Application of big-data for epidemiological studies of refractive error

Myopia progression patterns among paediatric patients in a clinical setting

PURPOSE

This retrospective analysis of electronic medical record (EMR) data investigated the natural history of myopic progression in children from optometric practices in Ireland.

METHODS

The analysis was of myopic patients aged 7-17 with multiple visits and not prescribed myopia control treatment. Sex- and age-specific population centiles for annual myopic progression were derived by fitting a weighted cubic spline to empirical quantiles. These were compared to progression rates derived from control group data obtained from 17 randomised clinical trials (RCTs) for myopia. Linear mixed models (LMMs) were used to allow comparison of myopia progression rates against outputs from a predictive online calculator. Survival analysis was performed to determine the intervals at which a significant level of myopic progression was predicted to occur.

RESULTS

Myopia progression was highest in children aged 7 years (median: -0.67 D/year) and progressively slowed with increasing age (median: -0.18 D/year at age 17). Female sex (p < 0.001), a more myopic SER at baseline (p < 0.001) and younger age (p < 0.001) were all found to be predictive of faster myopic progression. Every RCT exhibited a mean progression higher than the median centile observed in the EMR data, while clinic-based studies more closely matched the median progression rates. The LMM predicted faster myopia progression for patients with higher baseline myopia levels, in keeping with previous studies, which was in contrast to an online calculator that predicted slower myopia progression for patients with higher baseline myopia. Survival analysis indicated that at a recall period of 12 months, myopia will have progressed in between 10% and 70% of children, depending upon age.

CONCLUSIONS

This study produced progression centiles of untreated myopic children, helping to define the natural history of untreated myopia. This will enable clinicians to better predict both refractive outcomes without treatment and monitor treatment efficacy, particularly in the absence of axial length data.

Retrospective diagnosis of naked eye visual acuity (UCVA) variations in patients with refractive errors treated with SMILE, LASIK, and WF-LASIK refractive surgery

A retrospective assessment of the visual acuity (UCVA) variations in bare eyes of the refractive error cases treated with SMILE, LASIK and WF-LASIK. A retrospective selection of 126 patients with refractive error treated by refractive surgery admitted to our hospital between January 2019 and December 2021 were divided into three separate sets of patients according to their surgical methods: the SMILE cohort, the LASIK cohort, and the WF-LASIK cohort, and the three sets of patients were analyzed for bare eye visual acuity, refraction, higher-order aberration, BUT, SIt index, and complications, and the recovery effects of patients with the three surgical procedures. All three types of refractive surgery, SMILE, LASIK and WF-LASIK, can yield good surgical results in the reduction of refractive error, and patients with SMILE have better postoperative tear film stability, while patients with WF-LASIK have the best postoperative visual quality.

Application of big data in ophthalmology

The advents of information technologies have led to the creation of ever-larger datasets. Also known as big data, these large datasets are characterized by its volume, variety, velocity, veracity, and value. More importantly, big data has the potential to expand traditional research capabilities, inform clinical practice based on real-world data, and improve the health system and service delivery. This review first identified the different sources of big data in ophthalmology, including electronic medical records, data registries, research consortia, administrative databases, and biobanks. Then, we provided an in-depth look at how big data analytics have been applied in ophthalmology for disease surveillance, and evaluation on disease associations, detection, management, and prognostication. Finally, we discussed the challenges involved in big data analytics, such as data suitability and quality, data security, and analytical methodologies.

Using electronic medical record data to establish and monitor the distribution of refractive errors. JOURNAL OF OPTOMETRY 2022;15 Suppl 1:S32-S42. [PMID: 36220741 PMCID: PMC9732486 DOI: 10.1016/j.optom.2022.09.001]

OBJECTIVE

To establish the baseline distribution of refractive errors and associated factors amongst a population that attended primary care optometry clinics.

DESIGN

Retrospective cross sectional cohort study of electronic medical records (EMR).

METHODS

Electronic medical record data was extracted from forty optometry clinics, representing a mix of urban and rural areas in Ireland. The analysis was confined to demographic and clinical data gathered over a sixty-month period between 2015 and 2019. Distribution rates were calculated using the absolute and relative frequencies of refractive error in the dataset, stratified for age and gender using the following definitions: high myopia ≤ -6.00 D, myopia ≤ -0.50 D, hyperopia ≥ +0.50 D, astigmatism ≤ -0.75 DC and anisometropia ≥ 1.00 D. Visual acuity data was used to explore vision impairment rates in the population. Further analysis was carried out on a gender and age-adjusted subset of the EMR data, to match the proportion of patients in each age grouping to the population distribution in the most recent (2016) Irish census.

