Highlights
-
•
Annual OAGS-to-POAG conversion rate is low after year 1 (around 5% per year).
-
•
Older, treated suspects had an 8x greater risk than younger, untreated suspects.
-
•
Low-risk OAGS are monitored as frequently as high-risk patients.
-
•
Risk-stratified follow-up could safely extend visits for low-risk patients.
-
•
Per-visit risk thresholds enable efficient, individualized surveillance.
Objective
To assess annual rates and determinants of diagnostic conversion from open angle glaucoma suspects (OAGS) to primary open angle glaucoma (POAG) in the United States and develop a pragmatic, risk-stratified framework for tailoring follow-up frequency based on per-visit conversion risk.
Design
Retrospective cohort study.
Methods
Patients with newly diagnosed OAGS between 2007 and 2021 were identified in Optum’s de-identified Clinformatics Data Mart Database based on International Classification of Diseases codes. Inclusion required (1) continuous enrollment during a 3-year lookback period and 5-year study period from index (first) date of OAGS diagnosis, (2) diagnosis by an ophthalmologist or optometrist with ≥1 optical coherence tomography (OCT) or visual field test, and (3) no prior glaucoma-specific treatment or POAG diagnosis. Cox proportional hazards modeling was performed to assess factors associated with diagnostic conversion to POAG. In a secondary analysis, subgroups were created based on age and treatment status and predicted probability of conversion was estimated for each group.
Main Outcome(s) and Measure(s)
Diagnostic conversion of POAG.
Results
Among 83 305 OAGS patients (57.6% female; 5.4% Asian; 11.1% Black; 12.7% Hispanic; 70.8% non-Hispanic White), 17 134 (20.6%) converted to POAG, corresponding to an overall annual conversion rate of 6.1% (9.4% in year 1, 5.3% in years 2-5). On multivariable Cox regression, older age (HR ≥ 1.38), male sex (HR = 1.12), Black race (HR = 1.14), location outside the Northeast (HR ≥ 1.24), record of gonioscopy (HR = 1.90), and OAGS treatment (HR ≥ 1.31) were associated with greater hazard of conversion ( P ≤.02). Among the lowest-risk patients (<50 years; untreated), annual conversion was 2.0% with follow-up every 0.9 ± 0.6 years, whereas the highest-risk patients (>70 years; treated) had a 16.7% annual conversion rate with follow-up every 0.4 ± 0.4 years after year 1. When standardized to a 5.0% per-visit conversion threshold, corresponding monitoring intervals were 6.1 years and 0.6 years, respectively.
Conclusions
The rate of conversion from OAGS to POAG is <6.0% per year beyond the first year after diagnosis. Estimating per-visit conversion risk enables risk-stratified surveillance strategies that may safely reduce visit frequency for very low-risk patients while preserving timely detection in higher-risk groups, thereby improving clinical efficiency and resource allocation.
INTRODUCTION
Primary open angle glaucoma (POAG) is the leading cause of irreversible vision loss worldwide affecting an estimated 53 million people in 2020 and projected 80 million people by 2040. In the United States (US), POAG is the most common subtype of glaucoma affecting approximately 1.4% to 6.8% of adults over 40 years of age. Due to the ocular morbidity and public health impact of POAG, individuals exhibiting risk factors during routine eye exams, such as elevated intraocular pressure (IOP) or enlarged cup to disc ratio (CDR), are diagnosed as open angle glaucoma suspects (OAGS) and undergo long-term monitoring for conversion to POAG. Landmark glaucoma studies, such as the Ocular Hypertension Treatment Study (OHTS), have reported relatively low annual rates of conversion. Despite these low conversion rates, most patients diagnosed as OAGS are evaluated for glaucoma at least once per year, a practice that consumes substantial healthcare resources, increases patient anxiety, and contributes to lost workplace productivity. ,, As eye care expenditures rise and the size of the ophthalmology workforce declines relative to an aging population, strategies to optimize the allocation of limited healthcare resources are increasingly criticial. ,
Prior studies have estimated annual conversion rates from OAGS to POAG between 0.9% to 7.7%. However, the generalizability of these findings can be limited by the modest sample size, short follow-up, relative homogeneity of study cohorts, and focus on specific subgroups like ocular hypertensives (eg, OHTS). ,,,, In addition, OAGS cases diagnosed using strict research definitions may differ from real-world OAGS diagnoses, which are often based on clinician judgement and heterogenous practice patterns. , These gaps make it difficult to evaluate the cost-effectiveness of glaucoma monitoring strategies and to develop evidence-based guidelines that standardize OAGS care while optimizing resource utilization at scale. This challenge is particularly evident in low-risk OAGS, where the absence of clear clinical risk factors leads to broad and non-specific follow-up recommendations. For example, the American Academy of Ophthalmology (AAO) Preferred Practice Pattern (PPP) guidelines broadly recommend follow up “every 12 to 24 months” for lower-risk OAGS patients, thereby placing substantial discretion on individual clinicians and potentially contributing to inefficient practice variation and resource utilization. Therefore, there is a clear need for additional data on: (1) long-term rates and determinants of diagnostic conversion across diverse patient populations and practice settings; (2) pragmatic strategies for allocating eye care resources based on individualized conversion risk.
