4) | 15 (3.4) | 0.028 |
| Rash/vesicle | 25 (2.3) | 18 (2.8) | 7 (1.6) | 0.191 |
| Eye discharge | 14 (1.3) | 10 (1.6) | 4 (0.9) | 0.353 |
| Diplopia | 23 (2.1) | 12 (1.9) | 11 (2.5) | 0.481 |
| Other symptoms | 22 (2.0) | 9 (1.4) | 13 (3.0) | 0.076 |
Note: Values are expressed as numbers (with percentages in parentheses).
Trends in ophthalmic referrals, high severity cases, and effect of SDPs
As Figure 2a shows, during the early pandemic period, numbers of injury‐ and illness‐induced referrals reached their nadirs. This overlapped with the school closure phase, with subsequent rebounds. Meanwhile, a decline in high severity cases lagged behind such declines in the referrals. A trade‐off was observed between high severity and GRSI. To further analyze the trade‐off, we compared the school closure and attendance phases. High severity was more frequently observed in the school closure phase (closure, 92.0% [23/25] vs. attendance, 69.6% [201/289]; p = 0.017). In contrast, medium severity cases showed no difference (8.0% [2/25] vs. 14.9% [43/289]; p = 0.551), and low severity cases were more common in the school attendance phase (0% [0/25] vs. 15.6% [45/289]; p = 0.033).
FIGURE 2.
Trends in OPH referrals, high severity, and effects of social distancing policies. Vertical lines indicate the starting point of the coronavirus disease 2019 pandemic. (a) Quarterly trend of the referrals, high severity, and GRSI. Open and shaded bars indicate the numbers of the injury‐ and illness‐induced referrals, respectively, during each 3 month interval. “SC” denotes the school closure phase, coinciding with a nadir of referrals (open arrow; February–April 2020), and preceding a nadir of high severity (shaded arrow; May–July 2021). A trade‐off is shown between high severity (dashed line) and GRSI (dotted line). (b) Regression discontinuity in time analysis. Each circle and triangle indicates the monthly numbers of high severity. Shades indicate 95% CIs of slopes. A slope during the prepandemic (solid line) is 0.07 (95% CI, −0.04 to 0.18; p = 0.221), whereas a slope during the pandemic (dash line) is 0.10 (95% CI, −0.06 to 0.25; p = 0.226). An asterisk indicates an estimated change in monthly number of high severity (−7.49; 95% CI, −10.65 to −4.32; p < 0.001). CI, confidence interval; GRSI, Government Response Stringency Index; OPH, ophthalmic.
In the Poisson regression, after controlling for year and month, GRSI was negatively associated with the frequency of high severity (coefficient, −0.02; 95% confidence interval, −0.03 to −0.005), while not with medium or low severity (Table 3). The monthly number of children with high severity cases decreased by 7.49 during the pandemic period (Figure 2b).
TABLE 3.
Poisson regression models.
| Independent variable | Coefficient (95% confidence interval) | Standardized coefficient | Standard error | Z value | p |
|---|---|---|---|---|---|
| High severity a | |||||
| Intercept | 144.21 (−424.24 to 687.43) | — | 282.98 | 0.51 | 0.610 |
| Year | −0.07 (−0.34 to 0.21) | 0.93 | 0.14 | −0.50 | 0.615 |
| Month | 0.05 (0.01 to 0.09) | 1.05 | 0.02 | 2.21 | 0.027 |
| GRSI | −0.02 (−0.03 to −0.005) | 0.98 | 0.01 | −2.71 | 0.007 |
| Medium severity b | |||||
| Intercept | −625.60 (−2112.57 to 646.94) | — | 696.60 | −0.90 | 0.369 |
| Year | 0.31 (−0.32 to 1.04) | 1.36 | 0.34 | 0.90 | 0.368 |
| Month | 0.02 (−0.08 to 0.13) | 1.02 | 0.05 | 0.45 | 0.651 |
| GRSI | −0.02 (−0.05 to 0.01) | 0.98 | 0.02 | −1.19 | 0.235 |
| Low severity c | |||||
| Intercept | −2155.66 (−4044.89 to −639.27) | — | 865.68 | −2.49 | 0.013 |
| Year | 1.07 (0.32 to 2.00) | 2.91 | 0.43 | 2.49 | 0.013 |
| Month | 0.00 (−0.12 to 0.13) | 1.00 | 0.06 | 0.00 | 0.996 |
| GRSI | 0.00 (−0.03 to 0.04) | 1.00 | 0.02 | 0.10 | 0.921 |
Abbreviation: GRSI, Government Response Stringency Index.
