Causes of Childhood Blindness in the United States Using the IRIS® Registry (Intelligent Research in Sight).

PURPOSE: To investigate causes of childhood blindness in the United States using the IRIS® Registry (Intelligent Research in Sight). DESIGN: Cross-sectional study. PARTICIPANTS: Patients ≤ 18 years of age with visual acuity (VA) 20/200 or worse in their better-seeing eye in the IRIS Registry during 2018. METHODS: Causes of blindness were classified by anatomic site and specific diagnoses. MAIN OUTCOME MEASURES: Percentages of causes of blindness. RESULTS: Of 81 164 children with 2018 VA data in the IRIS Registry, 961 (1.18%) had VA 20/200 or worse in their better-seeing eye. Leading causes of blindness were retinopathy of prematurity (ROP) in 301 patients (31.3%), nystagmus in 78 patients (8.1%), and cataract in 64 patients (6.7%). The retina was the leading anatomic site (47.7%) followed by optic nerve (11.6%) and lens (10.0%). A total of 52.4% of patients had treatable causes of blindness. CONCLUSIONS: This analysis offers a unique cross-sectional view of childhood blindness in the United States using a clinical data registry. More than one-half of blind patients had a treatable cause of blindness. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

United States Population Disparities in Ophthalmic Care: Blindness and Visual Impairment in the IRIS® Registry (Intelligent Research in Sight).

PURPOSE: To evaluate associations of patient characteristics with United States eye care use and likelihood of blindness. DESIGN: Retrospective observational study. PARTICIPANTS: Patients (19 546 016) with 2018 visual acuity (VA) records in the American Academy of Ophthalmology's IRIS® Registry (Intelligent Research in Sight). METHODS: Legal blindness (20/200 or worse) and visual impairment (VI; worse than 20/40) were identified from corrected distance acuity in the better-seeing eye and stratified by patient characteristics. Multivariable logistic regressions evaluated associations with blindness and VI. Blindness was mapped by state and compared with population characteristics. Eye care use was analyzed by comparing population demographics with United States Census estimates and proportional demographic representation among blind patients versus a nationally representative US population sample (National Health and Nutritional Examination Survey [NHANES]). MAIN OUTCOME MEASURES: Prevalence and odds ratios for VI and blindness; proportional representation in the IRIS® Registry, Census, and NHANES by patient demographics. RESULTS: Visual impairment was present in 6.98% (n = 1 364 935) and blindness in 0.98% (n = 190 817) of IRIS patients. Adjusted odds of blindness were highest among patients ≥ 85 years old (odds ratio [OR], 11.85; 95% confidence interval [CI], 10.33-13.59 vs. those 0-17 years old). Blindness also was associated positively with rural location and Medicaid, Medicare, or no insurance vs. commercial insurance. Hispanic (OR, 1.59; 95% CI, 1.46-1.74) and Black (OR, 1.73; 95% CI, 1.63-1.84) patients showed a higher odds of blindness versus White non-Hispanic patients. Proportional representation in IRIS Registry relative to the Census was higher for White than Hispanic (2- to 4-fold) or Black (11%-85%) patients (P < 0.001). Blindness overall was less prevalent in NHANES than IRIS Registry; however, prevalence in adults aged 60+ was lowest among Black participants in the NHANES (0.54%) and second highest among comparable Black adults in IRIS (1.57%). CONCLUSIONS: Legal blindness from low VA was present in 0.98% of IRIS patients and associated with rural location, public or no insurance, and older age. Compared with US Census estimates, minorities may be underrepresented among ophthalmology patients, and compared with NHANES population estimates, Black individuals may be overrepresented among blind IRIS Registry patients. These findings provide a snapshot of US ophthalmic care and highlight the need for initiatives to address disparities in use and blindness. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Identifying Strategies to Reduce Low-Value Preoperative Testing for Low-Risk Procedures: a Qualitative Study of Facilities with High or Recently Improved Levels of Testing.