In this study, we assessed rates of diagnostic conversion from OAGS to POAG in the US using longitudinal, nationwide healthcare claims data. We further assessed sociodemographic and clinical factors associated with diagnostic conversion, including treatment with intraocular pressure (IOP)-lowering eye drops, laser, and glaucoma surgery. Finally, we developed a pragmatic, risk-stratified framework to approximate monitoring intervals based on per-visit conversion risk to support risk-stratified follow-up. Together, these analyses provide an empirical foundation for data-driven surveillance strategies and directly addresses the growing need to deploy limited eye care resources efficiently as the population ages and glaucoma prevalence continues to rise.
METHODS
Optum’s de-identified Clinformatics Data Mart Database (Optum CDM) is derived from administrative health claims data for members of large commercial and Medicare Advantage health plans. Optum Clinformatics utilizes medical and pharmacy claims to derive patient-level enrollment information, health care costs, and resource utilization information. The population is geographically diverse, spanning all 50 states and is de-identified under the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule’s Expert Determination method and managed according to Optum customer data use agreements. Optum CDM administrative claims submitted for payment by providers and pharmacies are verified, adjudicated, and de-identified prior to inclusion. Available sociodemographic data include sex, race/ethnicity, education level, household income, insurance product, and census division regions. Available clinical data include appointment and diagnosis dates, treatment dates and types, physician type, and first and last dates of enrollment in the CDM.
The University of Southern California Institutional Review Board determined that this study was from institutional review board (IRB) review. The study adhered to the tenets of the Declaration of Helsinki and complied with the Health Insurance Portability and Accountability Act.
STUDY DEFINITIONS
Inclusion in the study population required an index diagnosis of OAGS based on International Classification of Diseases, Ninth Revision (ICD9) or Tenth Revision (ICD-10) codes (Supplemental Table 1). The index date was defined as the date of the first claim associated with an OAGS diagnosis. Age was defined as the age at index diagnosis. Inclusion additionally required 96 months (8 years) of continuous enrollment in the Optum CDM, consisting of a 36-month lookback period before the index date and a 60-month follow-up period thereafter. Baseline diagnoses of myopia, hyperopia, cataract, and pseudophakia were assessed based on ICD-9 and ICD-10 codes (Supplemental Table 1) present on or before the index date. Prior receipt of gonioscopy was assessed using Current Procedural Terminology (CPT) codes recorded before POAG diagnosis (Supplemental Table 1).
Treatment of OAGS was defined as receipt of glaucoma-specific interventions within the study period. These treatments were grouped into IOP-lowering topical medications, laser trabeculoplasty, minimally invasive glaucoma surgery (MIGS; goniotomy, transluminal dilation/canaloplasty, trabecular bypass stent, suprachoroidal shunt), and conventional glaucoma surgery (trabeculectomy, aqueous shunt with or without reservoir, and cyclophotocoagulation) (Supplemental Table 1). Patients who received more than one treatment were classified according to their most intensive treatment, using the following hierarchy: conventional- glaucoma surgery, MIGS, laser trabeculoplasty, then IOP-lowering drops. IOP-lowering drops included alpha agonists, beta blockers, carbonic anhydrase inhibitors, miotics, prostaglandin analogs, and other topical antiglaucoma agents (Supplemental Table 1). Oral or intracameral IOP-lowering medications were excluded. As this study was based on claims data, granular clinical measurements (eg, IOP, cup-to-disc ratio) were not available. Therefore, treatment status was considered as a surrogate for clinician-assessed risk of glaucoma.