Akaike information criterion = 164.5; Hosmer‐Lemeshow goodness‐of‐fit test, p = 0.571.
Akaike information criterion = 104.2; Hosmer‐Lemeshow goodness‐of‐fit test, p = 0.829.
Akaike information criterion = 102.3; Hosmer‐Lemeshow goodness‐of‐fit test, p = 0.906.
Diagnosis‐based severity
The diagnoses are listed in Supporting Information, Table S2. Among the top five diagnoses, all injuries were rated as high severity without an overall difference (prepandemic, 42.7% vs. pandemic, 39.8%; p = 0.346; Table 4). In contrast, the illnesses showed heterogeneous severities with an overall decrease (23.2% vs. 17.8%; p = 0.033). In detail, hyphema and cellulitis decreased, whereas chemical burns and corneal abrasions due to illness increased in both percentages and numbers. Among the categories of cellulitis, periorbital cellulitis (the sole medium severity case) decreased by 6.4%, whereas the other categories showed a 0.2% to 0.4% change. During the pandemic period, among the top five diagnoses, injury was more common in the school‐closure phase (Figure 3).
TABLE 4.
Top five diagnoses.
| Diagnosis | Total (N = 1074) | Prepandemic (N = 637) | Pandemic (N = 437) | p |
|---|---|---|---|---|
| Injury | 446 (41.5) | 272 (42.7) | 174 (39.8) | 0.346 |
| Hyphema (H) | 144 (13.4) | 101 (15.9) | 43 (9.8) | 0.004 |
| Orbital fracture (H) | 141 (13.1) | 86 (13.5) | 55 (12.6) | 0.663 |
| Corneal abrasion (H) | 76 (7.1) | 42 (6.6) | 34 (7.8) | 0.456 |
| Laceration (H) | 45 (4.2) | 26 (4.1) | 19 (4.3) | 0.831 |
| Chemical burn (H) | 40 (3.7) | 17 (2.7) | 23 (5.3) | 0.027 |
| Illness | 226 (21.0) | 148 (23.2) | 78 (17.8) | 0.033 |
| Keratoconjunctivitis (H‐L) | 119 (11.1) | 74 (11.6) | 45 (10.3) | 0.499 |
| Herpes zoster (H) | 26 (2.4) | 24 (3.8) | 2 (0.5) | |
| Herpes simplex (H) | 8 (0.7) | 5 (0.8) | 3 (0.7) | |
| Glaucomatocyclitic crisis (H) | 2 (0.2) | 0 (0) | 2 (0.5) | |
| Bacterial (M) | 8 (0.7) | 4 (0.6) | 4 (0.9) | |
| Viral/idiopathic (M) | 21 (2.0) | 4 (0.6) | 17 (3.9) | |
| Keratoconjunctivitis sicca (L) | 2 (0.2) | 2 (0.3) | 0 (0) | |
| Epidemic (L) | 22 (2.0) | 20 (3.1) | 2 (0.5) | |
| Allergic (L) | 29 (2.7) | 15 (2.4) | 14 (3.2) | |
| Phlyctenular (L) | 1 (0.1) | 0 (0) | 1 (0.2) | |
| Cellulitis (H‐M) | 68 (6.3) | 57 (8.9) | 11 (2.5) | <0.001 |
| Orbital (H) | 7 (0.7) | 3 (0.5) | 4 (0.9) | |
| Dacrocystitis (H) | 4 (0.4) | 3 (0.5) | 1 (0.2) | |
| Dacryoadenitis (H) | 1 (0.1) | 1 (0.2) | 0 (0) | |
| Periorbital (M) | 56 (5.2) | 50 (7.8) | 6 (1.4) | |
| Hordeolum/chalazion (L) | 15 (1.4) | 8 (1.3) | 7 (1.6) | 0.635 |
| Corneal abrasion (H) | 14 (1.3) | 4 (0.6) | 10 (2.3) | 0.018 |
| Optic neuritis (M) | 10 (0.9) | 5 (0.8) | 5 (1.1) | 0.539 |
Note: Values are expressed as numbers (with percentages in parentheses). Severity is parenthesized. H, High; M, Medium; L, Low.