BACKGROUND: Healthcare agencies and perioperative professional organizations recommend avoiding preoperative screening tests for low-risk surgical procedures. However, low-value preoperative tests are still commonly ordered even for generally healthy patients and active strategies to reduce this testing have not been adequately described. OBJECTIVE: We sought to learn from hospitals with either high levels of testing or that had recently reduced use of low-value screening tests (aka "delta sites") about reasons for testing and active deimplementation strategies they used to effectively improve practice. DESIGN: Qualitative study of semi-structured telephone interviews. PARTICIPANTS: We identified facilities in the US Veterans Health Administration (VHA) with high or recently improved burden of potentially low-value preoperative testing for carpal tunnel release and cataract surgery. We recruited perioperative clinicians to participate. APPROACH: Questions focused on reasons to order preoperative screening tests for patients undergoing low-risk surgery and, more importantly, what strategies had been successfully used to reduce testing. A framework method was used to identify common improvement strategies and specific care delivery innovations. KEY RESULTS: Thirty-five perioperative clinicians (e.g., hand surgeons, ophthalmologists, anesthesiologists, primary care providers, directors of preoperative clinics, nurses) from 29 VHA facilities participated. Facilities that successfully reduced the burden of low-value testing shared many improvement strategies (e.g., building consensus among stakeholders; using evidence/norm-based education and persuasion; clarifying responsibility for ordering tests) to implement different care delivery innovations (e.g., pre-screening to decide if a preop clinic evaluation is necessary; establishing a dedicated preop clinic for low-risk procedures). CONCLUSIONS: We identified a menu of common improvement strategies and specific care delivery innovations that might be helpful for institutions trying to design their own quality improvement programs to reduce low-value preoperative testing given their unique structure, resources, and constraints.

Endophthalmitis Rate in Immediately Sequential versus Delayed Sequential Bilateral Cataract Surgery within the Intelligent Research in Sight (IRIS®) Registry Data.

PURPOSE: To compare the rate of postoperative endophthalmitis after immediately sequential bilateral cataract surgery (ISBCS) versus delayed sequential bilateral cataract surgery (DSBCS) using the American Academy of Ophthalmology Intelligent Research in Sight (IRIS®) Registry database. DESIGN: Retrospective cohort study. PARTICIPANTS: Patients in the IRIS Registry who underwent cataract surgery from 2013 through 2018. METHODS: Patients who underwent cataract surgery were divided into 2 groups: (1) ISBCS and (2) DSBCS (second-eye surgery ≥1 day after the first-eye surgery) or unilateral surgery. Postoperative endophthalmitis was defined as endophthalmitis occurring within 4 weeks of surgery by International Classification of Diseases (ICD) code and ICD code with additional clinical criteria. MAIN OUTCOME MEASURES: Rate of postoperative endophthalmitis. RESULTS: Of 5 573 639 IRIS Registry patients who underwent cataract extraction, 165 609 underwent ISBCS, and 5 408 030 underwent DSBCS or unilateral surgery (3 695 440 DSBCS, 1 712 590 unilateral surgery only). A total of 3102 participants (0.056%) met study criteria of postoperative endophthalmitis with supporting clinical findings. The rates of endophthalmitis in either surgery eye between the 2 surgery groups were similar (0.059% in the ISBCS group vs. 0.056% in the DSBCS or unilateral group; P = 0.53). Although the incidence of endophthalmitis was slightly higher in the ISBCS group compared with the DSBCS or unilateral group, the odds ratio did not reach statistical significance (1.08; 95% confidence interval, 0.87-1.31; P = 0.47) after adjusting for age, sex, race, insurance status, and comorbid eye disease. Seven cases of bilateral endophthalmitis with supporting clinical data in the DSBCS group and no cases in the ISBCS group were identified. CONCLUSIONS: Risk of postoperative endophthalmitis was not statistically significantly different between patients who underwent ISBCS and DSBCS or unilateral cataract surgery.

The American Academy of Ophthalmology IRIS Registry (Intelligent Research In Sight): current and future state of big data analytics.

PURPOSE OF REVIEW: To describe the drivers, development, and current state of the American Academy of Ophthalmology IRIS Registry (Intelligent Research In Sight), and analytics involving deidentified aggregate IRIS Registry data. RECENT FINDINGS: The IRIS Registry has a core mission of quality improvement and reporting. In addition, analytic projects performed to date have included characterizing patient populations and diseases, incidence, and prevalence; clinical outcomes and complications; risk factors and effect modifiers; practice patterns; and trends over time. Pipeline projects include application of artificial intelligence and machine learning approaches for predictive modeling and analytics, disease mapping, detecting patterns and identifying cohorts, and optimizing treatment based on patient-specific characteristics. SUMMARY: The IRIS Registry is the nation's largest single specialty clinical registry, with unique data elements specific to ophthalmology. It offers a wealth of opportunities involving big data analytics, including traditional inferential statistics as well as machine learning and artificial intelligence approaches scalable on massive amounts of data.

Bidirectional Association between Visual Impairment and Dementia Among Older Adults in the United States Over Time.