Newly diagnosed OAGS cases met the following criteria: (1) continuous enrollment with an eligible insurance provider during the lookback and study periods; (2) OAGS diagnosis by an ophthalmologist or optometrist with record of optical coherence tomography (OCT) and/or visual field (VF) test on or after the index date, identified using CPT codes (Supplemental Table 1); (3) no history of IOP-lowering medications, laser trabeculoplasty, or glaucoma surgery prior to the index date; (4) no diagnosis of POAG before or on the index date. Criterion 2 was implemented to increase diagnostic fidelity. Criterion 3 ensured that patients receiving prior glaucoma-directed were not misclassified as newly diagnosed OAGS. Included OAGS patients were also further classified as having ocular hypertension (OHTN) based on ICD-9/10 codes (Supplemental Table 1) or non-OHTN OAGS. Diagnostic conversion from OAGS to POAG, the primary study endpoint, was defined as a new and distinct POAG diagnosis based on ICD-9 and/or ICD-10 codes within the 60-month follow-up period after the index OAGS diagnosis (Supplemental Table 1). Because ICD-9 diagnostic codes lacked laterality, patient-level (rather than eye-level) analyses were conducted even when ICD-10 codes were available. Therefore, participants with POAG in the fellow eye prior to the study period were excluded, and those whose fellow eye was diagnosed with POAG during the study period were classified as progressors.
STATISTICAL ANALYSIS
The proportion of OAGS with diagnostic conversion to POAG was calculated and stratified by sociodemographic and clinical characteristics. Analyses were conducted on the patient level rather than eye level due to the lack of laterality data in ICD9. Continuous data were expressed as means with SDs. Categorical data were expressed as proportions and percentages.
In the primary analysis, survival time was defined as the duration between the index date and earliest of either the first POAG diagnosis or the last recorded visit. A sensitivity analysis was conducted assuming all OAGS cases without diagnostic conversion within the 5-year follow-up period did not convert regardless of last visit date. An additional sensitivity analysis excluding treatment status as a covariate was performed to identify other factors associated with conversion. Kaplan–Meier survival curves were plotted to visualize cumulative conversion over time. Cox proportional hazards regression models were developed to estimate hazard ratios (HRs) for factors associated with diagnostic conversion. Variables with a P value <.15 in univariable analysis were included in the multivariable analysis, with age, sex, and race/ethnicity forced into all models regardless of univariable significance. Variables were considered clinically and statistically significant if they demonstrated an HR > 1.10 or < 0.90 with a P value of <.01.
In a secondary analysis, the cohort was stratified into 6 subgroups based on age at OAGS diagnosis (<50 years old; 50-70 years old; >70 years old) and OAGS treatment status (treated; untreated). Age and treatment status were selected due to their strong associations with conversion risk and ease of assessment in routine clinical practice. For each subgroup, the predicted probability of conversion was estimated as the average annual conversion rate between years 2 and 5 multiplied by the visit interval (assumed fixed between follow-up visits). The mean observed visit interval for each subgroup was overlaid on the same. The first year of follow-up was excluded due to greater temporal variability in conversion rates, whereas years 2 through 5 demonstrated greater stability. All statistical analysis were performed using R version 4.2.1.
RESULTS
A total of 132 423 patients with a diagnosis of OAGS were identified in the Optum CDM (Supplemental Figure 1). Of these, 83 305 patients (62.9%) met the study definition of newly diagnosed OAGS. The mean age at index diagnosis was 64.9 ± 14.9 years. The cohort included 48 005 (57.6%) female and 35 290 (42.4%) male patients, of whom 4306 (5.4%) were Asian, 8947 (11.1%) Black, 10 213 (12.7%) Hispanic, and 56 922 (70.8%) non-Hispanic White. Of these patients, 10 660 (12.8%) were diagnosed as OHTN ( Table 1 ).