FIGURE 3.
Top five diagnoses during the pandemic period. Injury was more common in the school closure phase than in the school attendance phase (76.2% vs. 65.6%). A gap between the percentages of injury and illness was larger in the former phase (52.4% = 76.2% [16/21]–23.8% [5/21]) than in the latter phase (31.2% = 65.6% [139/212]–34.4% [73/212]).
DISCUSSION
This study has the following implications arisen from the long‐term changes in ophthalmologist practice in the ED due to the COVID‐19 pandemic period. During the pandemic, the stringency of SDPs imposed by Korean government was inversely associated with high severity cases. From the prepandemic period to the pandemic period, the reductions in numbers were more modest in high‐severity cases than in medium‐to‐low severity cases, and in injury cases than in illness cases.
This study is comparable to the relevant studies in terms of the age, diagnosis, and long‐term pandemic‐related changes. The median age of our study population approximates to the mean age reported in US and Turkish studies (8.0 vs. 7.9–9.7 years). 6 , 8 Many studies consistently cited trauma, conjunctivitis, and hordeolum as frequent diagnoses, 5 , 6 , 7 , 10 as well as foreign bodies, corneal abrasions, and cellulitis in some studies. 8 , 11 These are encompassed by our top five diagnoses (Table 4). However, the decrease in the number of ophthalmic referrals was smaller than the previously reported (31.4% vs. 42.8%–89.5%)5, 6, 7, 8, 9, 10, 11 The disparity might be partly explained by our longer study period compared with that of previous studies (35 months in our study vs. 2–13 months in previous studies5, 6, 7, 8, 10 The nadir during the early pandemic followed by the rebound of referrals in our study (Figure 2a) is in agreement with a Spanish study that showed a 57.4% decrease in annual eye‐related ED visits in 2019–2020, followed by a 68.5% increase in 2020–2021. 11
As shown by the regression model, the negative association of GRSI with severity means that for every 10‐point decrease in GRSI, there was a 20.0% increase on average in high severity cases (95% confidence interval, 5%–30%). In that model, the positive association of “month” with high severity is parallel to the time‐related increase in that severity during the pandemic period (Figure 2b). Korea's lowest GRSI among the developed countries (Table S1) proves low‐key SDPs in Korea, consistent with the absence of a stay‐at‐home order. High‐severity cases increased in studies performed in France and Türkiye, where the mean GRSIs were similar to that of Korea (Supporting Information, Table S1). 8 , 10 In contrast, studies in the USA and Spain, in which GRSIs were higher than that of Korea, showed decreases in injury and high severity. 5 , 11
We speculate that in Korea, the low‐key SDPs maintained children's outdoor activities to some level, thereby leading to the slight rise in high‐severity cases and the steady flow of injury during the pandemic period. A US nationwide study shows pandemic‐related decreases in injuries related to sports (from 6.5% to 3.9%) and school activities (from 3.5% to 1.5%). 19 In contrast, we observed no such decreases in “playground” and “educational facility” injuries (Table 2). This discrepancy might be due to the less stringent SDPs in Korea than in the USA (Supporting Information, Table S1). The low‐key SDPs in Korea likely allowed children to engage in injury‐prone activities, such as sports, more frequently than their US counterparts.