PURPOSE: Although visual impairment (VI) has been associated with worse cognitive performance among older adults, the temporal relationship between the 2 remains subject to debate. Our objective was to investigate the longitudinal impact of VI on cognitive function and vice versa. DESIGN: Retrospective, time-to-event study. PARTICIPANTS: National Health and Aging Trends Study (NHATS) participants from 2011 to 2018 cycles. METHODS: A total of 10 676 participants aged 65 years and older were included. Cox proportional hazards regression models evaluated the impact of baseline VI on subsequent dementia and impact of baseline dementia on subsequent VI. Models were adjusted for potential confounding variables, including demographics, clinical comorbidities, and hearing and physical function limitations. MAIN OUTCOME MEASURES: Hazard ratio (HR) for incident dementia among participants with baseline self-reported VI and HR for incident VI among participants with baseline dementia. RESULTS: Of the 10 676 participants included in the analysis, approximately 40% were aged 65-74 years, 40% were aged 75-84 years, and the remaining 20% were aged 85 years and older. The majority were female (59%), and 68% self-identified as non-Hispanic White. Among participants with normal cognitive status at baseline, subsequent dementia was observed in 1753 (16%), and among participants with normal self-reported vision at baseline, subsequent VI was reported in 2371 (22%). In adjusted regression models, participants with baseline VI had higher likelihood of developing dementia over subsequent follow-up (HR, 2.3; 95% confidence interval [CI], 2.0-2.6; P < 0.001). Likewise, participants with baseline dementia had a higher likelihood of developing self-reported VI over time (HR, 2.5; 95% CI, 2.2-2.8; P < 0.001). CONCLUSIONS: Self-reported VI in the US Medicare population is associated with greater dementia likelihood over time, and dementia is similarly associated with greater VI likelihood over time. Associations are likely multifactorial and bidirectional and could be explained by intervening variables in the path from VI to dementia, or vice versa, or by common risk factors for pathological processes in both eyes and brain. These findings suggest the need for early identification of older adults with visual compromise and consideration of visual disability in the cognitively impaired.

Systemic Corticosteroid Use after Central Serous Chorioretinopathy Diagnosis.

PURPOSE: To analyze the frequency of systemic corticosteroid prescriptions before and after central serous chorioretinopathy (CSC) diagnosis. DESIGN: Retrospective claims-based analysis. PARTICIPANTS: A nationally representative sample of commercial insurance beneficiaries who received care between 2007 and 2015. METHODS: We limited the study population to beneficiaries with incident CSC diagnosed by an eye care provider, excluding those with other major ophthalmologic comorbidities. We developed a non-CSC comparison cohort matched to CSC patients by age, sex, general health (Charlson Comorbidity Index), and geographic region. We compared systemic corticosteroid prescriptions before and after CSC diagnosis and by diagnosing provider (optometrist vs. ophthalmologist) and evaluated likelihood of steroids treatment among CSC versus matched control patients using logistic and Cox proportional hazard regression models. MAIN OUTCOME MEASURES: Systemic corticosteroid prescription frequency among CSC patients within 12 months pre-diagnosis and at 6, 12, and 24 months post-diagnosis, median time to steroid initiation and discontinuation, and odds of receiving steroids post-diagnosis among CSC and control patients. RESULTS: We identified 3418 CSC patients. Nearly 39% (n = 1326) were prescribed systemic steroids at some point during the analysis period, versus 23% of controls (4033 of 17 178 patients). Over 12% of CSC patients (n = 430) within 1 year pre-diagnosis, and nearly 12% (n = 404) within 1 year post-diagnosis. Most patients who received steroids after diagnosis were steroid naive (n = 231). Among those receiving steroids, CSC patients demonstrated longer median time to first post-diagnosis steroid prescription (1.82 years vs. 0.50 years for non-CSC patients) and longer time to last steroid prescription (1.62 years vs. 0.35 years for non-CSC patients). Although CSC patients were significantly less likely to receive steroids within 6 months post-diagnosis compared with non-CSC patients (odds ratio, 0.72; 95% confidence interval, 0.59-0.89), they were significantly more likely to receive steroids by 2 years post-diagnosis. Prescribing patterns were similar for patients diagnosed by an ophthalmologist versus optometrist. CONCLUSIONS: Despite evidence showing that steroids contribute to CSC development, many patients continue to be prescribed systemic corticosteroids after CSC diagnosis. Our results suggest a need for greater communication and collaboration among providers to ensure that clinical practice reflects evidence-based recommendations.

The Relative Impact of Patient, Physician, and Geographic Factors on Variation in Primary Rhegmatogenous Retinal Detachment Management.

