Article Text
PDF
Abstract
Objectives Amblyopia is the most common cause of unilateral visual impairment in children and requires long-term treatment. This study aimed to quantify the impact of pandemic control measures on amblyopia management.
Design and setting This was a retrospective cohort study of data from a large amblyopia management database at a major tertiary eye care centre in China.
Participants Outpatients with amblyopia who visited the hospital from 1 June 2019, through 28 February 2022.
Primary and secondary outcome measures The primary outcome was the number of first and follow-up in-person visits to the hospital for amblyopia treatment. Secondary outcomes included the time interval between consecutive visits and improvement of vision (visual acuity, contrast sensitivity and stereopsis). Patient records were grouped into prepandemic and during pandemic periods.
Results A total of 10 060 face-to-face visits for 5361 patients (median age 6.7 years, IQR 5.4, 8.9) that spanned two lockdown periods were included in the analysis, of which 28% were follow-up visits. Pandemic control measures caused a sharp decline in the number of outpatient visits (3% and 30% of prepandemic levels in the months directly after the start of the first (2020) and second (2021) periods of pandemic control measures, respectively). However, these drops were followed by pronounced rebounds in visits that exceeded prepandemic levels by 51.1% and 108.5%, respectively. The interval between consecutive visits increased significantly during the pandemic from a median (IQR) of 120 (112, 127) days in 2019 to 197 (179, 224) in 2020 (p<0.001) and 189 (182, 221) in 2021 (p<0.001). There were no significant differences in the improvement of visual function or treatment compliance between the prepandemic and postpandemic groups.
Conclusions The number of amblyopia patient hospital visits spiked well above prepandemic levels following lockdown periods. This pattern of patient behaviour can inform planning for amblyopia treatment services during and after public health-related disruptions.
- COVID-19
- paediatric ophthalmology
- public health
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
Use of a large real-world sample of 10 060 records.
The observation period covered 33 months and spanned two lockdown periods.
Amblyopia patient records contained measures of visual acuity, contrast sensitivity and stereoacuity when enabled an assessment of treatment outcomes before and after the epidemic.
The study involved only one centre and was retrospective.
Limited prepandemic data were available because the database we used for the study was established in 2019.
Introduction
Pandemic control measures implemented as a result of COVID-19, such as stay-at-home orders and cancellation of non-urgent medical services, caused unprecedented disruption to healthcare systems worldwide.1–3 These effects were compounded by the reluctance of patients to attend healthcare centres when COVID-19 infection rates were high. Previous studies have reported delayed treatment seeking and prolonged duration of symptoms during epidemics in patients with diseases such as myocardial infarction, acute cerebrovascular disease and retinal detachment.4 5 In addition, stay-at-home orders may have accelerated the progression of myopia in children and reduced paediatric ophthalmology and optometry clinic visits by over 80%.4 6–12
Amblyopia is a neurodevelopmental disorder of the visual system. With a prevalence varying from 1% to 6% across geographical regions,13–16 amblyopia is the most common cause of unilateral vision loss in children.17 18 Amblyopia treatment involves refractive correction and patching of the stronger non-amblyopic fellow eye. These treatments are most effective when delivered in early childhood when the visual system is developing19–22 and visual cortex neuroplasticity is high. Therefore, clinical management of amblyopia is particularly sensitive to disruptions or delays which may compromise long-term treatment outcomes.
To investigate the impact of the pandemic on amblyopia management, we retrospectively analysed records from a large amblyopia patient database held at a major tertiary eye hospital in China. We compared visit numbers and amblyopia treatment outcomes from 1 June 2019, through 28 February 2022, a period that encompassed prepandemic baseline data and data from two large-scale lockdowns that occurred in February 2020 and May 2021. Our analysis afforded a unique opportunity to examine the real-world effects of the pandemic on amblyopia management.
Methods
Study design and sampling
This investigation was a retrospective study of outpatients with amblyopia that used the Uniting Functions in Ophthalmology and Optometry (UFOs) database23 24 administered by Zhongshan Ophthalmic Centre (ZOC) of Sun Yat-sen University, a tertiary ophthalmic centre in Guangzhou, China. We included all consecutive, deidentified amblyopia record data on face-to-face patient visits from 1 June 2019 to 28 February 2022 grouped by month. From 1 February 2020, COVID-19 prevention protocols were initiated in Guangzhou.