RESULTS

153,598 clinic records were eligible for analysis. Refractive errors ranged from -26.00 to +18.50 D. Myopia was present in 32.7%, of which high myopia represented 2.4%, hyperopia in 40.1%, astigmatism in 38.3% and anisometropia in 13.4% of participants. The clinic distribution of hyperopia, astigmatism and anisometropia peaked in older age groups, whilst the myopia burden was highest amongst people in their twenties. A higher proportion of females were myopic, whilst a higher proportion of males were hyperopic and astigmatic. Vision impairment (LogMAR > 0.3) was present in 2.4% of participants. In the gender and age- adjusted distribution model, myopia was the most common refractive state, affecting 38.8% of patients.

CONCLUSION

Although EMR data is not representative of the population as a whole, it is likely to provide a reasonable representation of the distribution of clinically significant (symptomatic) refractive errors. In the absence of any ongoing traditional epidemiological studies of refractive error in Ireland, this study establishes, for the first time, the distribution of refractive errors observed in clinical practice settings. This will serve as a baseline for future temporal trend analysis of the changing pattern of the distribution of refractive error in EMR data. This methodology could be deployed as a useful epidemiological resource in similar settings where primary eyecare coverage for the management of refractive error is well established.

The Refractive Error and Vision Impairment Estimation with Spectacle Data Study

Purpose

To investigate whether spectacle lens sales data can be used to estimate the population distribution of refractive error among patients with ametropia and hence to estimate the current and future risk of vision impairment.

Design

Cross-sectional study.

Participants

A total of 141 547 436 spectacle lens sales records from an international European lens manufacturer between 1998 and 2016.

Methods

Anonymized patient spectacle lens sales data, including refractive error information, was provided by a major European spectacle lens manufacturer. Data from the Gutenberg Health Survey was digitized to allow comparison of a representative, population-based sample with the spectacle lens sales data. A bootstrap analysis was completed to assess the comparability of both datasets. The expected level of vision impairment resulting from myopia at 75 years of age was calculated for both datasets using a previously published risk estimation equation combined with a saturation function.

Main Outcome Measures

Comparability of spectacle lens sales data on refractive error with typical population surveys of refractive error and its potential usefulness to predict vision impairment resulting from refractive error.

Results

Equivalent estimates of the population distribution of spherical equivalent refraction can be provided from spectacle lens data within limits. For myopia, the population distribution was equivalent to the Gutenberg Health Survey (≤ 5% deviation) for levels of –2.0 diopters (D) or less, whereas for hyperopia, the distribution was equivalent (≤ 5% deviation) for levels of +3.0 D or more. The estimated rates of vision impairment resulting from myopia were not statistically significantly different (chi-square, 182; degrees of freedom, 169; P = 0.234) between the spectacle lens dataset and Gutenberg Health Survey dataset.

Conclusions

The distribution of refractive error and hence the risk of vision impairment resulting from refractive error within a population can be determined using spectacle lens sales data. Pooling this type of data from multiple industry sources could provide a cost-effective, timely, and globally representative mechanism for monitoring the evolving epidemiologic features of refractive error and associated vision impairment.

Big Data analysis of vision screening standards used to evaluate fitness to drive

PURPOSE

Visual acuity assessment is the most commonly performed vision screening method for drivers. The standards and repeat assessment intervals used, however, are arbitrary, lack an evidence base and are highly variable across different countries. This study utilises the power of Big Data to provide evidence-based recommendations for standardised driver vision screening.

METHODS

Anonymised electronic medical record data was gathered from 40 Irish optometry practices comprising 81,184 unique patients. A Kaplan-Meier Survival (KMS) analysis was used to determine the effect of increasing age and time since screening on the likelihood of passing the visual acuity standard for driving. A logistic function was fit to assess the effect of varying the minimum visual acuity standard required to drive on the screening pass rate within the population.

RESULTS

The likelihood of failing repeat screening increased as a function of time since initial screening for all age groups (χ²=1447, df =6, p < 0.001), with older patients most affected. Rescreening intervals for individuals who initially met the vision standard unaided reduced as a function of age. Using an 80% survivability threshold, intervals ranged from every eight years for drivers under 50, reducing to every two years for those aged over 80. Rescreening intervals for drivers requiring optical correction to meet the standard, also decreased with age. Approximately 1% of individuals are excluded from driving using a 0.3 logMAR visual acuity standard with correction.

CONCLUSION

Visual acuity-based screening should take place at regular intervals for all drivers, not just those over 70. Re-screening intervals should be based on age, with shorter intervals for older drivers due to the combined effect of age and time on the likelihood of passing the driving visual acuity standards. The most commonly used standard of 0.3 logMAR results in a minimal number of potential drivers being excluded from driving.