PURPOSE: To evaluate geographic variation and characterize the relative contributions of patient characteristics, physician practice, and geographic region on variation in primary rhegmatogenous retinal detachment (RRD) management. DESIGN: Retrospective claims-based analysis. PARTICIPANTS: Commercially insured patients with incident RRD diagnosed between 2008 and 2016 (12 779 patients). METHODS: We determined whether patients underwent primary RRD repair within 60 days of diagnosis and identified repair type. We characterized physicians using physician identifier variables and characterized geography by Combined Statistical Areas or Core-Based Statistical Area. We used multilevel mixed effects logistic regression models to evaluate patient-, physician-, and geographic-level variation in whether patients underwent RRD repair and used multilevel mixed effects multinomial models to characterize variation in repair type. For each model, we evaluated patient fixed effects and physician random effects nested within geographic random effects. We estimated intraclass correlation coefficients and variance partition coefficients, respectively, to compare relative contributions of patient, physician, and geography to overall variation. MAIN OUTCOME MEASURES: Odds ratios for RRD repair and variation estimates for patient, physician, and geography. RESULTS: Most incident RRD patients received treatment within 60 days post-diagnosis. Pars plana vitrectomy was most common (49%), followed by laser barricade (23%), scleral buckle and pneumatic retinopexy (both 11%), and cryotherapy (5%). Physician-level variation showed greater impact on receipt of any treatment than geographic-level variation (estimated variance coefficients of 1.09 and 0.32, respectively). Patient-level characteristics represented approximately 82% of overall variation in receipt of any repair, versus 16% from physician-level and 2% from geographic-level factors. Among RRD patients who underwent repair, estimated variance coefficients were 0.07 for geography and 3.37 for physician. Physician-level factors represented approximately 50% of total variation in repair type, followed by patient-level (49%), and geographic-level (1%) factors. CONCLUSIONS: Rhegmatogenous retinal detachment repair decisions are influenced by patient-level and physician-level factors, less so by geographic variation. Patient characteristics account for most of the variation in receipt of repair, and physician practice accounts for most of the variation in choice of procedure. These findings indicate a need for additional studies to understand drivers behind differences in care and clinical outcomes and to identify barriers in access to care.

Receipt of Eye Care Services among Medicare Beneficiaries with and without Dementia.

PURPOSE: To examine the relationship between dementia status and receipt of eye care among US Medicare beneficiaries. DESIGN: Retrospective, claims-based analysis. PARTICIPANTS: A 20% representative sample of Medicare beneficiaries who received care between January 1, 2006, and December 31, 2015. METHODS: Dementia was identified from diagnosis codes documented in a beneficiary's first 3 years of observed Medicare enrollment. Eye care visits were identified from provider specialty codes on each encounter claim. We used multivariable Cox proportional hazards regression models with time-varying covariates to compare the likelihood of receiving eye care between beneficiaries with and without dementia. All models were adjusted for potential confounders, including demographics, urban/rural residence, systemic health (Charlson Index), and ocular comorbidities. MAIN OUTCOME MEASURES: Hazard ratio (HR) and 95% confidence interval (CI) for (1) being seen by any eye care provider (ophthalmologist or optometrist); (2) being seen by an ophthalmologist specifically; and (3) receiving cataract surgery (among beneficiaries with ophthalmologist encounters). RESULTS: A total of 4 451 200 beneficiaries met inclusion criteria; 3 805 718 (85.5%) received eye care during the study period, and 391 556 (8.8%) had diagnosed dementia. Some 73.4% of beneficiaries diagnosed with dementia saw an eye care provider during the study period and 55.4% saw an ophthalmologist versus 86.7% and 74.0% of beneficiaries, respectively, without dementia diagnoses. Compared with those without dementia diagnoses, beneficiaries with diagnosed dementia had lower likelihood of seeing any eye care provider (adjusted HR, 0.69; 95% CI, 0.69-0.70) and were less likely to see an ophthalmologist (adjusted HR, 0.55; 95% CI, 0.55-0.55). Among the subset of beneficiaries who did see ophthalmologists, those with diagnosed dementia were also less likely to receive cataract surgery than beneficiaries without diagnosed dementia (HR, 0.62; 95% CI, 0.62-0.63) and less likely to receive a cataract diagnosis (18% vs. 82%). CONCLUSIONS: US Medicare beneficiaries diagnosed with dementia are less likely to receive eye care than those without diagnosed dementia. Depending on visual acuity and functional status, this may have implications for injury prevention, physical and cognitive function, and quality of life. Further work is needed to identify barriers to receiving eye care, determine eye care services and settings that provide greatest value to patients with dementia, and implement measures to improve access to appropriate eye care.

Endophthalmitis after Cataract Surgery in the United States: A Report from the Intelligent Research in Sight Registry, 2013-2017.