We analysed the number of patient first visits, the number of follow-up visits, follow-up intervals and vision improvement with treatment. We also analysed treatment adherence data and the geographical spread of patient visits. For first and follow-up visits, we compared the number of visits per month over 33 months from 1 June 2019, through 28 February 2022. We also calculated the number of patient visits from 1 December 2019, through 31 December 2019, as a baseline comparison. December was chosen as the baseline because the number of working days in December is closer to the average, avoiding months with long public holidays and student vacation periods. Follow-up intervals were calculated as the time from a patient’s follow-up visit to their immediately preceding visit. The follow-up interval for a month was the mean of the intervals for all of the follow-up visits in that month.
To compare amblyopia treatment efficacy prepandemic and during the pandemic, we identified two 6-month periods in our data; a pre-COVID interval (1 August 2019 to 31 January 2020) and a post-COVID interval (1 February 2020 to 31 July 2020). Participants who had both a first visit and a follow-up visit within one of these intervals were included in the analysis. We compared clinical characteristics, baseline visual function and the visual function changes of the two groups. Clinical characteristics included sex, age, type of amblyopia, treatment method, treatment compliance and geographical distribution. Baseline visual function included first visit best-corrected visual acuity (BCVA), the area under the log contrast sensitivity function (AULCSF) and distance stereoacuity. Visual function improvement included the change in BCVA, AULCSF for the amblyopic eye (AE) and distance stereoacuity. Test details are provided in online supplemental material.
Supplemental material
Statistical analysis
IBM SPSS software V.23 (IBM Corporation, Armonk, NY) was used for statistical analysis. Decimal VAs were converted to the logarithm of the minimum angle of resolution (logMAR) scale for all statistical analyses. We used the Shapiro-Wilk test to evaluate the distribution of each sample. Median and IQR were used for the description of continuous variables that did not conform to a normal distribution. Frequency and proportion were used to describe categorical variables, and the χ2 test was used for comparison between groups. A p value of less than 0.05 was considered statistically significant for all tests.
Patient and public involvement
Patients or the public were not involved in the design, conduct or reporting, or dissemination plans of our research.
Results
Sample characteristics
From 1 June 2019, through 28 February 2022, 10 060 face-to-face visits records from 5361 amblyopia patients were recorded in the database, of which 28% were follow-up patients. The median (IQR) age of the patients was 6.7 (5.4, 8.9) years and 3059 (57.1%) were boys.
Overall hospital visit frequency
Figure 1 shows the number of first and follow-up in-person hospital visits by amblyopia patients during the study period. All specialist outpatient clinics at ZOC were closed on 1 February 2020 and the ZOC public number advised amblyopia patients to postpone their treatment. A small number of specialist outpatient appointment resumed on 15 February 2020 and normal hospital operations were resumed on 16 March 2020. The pandemic control measures caused a 97% decrease in amblyopia patient visits in February 2020 compared with the pre-epidemic baseline level in 2019 (259 visits per month). After the hospital opened, visit numbers increased and in July 2020 exceeded the prepandemic baseline level (16% more). Visit numbers peaked in August 2020 (88% more visits than the prepandemic baseline level). The timing of the peak coincided with the school summer holidays, which is a popular time for amblyopia patient hospital visits. However, the pandemic control measures amplified this annual peak with the number of visits in August 2020 being 35% higher than the number of prepandemic visits in August 2019.
Number of amblyopia patient visits per month.
In late May 2021, a second wave of the pandemic occurred in Guangzhou. Specialist outpatient clinics were not closed, but patient visits still dropped sharply. However, the number of visits picked up in July and reached a second peak in August 2021, which surpassed the first peak in August 2020 with an increase of 55% from the previous year. An increased number of visits compared with prepandemic levels characterised much of 2021 and 2022, where the mean number of visits were 46.6% and 78.8% above those in December 2019, respectively. We also observed a significantly lower proportion of follow-up visits than first visits, with follow-up visits averaging 29% of the total number of visits. Trends in visit numbers differed for males versus females, with males having a greater percentage of overall visits and per-month visits. This appears to be unrelated to the pandemic, as the number of male patients was higher than female patients at all times prepandemic, during pandemic and postpandemic.
Follow-up intervals
We excluded months with less than 100 total visits (February 2020, March 2020 and June 2021) from the follow-up interval analysis because the number of follow-up visits in these months was too low to reliably estimate follow-up rates (blank area in figure 2). The median (IQR) follow-up visit interval in 2019 before the pandemic was 120 (112, 127) days. This reflects the usual follow-up visit interval of 2–3 months for amblyopia treatment. The interval gradually increased after the start of the pandemic until it peaked in August 2020 (figure 2). After August, the interval between follow-ups fell slightly but remained longer than prepandemic levels.