We need to discuss the background of the rise in high severity along with steady flow of injury, amid an overall decrease in eye‐related ED visits. First, an ocular injury inherently indicates a higher severity than an illness affecting the eyes. For example, laceration or foreign body is considered more urgent than conjunctivitis or hordeolum. 5 , 18 An injury like a corneal penetrating laceration or a chemical burn may directly jeopardize visual acuity. From the perspective of guardians, such injuries drive them to bring their children to EDs even during a pandemic period. This contrasts with conjunctivitis, which can be treated temporarily at home.
Second, hand washing or mask mandates primarily reduce the spread of viruses, 20 , 21 , 22 as shown by a decrease in conjunctivitis. 5 , 11 , 23 This phenomenon was exemplified by the fact that the injury's predominance over illness was greater in the school closure phase than in the school attendance phase (Figure 3). As shown in Figure 2a, the school closure overlapped with the nadir of overall referrals (i.e., injury plus illness). Thus, the SDP‐induced prevention of viral infection may explain why the nadir of referrals preceded that of high severity. In contrast with the illnesses, which declined sharply, injury showed conflicting changes (Supporting Information, Table S1). We noted increases in numbers of domestic injury (Table 2) and chemical burns caused by hand sanitizers or disinfectants (Table 4) (prepandemic, one case vs. pandemic, five cases; data not shown). Such increases likely reflected pandemic‐related behavioral changes, and counterbalanced the school‐closure‐induced decline in injury. 7 , 24 , 25 , 26
Third, the low‐key SDPs and relatively high activity level in Korea might have led to the steady occurrence of injury (Table 1). This feature may be related to a 22.4% gap in high severity between the school closure and attendance phases (92.0% vs. 69.6%, respectively), which is larger than the 8.0% gap in the severity between the prepandemic and pandemic periods (63.3% vs. 71.3%; Table 1).
This study has several limitations. First, being based in a single ED, the results may not be widely generalizable. Medical and social responses to a pandemic vary based on local circumstances. Thus, the implications of this study should be carefully applied to similar academic hospitals with the capability of providing a 24/7 on‐call ophthalmologist practice. Second, the lack of standardized severity criteria for pediatric ophthalmic diagnoses suggests a potential for classification bias, despite our efforts to rate severity as rigorously as possible. Third, there might be another factor confounding the association of the SDPs with severity. We attempted to minimize this issue by controlling for month, year, and GRSI in the regression models. Social factors were probably controlled to some degree by the use of GRSI, which encompasses many social factors.
In conclusion, Korean government stringency with regard to SDPs was inversely associated with high severity cases during the COVID‐19 pandemic period. During the period, children with high severity cases or injuries were referred to the ophthalmologists at a similar or higher frequency than those with medium‐to‐low severity or illness. This tendency could be related to the low‐key SDPs in Korea and the fact that outdoor activity levels of children were maintained during the pandemic. During future pandemics, allowing for some outdoor activity may require preparing pediatric emergency resources for ophthalmologist practice, focusing on vision‐threatening injuries. Targeted education on the prevention of ocular injury should also be implemented and adapted to the intensity of SDPs. During periods of high stringency (e.g., high GRSI or lockdown), guardians should be vigilant about domestic injuries, particularly from falls or hand sanitizers. As the restrictions reduce, outdoor activity can be a major source of ocular injury. Thus, the focus of education should shift towards awareness of injury‐prone activities, along with relevant preventive measures (e.g., use of protective eyewear for ball sports).