PURPOSE: To determine recent incidence and visual outcomes for acute-onset endophthalmitis after cataract surgery performed in the United States. DESIGN: Retrospective cohort study. PARTICIPANTS: United States cataract surgery patients, 2013-2017 (5 401 686 patients). METHODS: Cases of acute-onset postoperative endophthalmitis occurring within 30 days after cataract surgery were identified using diagnosis codes in the American Academy of Ophthalmology IRIS (Intelligent Research in Sight) Registry database, drawn from electronic health records in ophthalmology practices across the nation. Annual and aggregate 5-year incidences were determined for all cataract surgeries and specifically for standalone procedures versus those combined with other ophthalmic surgeries. Patient characteristics were compared. Mean and median visual acuity was determined at 1 month preoperative as well as 1 week, 1 month, and 3 months postoperative among patients with and without endophthalmitis. MAIN OUTCOME MEASURES: Incidence of acute-onset postoperative endophthalmitis after cataract surgery. RESULTS: A total of 8 542 838 eyes underwent cataract surgery, 3629 of which developed acute-onset endophthalmitis (0.04%; 95% confidence interval, 0.04%-0.04%). Endophthalmitis incidence was highest among patients aged 0 to 17 years (0.37% over 5 years), followed by patients aged 18 to 44 years (0.18% over 5 years; P < 0.0001). Endophthalmitis occurred 4 times more often after combined cases (cataract with other ophthalmic procedures) than after standalone cataract surgeries (0.20% vs. 0.04% of cases), and occurred in 0.35% of patients receiving anterior vitrectomy. Mean 3-month postoperative visual acuity was 20/100 (median, 20/50) among endophthalmitis patients, versus a mean of approximately 20/40 (median, 20/30) among patients without endophthalmitis. However, 4% of endophthalmitis patients still achieved 20/20 or better visual acuity, and 44% achieved 20/40 or better visual acuity at 3 months. CONCLUSIONS: Acute-onset endophthalmitis occurred in 0.04% of 8 542 838 cataract surgeries performed in the United States between 2013 and 2017. Risk factors may include younger age, cataract surgery combined with other ophthalmic surgeries, and anterior vitrectomy. Visual acuity outcomes vary; however, patients can recover excellent vision after surgery. Big data from clinical registries like the IRIS Registry has great potential for evaluating rare conditions such as endophthalmitis, including developing benchmarks, longer-term time trend investigation, and comprehensive analysis of risk factors and prophylaxis.

Big data requirements for artificial intelligence.

PURPOSE OF REVIEW: To summarize how big data and artificial intelligence technologies have evolved, their current state, and next steps to enable future generations of artificial intelligence for ophthalmology. RECENT FINDINGS: Big data in health care is ever increasing in volume and variety, enabled by the widespread adoption of electronic health records (EHRs) and standards for health data information exchange, such as Digital Imaging and Communications in Medicine and Fast Healthcare Interoperability Resources. Simultaneously, the development of powerful cloud-based storage and computing architectures supports a fertile environment for big data and artificial intelligence in health care. The high volume and velocity of imaging and structured data in ophthalmology and is one of the reasons why ophthalmology is at the forefront of artificial intelligence research. Still needed are consensus labeling conventions for performing supervised learning on big data, promotion of data sharing and reuse, standards for sharing artificial intelligence model architectures, and access to artificial intelligence models through open application program interfaces (APIs). SUMMARY: Future requirements for big data and artificial intelligence include fostering reproducible science, continuing open innovation, and supporting the clinical use of artificial intelligence by promoting standards for data labels, data sharing, artificial intelligence model architecture sharing, and accessible code and APIs.

Development and validation of a predictive model for American Society of Anesthesiologists Physical Status.