Follow-up visit intervals from 1 June 2019 to 28 February 2022. The shaded area represents IQR.
The median (IQR) follow-up interval was 197 (IQR, 179, 224) days in 2020 and 189 (IQR, 182, 221) days in 2021, both well above the interval in 2019 (197 (179, 224) vs 120 (112, 127); p<0.001; 189 (182, 221) vs 120 (112, 127); p<0.001). There was no statistically significant difference between the 2020 and 2021 follow-up intervals (197 (179, 224) vs 189 (182, 221); p=0.94). We further compared the frequency of follow-up visits in 6-month windows before and after 1 February 2020 (the start of pandemic control measures). We observed a statistically significant difference between 1.19±0.61 visits per capita pre the epidemic and 1.09±0.36 per capita during the epidemic (1.19±0.61 vs 1.09±0.36; p=0.009).
Clinical and visual function outcomes between pre-epidemic and during-epidemic cohorts
A total of 130 patients met the criteria for inclusion in this analysis. The dataset included 102 eyes from 61 patients in the pre-epidemic cohort and 107 eyes from 69 patients in the during-epidemic cohort. The average age in the precohort was 7.50±4.24 years and 7.86±2.71 years in the during-cohort. There were no statistically significant differences between the demographics of the two groups or the type of amblyopia (table 1). Visual function measurement included visual acuity, contrast sensitivity and stereopsis. These tests were performed at each patient’s visit as a routine procedure and the results were stored in the UFOs database.23–26 We compared the improvement in visual acuity and contrast sensitivity between the two groups after 6-month follow-up intervals. The improvement was defined as the value at the follow-up visit minus the value at the initial visit. Stereopsis was compared between the two groups as the percentage of patients with improved stereopsis. We observed that patients in the during cohort showed better BCVA, AULCSF and stereoacuity at their initial visits compared with the pre-epidemic cohort (BCVA: 0.20±0.24 vs 0.31±0.30; p=0.004; AULCSF: 1.32±0.37 vs 1.14±0.43; p=0.017; stereoacuity: 5000 (400, 5000) vs 5000 (5000, 5000); p=0.016).
Characteristics of precohort and during-cohort at 6-month follow-up intervals
After a 6-month follow-up, there was no significant group difference in the improvement of visual function including AE BCVA, AULCSF and stereoacuity (BCVA: −0.04 (−0.1, 0) vs −0.04 (−0.1, 0); p=0.41; AULCSF: 0.1 (−0.02, 0.23) vs 0.08 (−0.13, 0.27); p=0.76; stereoacuity: 16.4% vs 22.4%; p=0.41). No patient experienced reduced visual acuity of more than two logMAR lines in either group.
To understand the impact of pandemic control measures on patients’ behaviour, we compared patching treatment adherence and the geographic distribution of visiting patients’ homes. There was no difference in treatment adherence between the two groups measured using self-report questionnaires (75±19 score vs 71±16 score; p=0.37). Patients were mainly from Guangzhou city, followed by other urban areas in Guangdong Province, and finally, from other provinces. There was no statistical difference in the geographical distribution of visits before and during the pandemic (table 1).
Discussion
The COVID-19 pandemic impacted patients and healthcare providers, however, the effect of the pandemic on amblyopia management has not yet been reported, although it has attracted much attention.8 27 We investigated the effect of the pandemic on 10 060 amblyopia patient visits with records stored in a large database (the UFOs database) across 33 months in a tertiary ophthalmology centre in China. We found a sharp reduction of face-to-face visits in the short term after each pandemic wave; however, there was a rebound of visits in the long term. Follow-up periods were significantly longer, and follow-up frequency was reduced after the start of the pandemic. Patients in the during-epidemic cohort maintained stable visual function at 6-month follow-up time intervals, with no difference in vision improvement compared with patients before the epidemic.
During periods of lockdown, patient visit numbers were affected by hospital closures, public health messaging asking people to stay at home, and parents were likely to have been worried about taking their children to hospitals when infection rates were high.6 12 A key feature of our data is a substantial increase in visits following lockdown periods. Many studies on the COVID-19 pandemic have suggested a possible rebound effect for visit numbers after an outbreak,28 but this has not previously been confirmed due to short observation periods.