AUTHOR CONTRIBUTIONS
Hyung Kyoo Woo, Seung Ah Chung, and Jung Heon Kim designed the study. Hyung Kyoo Woo, Seung Ah Chung, and Jung Heon Kim collected and analyzed the data. Hyelynn Jeon, Bumhee Park, and Jung Heon Kim performed the statistical analyses. Hyung Kyoo Woo, Seung Ah Chung, and Jung Heon Kim drafted the manuscript. All authors read and approved the final manuscript.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.
ETHICAL APPROVAL
This study was approved by the institutional review board with a waiver for informed consent (IRB no. AJOUIRB‐DB‐2024‐066).
Supporting information
Table S1.
Table S2.
Woo HK, Chung SA, Jeon H, Park B, Kim JH. Impact of social distancing policy on pediatric emergency ophthalmic severity during the coronavirus disease 2019 pandemic. Pediatr Int. 2024;66:e15845. 10.1111/ped.15845
REFERENCES
- 1. Dopfer C, Wetzke M, Zychlinsky Scharff A, Mueller F, Dressler F, Baumann U, et al. COVID‐19 related reduction in pediatric emergency healthcare utilization ‐ a concerning trend. BMC Pediatr. 2020;20:427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Kruizinga MD, Peeters D, van Veen M, van Houten M, Wieringa J, Noordzij JG, et al. The impact of lockdown on pediatric ED visits and hospital admissions during the COVID19 pandemic: a multicenter analysis and review of the literature. Eur J Pediatr. 2021;180:2271–2279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Choi DH, Jung JY, Suh D, Choi JY, Lee SU, Choi YJ, et al. Impact of the COVID‐19 outbreak on trends in emergency department utilization in children: a multicenter retrospective observational study in Seoul metropolitan area, Korea. J Korean Med Sci. 2021;36:e44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hwang SY, Lee JK, Ryu HS, Park SS, Choi JY, Lee HJ, et al. Long‐term impact of coronavirus disease 2019 pandemic on emergency department utilization in a metropolitan emergency department in Korea. Pediatr Emerg Med J. 2021;8:57–65. (in Korean). [Google Scholar]
- 5. Jin J, Bules L, Doctor K, Hendricks D, Callaghan K, Reid JE, et al. Visits to the pediatric emergency department for eye conditions before and during the COVID‐19 pandemic. West J Emerg Med. 2022;23:424–431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Shah K, Camhi SS, Sridhar J, Cavuoto KM. Impact of the coronavirus pandemic on pediatric eye‐related emergency department services. J AAPOS. 2020;24:367–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Levy I, Dollberg D, Berant R, Friling R. Trends in pediatric ophthalmic emergency department visits during the COVID‐19 pandemic. Isr Med Assoc J. 2022;24:289–292. [PubMed] [Google Scholar]
- 8. Durmaz Engin C, Ozturk T, Akbulut Yagci B, Ozcelik O, Ecer R. The impact of the COVID‐19 pandemic on emergency department visits resulting in ophthalmology consultations. Cureus. 2022;14:e30598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kilic D. The effect of COVID‐19 pandemic on the characteristics of patients presenting to the pediatric ocular emergency department in Türkiye: demographic and diagnoses. Ulus Travma Acil Cerrahi Derg. 2022;29:46–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Martin GC, Boulanger E, Maalej R, Partouche S, Dentel A, Grosselin M, et al. Specificities of pediatric ocular emergencies before and during the COVID‐19 era: a retrospective comparative study in an eye‐related emergency department in Paris. Arch Pediatr. 2023;30:396–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Álvarez‐Bulnes O, Casals‐Sánchez A, Roig‐Albertos M. Epidemiology of eye‐related emergencies in children under 18 years of age at a general Hospital in Spain from 2019 to 2021. Arch Soc Esp Oftalmol (Engl Ed). 2023;98:564–567. [DOI] [PubMed] [Google Scholar]
- 12. Cha Y, Kim JH, Chung SA. Profile of pediatric ophthalmic referrals in a single emergency department in Korea. Pediatr Emerg Med J. 2024;11:28–38. [Google Scholar]
- 13. Hutchinson AK, Morse CL, Hercinovic A, Cruz OA, Sprunger DT, Repka MX, et al. Pediatric eye evaluations preferred practice pattern. Ophthalmology. 2023;130:P222–P270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Prentiss KA, Dorfman DH. Pediatric ophthalmology in the emergency department. Emerg Med Clin North Am. 2008;26:181–198. [DOI] [PubMed] [Google Scholar]
- 15. Korean Ministry of Education . Press release: announcement of in‐person classes for schools at each level vol. 2024 Ministry of Education. 2020. [Cited 2024 Jan 31]. Available from: https://www.moe.go.kr/boardCnts/viewRenew.do?boardID=294&boardSeq=80510&lev=0&searchType=null&statusYN=W&page=41&s=moe&m=020402&opType=N.(in Korean).