BACKGROUND: The American Society of Anesthesiologists Physical Status (ASA-PS) classification system was developed to categorize the fitness of patients before surgery. Increasingly, the ASA-PS has been applied to other uses including justification of inpatient admission. Our objectives were to develop and cross-validate a statistical model for predicting ASA-PS; and 2) assess the concurrent and predictive validity of the model by assessing associations between model-derived ASA-PS, observed ASA-PS, and a diverse set of 30-day outcomes. METHODS: Using the 2014 American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) Participant Use Data File, we developed and internally cross-validated multinomial regression models to predict ASA-PS using preoperative NSQIP data. Accuracy was assessed with C-Statistics and calibration plots. We assessed both concurrent and predictive validity of model-derived ASA-PS relative to observed ASA-PS and 30-day outcomes. To aid further research and use of the ASA-PS model, we implemented it into an online calculator. RESULTS: Of the 566,797 elective procedures in the final analytic dataset, 8.9% were ASA-PS 1, 48.9% were ASA-PS 2, 39.1% were ASA-PS 3, and 3.2% were ASA-PS 4. The accuracy of the 21-variable model to predict ASA-PS was C = 0.77 +/- 0.0025. The model-derived ASA-PS had stronger association with key indicators of preoperative status including comorbidities and higher BMI (concurrent validity) compared to observed ASA-PS, but less strong associations with postoperative complications (predictive validity). The online ASA-PS calculator may be accessed at https://s-spire-clintools.shinyapps.io/ASA_PS_Estimator/ CONCLUSIONS: Model-derived ASA-PS better tracked key indicators of preoperative status compared to observed ASA-PS. The ability to have an electronically derived measure of ASA-PS can potentially be useful in research, quality measurement, and clinical applications.

Integrating the Internship into Ophthalmology Residency Programs: Association of University Professors of Ophthalmology American Academy of Ophthalmology White Paper.

Future ophthalmologists will need to have broad skills to thrive in complex health care organizations. However, training for ophthalmologists does not take advantage of all of the postgraduate years (PGYs). Although the traditional residency years seem to have little excess capacity, enhancing the internship year does offer an opportunity to expand the time for ophthalmology training in the same 4 PGYs. Integrating the internship year into residency would allow control of all of the PGYs, allowing our profession to optimize training for ophthalmology. In this white paper, we propose that we could capture an additional 6 months of training time by integrating basic ophthalmology training into the intern year. This would allow 6 additional months to expand training in areas such as quality improvement or time for "mini-fellowships" to allow graduates to develop a deeper set of skills.

Cost-effectiveness of treatment of diabetic macular edema.

BACKGROUND: Macular edema is the most common cause of vision loss among patients with diabetes. OBJECTIVE: To determine the cost-effectiveness of different treatments of diabetic macular edema (DME). DESIGN: Markov model. DATA SOURCES: Published literature and expert opinion. TARGET POPULATION: Patients with clinically significant DME. TIME HORIZON: Lifetime. PERSPECTIVE: Societal. INTERVENTION: Laser treatment, intraocular injections of triamcinolone or a vascular endothelial growth factor (VEGF) inhibitor, or a combination of both. OUTCOME MEASURES: Discounted costs, gains in quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs). RESULTS OF BASE-CASE ANALYSIS: All treatments except laser monotherapy substantially reduced costs, and all treatments except triamcinolone monotherapy increased QALYs. Laser treatment plus a VEGF inhibitor achieved the greatest benefit, gaining 0.56 QALYs at a cost of $6975 for an ICER of $12 410 per QALY compared with laser treatment plus triamcinolone. Monotherapy with a VEGF inhibitor achieved similar outcomes to combination therapy with laser treatment plus a VEGF inhibitor. Laser monotherapy and triamcinolone monotherapy were less effective and more costly than combination therapy. RESULTS OF SENSITIVITY ANALYSIS: VEGF inhibitor monotherapy was sometimes preferred over laser treatment plus a VEGF inhibitor, depending on the reduction in quality of life with loss of visual acuity. When the VEGF inhibitor bevacizumab was as effective as ranibizumab, it was preferable because of its lower cost. LIMITATION: Long-term outcome data for treated and untreated diseases are limited. CONCLUSION: The most effective treatment of DME is VEGF inhibitor injections with or without laser treatment. This therapy compares favorably with cost-effective interventions for other conditions. PRIMARY FUNDING SOURCE: Agency for Healthcare Research and Quality.

Impact of Type 2 diabetes mellitus and insulin use on progression to glaucoma surgery in primary open angle glaucoma.