TABLE 1
Proportion With Diagnostic Conversion From Open-Angle Glaucoma Suspect (OAGS) to Primary Open-Angle Glaucoma (POAG)
| N | n | % | ||
|---|---|---|---|---|
| Age | <40 | 4915 | 238 | 4.8% |
| 40-50 | 7924 | 756 | 9.5% | |
| 50-60 | 14 709 | 2096 | 14.2% | |
| 60-70 | 13 630 | 2759 | 20.2% | |
| 70-80 | 30 410 | 7757 | 25.5% | |
| >80 | 11 717 | 3528 | 30.1% | |
| Sex | Female | 48 005 | 9750 | 20.3% |
| Male | 35 290 | 7384 | 20.9% | |
| Race/Ethnicity | Asian | 4306 | 695 | 16.1% |
| Black | 8947 | 2081 | 23.3% | |
| Hispanic | 10 213 | 2174 | 21.3% | |
| White | 56 922 | 11 613 | 20.4% | |
| Education | H.S. degree or less | 424 | 107 | 25.2% |
| H.S. degree | 16 752 | 3754 | 22.4% | |
| Some college | 45 800 | 9617 | 21.0% | |
| College graduate | 18 851 | 3344 | 17.7% | |
| Household Income | <$40,000 | 19 053 | 4553 | 23.9% |
| $40 000-$49 000 | 6364 | 1513 | 23.8% | |
| $50 000-$59 000 | 7151 | 1616 | 22.6% | |
| $60 000-$74 000 | 9263 | 1995 | 21.5% | |
| $75 000-$99,000 | 13 293 | 2765 | 20.8% | |
| >$100 000 | 24 499 | 3900 | 15.9% | |
| Insurance | COM EPO | 3941 | 528 | 13.4% |
| COM HMO | 4333 | 647 | 14.9% | |
| COM Other | 26 501 | 3569 | 13.5% | |
| COM PPO | 1084 | 167 | 15.4% | |
| MCR HMO | 27 640 | 7429 | 26.9% | |
| MCR Other | 15 892 | 3839 | 24.2% | |
| MCR PPO | 3909 | 954 | 24.4% | |
| Location | Northeast | 9820 | 1518 | 15.5% |
| Midwest | 16 091 | 3109 | 19.3% | |
| Mountain | 12 490 | 3054 | 24.5% | |
| Pacific | 11 315 | 2756 | 24.4% | |
| South | 33 488 | 6674 | 19.9% | |
| Myopia | No | 72 986 | 15 054 | 20.6% |
| Yes | 10 319 | 2080 | 20.2% | |
| Hyperopia | No | 75 645 | 15 542 | 20.5% |
| Yes | 7660 | 1592 | 20.8% | |
| Lens Status | No lens diagnosis | 37 346 | 6225 | 16.7% |
| Cataract | 36 532 | 8287 | 22.7% | |
| Pseudophakia | 9427 | 2622 | 27.8% | |
| Gonioscopy | No | 52 707 | 7586 | 14.4% |
| Yes | 30 598 | 9548 | 31.2% | |
| Cataract Surgery | No | 68 875 | 15 441 | 22.4% |
| Yes | 14 430 | 1693 | 11.7% | |
| OAGS Treatment | Untreated | 74 643 | 12 914 | 17.3% |
| IOP-lowering Drops | 8137 | 4006 | 49.2% | |
| Laser Trabeculoplasty | 400 | 162 | 40.5% | |
| MIGS | 40 | 12 | 30.0% | |
| Glaucoma Surgery | 85 | 40 | 47.1% | |
| OHTN | Non-OHTN OAGS | 72 645 | 14 662 | 20.2% |
| OHTN | 10 660 | 2472 | 23.2% |
Abbreviations: N = Open angle glaucoma suspect; n = Detected conversion to primary open angle glaucoma; OAGS = Open angle glaucoma suspect; H.S. = High school; COM = Commercial insurance; MCR = Medicare insurance; EPO = Exclusive provider organization; HMO = Health maintenance organization; PPO = Preferred provider organization; IOP = Intraocular pressure; MIGS = Micro-invasive glaucoma surgery; OHTN = Ocular hypertension.
Despite being continuously enrolled in the Optum CDM for the entire 5-year study period, 6600 patients (7.9%) had no follow-up visits after the index OAGS diagnosis. The remaining 76 705 patients (92.1%) returned for at least one follow-up visit, and 45 822 patients (55.0%) had at least one visit recorded beyond the 5-year study period. At baseline, 10 319 patients (12.4%) had myopia, 7662 (9.2%) hyperopia, 36 532 (43.9%) cataract, and 9427 (11.3%) pseudophakia. A record of gonioscopy prior to POAG diagnosis or the end of follow-up was present in 30 598 patients (36.7%).