- 16. Hale T. Oxford COVID‐19 government response traker. 2024. Oxford (United Kingdom): Blavatnik School of Government. [Cited 2024 Jan 22]. Available from: http://www.bsg.ox.ac.uk/covidtracker
- 17. Lim T, Park J, Je S. Pediatric Korean triage and acuity scale. Pediatr Emerg Med J. 2015;2:53–58. (in Korean). [Google Scholar]
- 18. Channa R, Zafar SN, Canner JK, Haring RS, Schneider EB, Friedman DS. Epidemiology of eye‐related emergency department visits. JAMA Ophthalmol. 2016;134:312–319. [DOI] [PubMed] [Google Scholar]
- 19. Halawa OA, Friedman DS, Roldan AM, Zebardast N. Changing trends in ocular trauma during the COVID‐19 pandemic in the USA. Br J Ophthalmol. 2023;107:295–298. [DOI] [PubMed] [Google Scholar]
- 20. Kuitunen I, Artama M, Mäkelä L, Backman K, Heiskanen‐Kosma T, Renko M. Effect of social distancing due to the COVID‐19 pandemic on the incidence of viral respiratory tract infections in children in Finland during early 2020. Pediatr Infect Dis J. 2020;39:e423–e427. [DOI] [PubMed] [Google Scholar]
- 21. Hatoun J, Correa ET, Donahue SMA, Vernacchio L. Social distancing for COVID‐19 and diagnoses of other infectious diseases in children. Pediatrics. 2020;146:e2020006460. [DOI] [PubMed] [Google Scholar]
- 22. Ahn JG. Epidemiological changes in infectious diseases during the coronavirus disease 2019 pandemic in Korea: a systematic review. Clin Exp Pediatr. 2022;65:167–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Kim JY, Jung JU, Jo YC, Park MH, Kim KY, Kim HK. Impact of social distancing and personal hygiene on the prevalence of epidemic keratoconjunctivitis during the COVID‐19 pandemic. J Korean Ophthalmol Soc. 2022;63:126–133. (in Korean). [Google Scholar]
- 24. Yu H, Xu M, Zhao Y, Li J, Wu W, Feng Y. COVID‐19 changed prevalence, disease spectrum and management strategies of ocular trauma. Front Med (Lausanne). 2021;8:774493. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Cavuoto KM, Vanner EA, Osigian CJ. Trends in pediatric ocular trauma presenting to an ophthalmology‐specific emergency department during the COVID‐19 pandemic. J AAPOS. 2021;25:170–172. [DOI] [PubMed] [Google Scholar]
- 26. Claudet I, Marchand‐Tonel C, Ricco L, Houzé‐Cerfon C‐H, Lang T, Bréhin C. During the COVID‐19 quarantine, home has been more harmful than the virus for children! Pediatr Emerg Care. 2020;36:e538–e540. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1.
Table S2.