PURPOSE: To investigate outcomes of primary open-angle glaucoma (POAG) patients with and without type 2 diabetes mellitus (T2DM). METHODS: Retrospective observational study using U.S. nationwide healthcare insurance claims database. Patients ≥40 years old with at least one HbA1c within one year of POAG diagnosis were included. Diabetic factors associated with POAG progression requiring glaucoma surgery were evaluated using multivariable Cox proportional hazards regression models adjusted for demographic, diabetic and glaucoma factors. T2DM diagnosis and use of either oral hypoglycaemic agents or insulin therapy were assessed in association with POAG progression requiring glaucoma surgery. RESULTS: 104,515 POAG patients were included, of which 70,315 (67%) had T2DM. The mean age was 68.9 years (Standard deviation 9.2) and 55% were female. Of those with T2DM, 93% were taking medication (65,468); 95% (62,412) taking oral hypoglycaemic agents, and 34% (22,028) were on insulin. In multivariable analyses, patients with T2DM had a higher hazard of requiring glaucoma surgery (Hazard ratio, HR 1.15, 95% CI 1.09-1.21, p < 0.001). Higher mean HbA1c was also a significant predictor of progression requiring glaucoma surgery (HR 1.02, 95% CI 1.01-1.03, p < 0.001). When evaluating only patients who were taking antidiabetic medication, after adjusting for confounders, insulin use was associated with a 1.20 higher hazard of requiring glaucoma surgery compared to oral hypoglycaemic agents (95% CI 1.14-1.27, p < 0.001), but when stratified by HbA1c, this effect was only significant for those with HbA1c > 7.5%. CONCLUSIONS: Higher baseline HbA1c, particularly in patients taking insulin may be associated with higher rates of glaucoma surgery in POAG.

Dr. Google vs. Dr. ChatGPT: Exploring the Use of Artificial Intelligence in Ophthalmology by Comparing the Accuracy, Safety, and Readability of Responses to Frequently Asked Patient Questions Regarding Cataracts and Cataract Surgery.

PURPOSE: Patients are using online search modalities to learn about their eye health. While Google remains the most popular search engine, the use of large language models (LLMs) like ChatGPT has increased. Cataract surgery is the most common surgical procedure in the US, and there is limited data on the quality of online information that populates after searches related to cataract surgery on search engines such as Google and LLM platforms such as ChatGPT. We identified the most common patient frequently asked questions (FAQs) about cataracts and cataract surgery and evaluated the accuracy, safety, and readability of the answers to these questions provided by both Google and ChatGPT. We demonstrated the utility of ChatGPT in writing notes and creating patient education materials. METHODS: The top 20 FAQs related to cataracts and cataract surgery were recorded from Google. Responses to the questions provided by Google and ChatGPT were evaluated by a panel of ophthalmologists for accuracy and safety. Evaluators were also asked to distinguish between Google and LLM chatbot answers. Five validated readability indices were used to assess the readability of responses. ChatGPT was instructed to generate operative notes, post-operative instructions, and customizable patient education materials according to specific readability criteria. RESULTS: Responses to 20 patient FAQs generated by ChatGPT were significantly longer and written at a higher reading level than responses provided by Google (p < .001), with an average grade level of 14.8 (college level). Expert reviewers were correctly able to distinguish between a human-reviewed and chatbot generated response an average of 31% of the time. Google answers contained incorrect or inappropriate material 27% of the time, compared with 6% of LLM generated answers (p < .001). When expert reviewers were asked to compare the responses directly, chatbot responses were favored (66%). CONCLUSIONS: When comparing the responses to patients' cataract FAQs provided by ChatGPT and Google, practicing ophthalmologists overwhelming preferred ChatGPT responses. LLM chatbot responses were less likely to contain inaccurate information. ChatGPT represents a viable information source for eye health for patients with higher health literacy. ChatGPT may also be used by ophthalmologists to create customizable patient education materials for patients with varying health literacy.

Evaluating Visual Acuity in the American Academy of Ophthalmology IRIS® Registry.

Objective: To describe visual acuity data representation in the American Academy of Ophthalmology Intelligent Research in Sight (IRIS) Registry and present a data-cleaning strategy. Design: Reliability and validity study. Participants: Patients with visual acuity records from 2018 in the IRIS Registry. Methods: Visual acuity measurements and metadata were identified and characterized from 2018 IRIS Registry records. Metadata, including laterality, assessment method (distance, near, and unspecified), correction (corrected, uncorrected, and unspecified), and flags for refraction or pinhole assessment were compared between Rome (frozen April 20, 2020) and Chicago (frozen December 24, 2021) versions. We developed a data-cleaning strategy to infer patients' corrected distance visual acuity in their better-seeing eye. Main Outcome Measures: Visual acuity data characteristics in the IRIS Registry. Results: The IRIS Registry Chicago data set contains 168 920 049 visual acuity records among 23 001 531 unique patients and 49 968 974 unique patient visit dates in 2018. Visual acuity records were associated with refraction in 5.3% of cases, and with pinhole in 11.0%. Mean (standard deviation) of all measurements was 0.26 (0.41) logarithm of the minimum angle of resolution (logMAR), with a range of - 0.3 to 4.0 A plurality of visual acuity records were labeled corrected (corrected visual acuity [CVA], 39.1%), followed by unspecified (37.6%) and uncorrected (uncorrected visual acuity [UCVA], 23.4%). Corrected visual acuity measurements were paradoxically worse than same day UCVA 15% of the time. In aggregate, mean and median values were similar for CVA and unspecified visual acuity. Most visual acuity measurements were at distance (59.8%, vs. 32.1% unspecified and 8.2% near). Rome contained more duplicate visual acuity records than Chicago (10.8% vs. 1.4%). Near visual acuity was classified with Jaeger notation and (in Chicago only) also assigned logMAR values by Verana Health. LogMAR values for hand motion and light perception visual acuity were lower in Chicago than in Rome. The impact of data entry errors or outliers on analyses may be reduced by filtering and averaging visual acuity per eye over time. Conclusions: The IRIS Registry includes similar visual acuity metadata in Rome and Chicago. Although fewer duplicate records were found in Chicago, both versions include duplicate and atypical measurements (i.e., CVA worse than UCVA on the same day). Analyses may benefit from using algorithms to filter outliers and average visual acuity measurements over time. Financial Disclosures: Proprietary or commercial disclosure may be found found in the Footnotes and Disclosures at the end of this article.