Interestingly, we also found a higher outpatient volume for male patients than for females. This seems to be a constant characteristic during the 33-month observation period, although there is no evidence for a significantly higher prevalence of amblyopia in males than in females. The cause of this effect is unknown and may be specific to the population being studied.
Many efforts were taken during the pandemic to balance infection control and the supply of ophthalmology services at ZOC. ZOC established a virtual clinical service using several digital technologies to deliver online ophthalmic diagnosis and treatment services. In Wu et al’s report,29 which analysed the use of telehealth services in ZOC before and after the epidemic, retinopathy was the most common reason for a virtual consultation (22.6%), followed by ocular surface diseases (24.2%) and glaucoma (19.4%). Refractive problems were the most common reason for face-to-face visits before the epidemic (26.1%). However, Wu et al did not analyse the subspecialist online data for amblyopia patients in detail. Patients with refractive problems that include amblyopia appear to be less likely to have a virtual consultation because essential examination elements such as visual acuity assessment can be assessed only in person. This may be why the families of amblyopia patients prefer to come to the hospital for a traditional consultation in person rather than using telemedicine. In this scenario, promoting the development of digital clinical tools, such as home self-assessments and amblyopia specialist doctor–patient interaction platforms may provide longer-term solutions to ophthalmology service disruptions. In fact, many digital health solutions were deployed during the pandemic, including artificial intelligence-enabled care, digital home monitoring, and more, in addition to virtual consultations similar to those used at ZOC.30–33
The UFOs database contains longitudinal data from adult and paediatric patients including those with amblyopia, strabismus, abnormal binocular vision, myopia and low vision, with paediatric amblyopia patients being the largest group. Amblyopia patient records contain basic clinical data, medical history, refractive error, visual acuity, contrast sensitivity, stereoacuity and demographic information. In addition, some records include electroencephalogram and visual evoked potentials measures, psychophysical testing results and vision training outcomes. These longitudinal data allow for a multidimensional assessment of changes in a patient’s visual function. Using these data, we compared visual function between prepandemic and during-pandemic patient groups. We found similar improvements in BCVA, AULCSF and stereopsis in both groups within a 6-month follow-up interval. Visual acuity gains were small in both groups during our observation period, which may be because most patients included in the analysis had mild amblyopia. In addition, there was no difference in treatment adherence between the two groups. In terms of geographical distribution, face-to-face visits during the epidemic were geographically concentrated around ZOC clinics from Guangzhou city (65% of Guangdong Province); this is in line with Wu et al.29 However, Wu et al found that the pandemic was associated with fewer hospital visits from outside Guangzhou compared with prepandemic visits, whereas our analysis revealed no change in the geographic distribution of patients. This may be because we restricted our analysis to patients diagnosed with amblyopia.
Limitations
Limitations include data being from a single centre in China and limited availability of prepandemic data because the database we used for the study was established recently.
Conclusion
Pandemic control measures that caused a sharp decline in amblyopia clinic visits were followed by a rebound in visits, with sustained visit volumes exceeding prepandemic levels by 50% to over 100%. Paediatric ophthalmology services should plan to allocate extra resources following public health measures that temporarily restrict access to in-person clinical appointments for amblyopia management.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
The study was approved by the Ethical Committee of ZOC and adhered to the principles of the Declaration of Helsinki. The Zhongshan Ophthalmic Center (ZOC) institutional review board waived the need for written informed consent from participants since the study required no direct contact with human subjects (no interview or sample collection) and only used amblyopia database records, and the personal information has been recorded anonymously where subjects cannot be identified directly or indirectly.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
JL, QY and CX are joint first authors.
JL, QY and CX contributed equally.
Contributors JLiu, QY and CX are joint first authors. JLiu, BT and JLi conceptualised and designed the study. JLi created the database and provided the raw data. JLiu, CX contributed to the data analysis and interpretation. QY contributed to the data acquisition. YZhuang, LF, YH, ZX, YZhou, XC, YY, RJ, YP, WY, YW and JY assisted with data collection. JLiu drew and wrote the first draft of the article. BT contributed critical feedback and edits to article drafts. BT and JLi are responsible for the overall content as the guarantor. All authors read and approved the final manuscript.
Funding This work was supported by the National Key Research & Development Project number 2020YF2003905; InnoHK and the Hong Kong SAR Government number N/A.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.