Among the full cohort, 74 643 patients (89.6%) received no glaucoma-specific treatment, whereas 8662 (10.4%) received treatment during the study period. Among treated patients, 85 (1.0%) underwent conventional glaucoma surgery, 40 (0.5%) underwent MIGS, 400 (4.6%) underwent laser trabeculoplasty, and 8137 (93.9%) received IOP-lowering drops as their most intensive treatment modality. Progression appeared similar between OAGS patients treated with IOP-lowering drops and those treated with conventional glaucoma surgery, MIGS, and laser trabeculoplasty (Supplemental Figure 2). A total of 14 430 patients (17.3%) underwent cataract surgery prior to conversion or the end of follow-up.
Overall, 17 134 patients (20.6%) converted from OAGS to POAG, corresponding to an average annual conversion rate of 6.1% ( Figure 1 ). The conversion rate was highest in the first year (9.4%) and declined thereafter, stabilizing at an average of 5.3% per year during years 2 through 5. Conversion occurred at a mean age of 71.4 ± 11.0 years, and a mean time from index diagnosis to POAG of 1.6 ± 1.5 years. The annual conversion rate was substantially higher among treated patients (15.2%) compared with untreated patients (5.0%). OHTN patients had a higher annual conversion rate (7.0%) than non-OHTN OAGS patients (5.8%).
Kaplan–Meier survival curves of diagnostic conversion from open-angle glaucoma suspect (OAGS) to primary open-angle glaucoma (POAG).
On univariable analysis, all examined variables except hyperopia met criteria for inclusion in the multivariable model ( Table 2 ). On multivariable analysis, age ≥50 years (HR ≥ 1.38), male sex (HR = 1.12), Black race (HR = 1.14), location outside the Northeast (HR ≥ 1.24), a record of gonioscopy (HR = 1.90), and treatment with IOP-lowering drops (HR = 2.21), laser trabeculoplasty (HR = 1.31), MIGS (HR = 1.95), or conventional glaucoma surgery (HR = 2.47) were associated with significantly greater hazard of conversion ( P ≤.02). In contrast, age <40 years (HR = 0.62), Asian race (HR = 0.87), household income above $100 000 (HR = 0.89), HMO or Other commercial health insurance (HR ≤ 0.87), and prior cataract surgery (HR = 0.36) were associated with significantly lower hazard of conversion ( P ≤.001). OHTN was associated with greater hazard of POAG (HR = 1.12; P <.001) on univariable analysis but was not clinically significant on multivariable analysis (HR = 0.94; P =.004). Baseline myopia (HR = 1.09; P =.001) and pseudophakia (HR = 0.91; P <.001) approached but did not meet the predefined threshold for clinical significance.
TABLE 2
Univariable and Multivariable Analysis of Factors Associated With Diagnostic Conversion From Open-Angle Glaucoma Suspect (OAGS) to Primary Open-Angle Glaucoma (POAG) with Censorship at the Last Appointment
| Univariable Analysis | Multivariable Analysis | ||||
|---|---|---|---|---|---|
| HR (CI) | P-Value | HR (CI) | P-Value | ||
| Age | <40 | 0.60 (0.52-0.70) | <.001 | 0.62 (0.54-0.73) | <.001 |
| 40-50 | Ref | – | Ref | – | |
| 50-60 | 1.40 (1.29-1.53) | <.001 | 1.38 (1.26-1.50) | <.001 | |