Using Natural Language Processing to Identify Different Lens Pathology in Electronic Health Records.

PURPOSE: Nearly all published ophthalmology-related Big Data studies rely exclusively on International Classification of Diseases (ICD) billing codes to identify patients with particular ocular conditions. However, inaccurate or nonspecific codes may be used. We assessed whether natural language processing (NLP), as an alternative approach, could more accurately identify lens pathology. DESIGN: Database study comparing the accuracy of NLP versus ICD billing codes to properly identify lens pathology. METHODS: We developed an NLP algorithm capable of searching free-text lens exam data in the electronic health record (EHR) to identify the type(s) of cataract present, cataract density, presence of intraocular lenses, and other lens pathology. We applied our algorithm to 17.5 million lens exam records in the Sight Outcomes Research Collaborative (SOURCE) repository. We selected 4314 unique lens-exam entries and asked 11 clinicians to assess whether all pathology present in the entries had been correctly identified in the NLP algorithm output. The algorithm's sensitivity at accurately identifying lens pathology was compared with that of the ICD codes. RESULTS: The NLP algorithm correctly identified all lens pathology present in 4104 of the 4314 lens-exam entries (95.1%). For less common lens pathology, algorithm findings were corroborated by reviewing clinicians for 100% of mentions of pseudoexfoliation material and 99.7% for phimosis, subluxation, and synechia. Sensitivity at identifying lens pathology was better for NLP (0.98 [0.96-0.99] than for billing codes (0.49 [0.46-0.53]). CONCLUSIONS: Our NLP algorithm identifies and classifies lens abnormalities routinely documented by eye-care professionals with high accuracy. Such algorithms will help researchers to properly identify and classify ocular pathology, broadening the scope of feasible research using real-world data.

Analysis of the Readability and Accountability of Online Patient Education Materials Related to Glaucoma Diagnosis and Treatment.

Purpose: To assess the readability and accountability of online patient education materials related to glaucoma diagnosis and treatment. Methods: We conducted a Google search for 10 search terms related to glaucoma diagnosis and 10 search terms related to glaucoma treatment. For each search term, the first 10 patient education websites populated after Google search were assessed for readability and accountability. Readability was assessed using five validated measures: Flesch Reading Ease (FRE), Gunning Fog Index (GFI), Flesch-Kincaid Grade Level (FKGL), Simple Measure of Gobbledygook (SMOG), and New Dale-Chall (NDC). Accountability was assessed using the Journal of the American Medical Association (JAMA) benchmarks. The source of information for each article analyzed was recorded. Results: Of the 200 total websites analyzed, only 11% were written at or below the recommended 6th grade reading level. The average FRE and grade level for 100 glaucoma diagnosis-related articles were 42.02 ± 1.08 and 10.53 ± 1.30, respectively. The average FRE and grade level for 100 glaucoma treatment-related articles were 43.86 ± 1.01 and 11.29 ± 1.54, respectively. Crowdsourced articles were written at the highest average grade level (12.32 ± 0.78), followed by articles written by private practice/independent users (11.22 ± 1.74), national organizations (10.92 ± 1.24), and educational institutions (10.33 ± 1.35). Websites averaged 1.12 ± 1.15 of 4 JAMA accountability metrics. Conclusion: Despite wide variation in the readability and accountability of online patient education materials related to glaucoma diagnosis and treatment, patient education materials are consistently written at levels above the recommended reading level and often lack accountability. Articles from educational institutions and national organizations were often written at lower reading levels but are less frequently encountered after Google search. There is a need for accurate and understandable online information that glaucoma patients can use to inform decisions about their eye health.