| 60-70 | 1.93 (1.78-2.09) | <.001 | 1.89 (1.73-2.06) | <.001 | |
| 70-80 | 2.43 (2.25-2.61) | <.001 | 2.42 (2.21-2.66) | <.001 | |
| >80 | 3.06 (2.83-3.31) | <.001 | 3.14 (2.85-3.46) | <.001 | |
| Sex | Female | Ref | – | Ref | – |
| Male | 1.08 (1.04-1.11) | <.001 | 1.12 (1.08-1.15) | <.001 | |
| Race | Asian | 0.81 (0.75-0.87) | <.001 | 0.87 (0.81-0.95) | .001 |
| Black | 1.21 (1.16-1.27) | <.001 | 1.14 (1.08-1.20) | <.001 | |
| Hispanic | 1.04 (0.99-1.09) | .12 | 0.92 (0.87-0.97) | .001 | |
| White | Ref | – | Ref | – | |
| Education | H.S. degree or less | Ref | – | Ref | – |
| H.S. degree | 0.92 (0.76-1.11) | .39 | 0.96 (0.78-1.17) | .65 | |
| Some college | 0.82 (0.68-1.00) | .05 | 0.93 (0.76-1.13) | .47 | |
| College graduate | 0.68 (0.56-0.83) | <.001 | 0.87 (0.71-1.07) | .19 | |
| Household Income | <$40 000 | Ref | – | Ref | – |
| $40 000-$49 000 | 0.97 (0.92-1.03) | .39 | 1.00 (0.94-1.06) | .99 | |
| $50 000-$59 000 | 0.93 (0.88-0.98) | .01 | 0.97 (0.92-1.03) | .38 | |
| $60 000-$74 000 | 0.87 (0.83-0.92) | <.001 | 0.96 (0.91-1.01) | .14 | |
| $75 000-$99 000 | 0.82 (0.78-0.86) | <.001 | 0.95 (0.90-1.00) | .04 | |
| > $100 000 | 0.64 (0.61-0.67) | <.001 | 0.89 (0.84-0.93) | <.001 | |
| Insurance | COM EPO | 0.56 (0.52-0.61) | <.001 | 0.92 (0.83-1.02) | .12 |
| COM HMO | 0.55 (0.51-0.59) | <.001 | 0.78 (0.71-0.86) | <.001 | |
| COM Other | 0.54 (0.52-0.56) | <.001 | 0.87 (0.82-0.92) | <.001 | |
| COM PPO | 0.58 (0.50-0.68) | <.001 | 0.89 (0.76-1.05) | .16 | |
| MCR HMO | Ref | – | Ref | – | |
| MCR Other | 0.97 (0.94-1.01) | .16 | 1.02 (0.97-1.06) | .49 | |
| MCR PPO | 1.01 (0.94-1.08) | .81 | 1.03 (0.95-1.10) | .50 | |
| Division | Northeast | Ref | – | Ref | – |
| Midwest | 1.20 (1.13-1.28) | <.001 | 1.25 (1.17-1.33) | <.001 | |
| Mountain | 1.44 (1.35-1.53) | <.001 | 1.25 (1.16-1.33) | <.001 | |
| Pacific | 1.56 (1.47-1.66) | <.001 | 1.34 (1.25-1.44) | <.001 | |
| South | 1.32 (1.25-1.40) | <.001 | 1.30 (1.22-1.38) | <.001 | |
| Myopia | No | Ref | – | Ref | – |
| Yes | 0.94 (0.90-0.99) | .02 | 1.09 (1.04-1.14) | .001 | |
| Hyperopia * | No | Ref | – | NA | NA |
| Yes | 1.00 (0.95-1.05) | .89 | NA | NA | |
| Lens Status | No lens diagnosis | 0.78 (0.75-0.80) | <.001 | 1.03 (0.99-1.07) | .15 |
| Cataract | Ref | – | Ref | – | |
| Pseudophakia | 1.25 (1.19-1.30) | <.001 | 0.91 (0.87-0.96) | <.001 | |
| Gonioscopy | No | Ref | – | Ref | – |
| Yes | 1.96 (1.90-2.02) | <.001 | 1.90 (1.84-1.96) | <.001 | |
| Cataract Surgery | No | Ref | – | Ref | – |
| Yes | 0.47 (0.45-0.50) | <.001 | 0.36 (0.34-0.38) | <.001 | |
| OAGS Treatment | Untreated | Ref | – | Ref | – |
| IOP-lowering Drops | 2.54 (2.46-2.64) | <.001 | 2.21 (2.13-2.30) | <.001 | |
| Laser Trabeculoplasty | 1.79 (1.54-2.10) | <.001 | 1.31 (1.11-1.54) | .001 | |
| MIGS | 1.48 (0.84-2.60) | .18 | 1.95 (1.10-3.43) | .02 | |
| Glaucoma Surgery | 2.77 (2.03-3.77) | <.001 | 2.47 (1.80-3.38) | <.001 | |
| OHTN | Non-OHTN OAGS | Ref | – | Ref | – |
| OHTN | 1.12 (1.07-1.17) | <.001 | 0.94 (0.89-0.98) | .004 | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
