We analyze daily Airbnb service-fee shares across eleven settlement currencies, a compositional series that shows bursts of volatility after shocks such as the COVID-19 pandemic. Standard Dirichlet time series models assume constant precision and therefore miss these episodes. We introduce B-DARMA-DARCH, a Bayesian Dirichlet autoregressive moving average model with a Dirichlet ARCH component, which lets the precision parameter follow an ARMA recursion. The specification preserves the Dirichlet likelihood so forecasts remain valid compositions while capturing clustered volatility. Simulations and out-of-sample tests show that B-DARMA-DARCH lowers forecast error and improves interval calibration relative to Dirichlet ARMA and log-ratio VARMA benchmarks, providing a concise framework for settings where both the level and the volatility of proportions matter.

Importance Risk stratification is a key component of syncope management in the emergency department (ED), but objective risk-stratification tools have not been widely adopted. Objective To externally validate 2 risk-stratification tools, the Canadian Syncope Risk Score (CSRS) and FAINT score, and compare their performance with unstructured physician-estimated risk. Design, Setting, and Participants This prospective, observational, multicenter cohort study was conducted across the US from September 2020 to September 2024, enrolling ED patients aged 40 years or older with syncope or presyncope and no serious diagnosis during initial evaluation. Data were analyzed from September to December 2025. Exposure Presence of syncope or presyncope. Main Outcomes and Measures The primary outcome was a serious adverse outcome within 30 days of the ED visit. This included both cardiac and noncardiac outcomes for the CSRS and exclusively cardiac outcomes for the FAINT score. Unstructured estimates of each patient's risk for serious adverse outcome were obtained from the treating attending physician. Risk scores and physician-estimated risk were assessed using the area under the receiver operating characteristic curve (AUROC). Sensitivity and negative predictive value (NPV) were also calculated. Results A total of 1263 patients were analyzed (median [IQR] age, 66.0 [55.0-74.0] years; 676 female [53.5%]), of whom 74 (5.9%) had a serious adverse outcome and 62 (4.9%) had a serious cardiac outcome. The CSRS AUROC for any serious outcome was 0.72 (95% CI, 0.67-0.78). A low-risk score (CSRS <0) had a sensitivity of 91.9% (95% CI, 85.7%-98.1%) and an NPV of 97.5% (95% CI, 96.4%-98.8%). The FAINT score AUROC for a serious cardiac outcome was 0.76 (95% CI, 0.71-0.81). A low-risk score (FAINT = 0) had a sensitivity of 96.7% (95% CI, 92.4%-100%) and an NPV of 98.8% (95% CI, 98.0%-99.7%). In comparison, physician-estimated risk less than 2% had a sensitivity of 87.5% (95% CI, 79.4%-95.6%) and an NPV of 97.4% (95% CI, 95.5%-98.9%) for any serious adverse outcome. Physician-estimated risk less than 2% had a sensitivity of 79.2% (95% CI, 68.3%-90.2%) and an NPV of 97.4% (95% CI, 95.9%-98.9%) for serious cardiac outcomes. The FAINT score, but not the CSRS, was more sensitive than physician-estimated risk (difference, 18.7%; 95% CI, 6.8%-30.6%; P = .002). Conclusions and Relevance In this cohort study of 1263 patients, both the CSRS and the FAINT score identified patients at low risk for adverse outcomes, and FAINT outperformed physicians at identifying patients at low risk for serious cardiac outcomes. Further research on the effect of score implementation is required to understand its impact on patient care.

BACKGROUND:Syncope is common in the Emergency Department (ED) and can be associated with structural heart disease (SHD). Transthoracic echocardiography (TTE) is commonly ordered to assess for SHD. OBJECTIVE:To externally validate the previously developed ROMEO score, which identifies patients with syncope who are at very low risk of significant findings on TTE. DESIGN:Secondary analysis of a multicenter, prospective, observational cohort study. SETTING:One community and five academic EDs in the United States. PARTICIPANTS:Adults (≥ 40 years old) presenting to the ED with syncope or presyncope, without a serious ED diagnosis. INTERVENTIONS:Receipt of TTE within 30 days of the index visit. MEASUREMENTS:Our primary outcome was the rate of significant findings on TTE within 30 days of the index ED visit. We calculated the sensitivity, specificity, negative and positive predictive values (NPV, PPV), and the area under the curve (AUC) of the ROMEO score. RESULTS:We enrolled 1287 patients, of whom 427 underwent TTE. 88 (20.6%) had a significant finding. A ROMEO score of zero had a sensitivity of 98.9% and a NPV of 98.6%. The specificity and PPV were 20.2% and 24.3%, respectively. The AUC was 0.83 (95% CI: 0.79 to 0.87). LIMITATIONS:Given the rate of non-enrollment of screened patients, there is potential for selection bias. Our study sample was biased toward urban, academic centers; the results may not apply to community settings. CONCLUSIONS:The ROMEO score demonstrates strong predictive performance and may be useful to help clinicians identify patients who are unlikely to benefit from TTE. TRIAL REGISTRATION:ClinicalTrial.gov identifier: NCT04533425.

OBJECTIVES:Syncope and presyncope are common emergency department (ED) presentations that may indicate dangerous underlying conditions. For patients without serious ED diagnoses, hospitalization for monitoring and further workup is common, yet the benefits remain unclear. We sought to determine the diagnostic yield of hospitalization among adults with unexplained syncope or presyncope. METHODS:We conducted a secondary analysis of data from a prospective, multicenter, observational study enrolling ED patients aged ≥ 40 years with syncope or presyncope and no serious ED diagnosis. We collected 30-day serious adverse outcomes (SAOs), both cardiac and non-cardiac, including only in-hospital SAOs for admitted patients and all SAOs for discharged patients. Logistic and Cox regression analyses compared admitted and discharged participants using propensity score adjustment and matching. RESULTS:Among 1263 patients (mean age 64.8 ± 13.1 years), 74 (5.9%) experienced any SAO, including 62 (4.9%) with serious cardiac outcomes. After propensity-score adjustment, Bayesian logistic regression showed a significant difference in diagnostic yield (OR 3.70 [95% CrI 1.85-6.82]) in the hospitalized cohort. In a propensity-score-matched Cox regression analysis, hospitalization was associated with a shorter time to SAO diagnosis, with an HR of 12.43 (95% CI 2.94-52.48). CONCLUSION:Hospitalization increased the diagnostic yield for SAOs and accelerated time to diagnosis. It is reasonable to hospitalize select, higher-risk adult patients over 40 years old with presyncope or syncope, even if no dangerous diagnosis is found in the ED. TRIAL REGISTRATION:ClinicalTrials.gov identifier: NCT04533425.

STUDY OBJECTIVE:Previous research suggests that the short-term incidence of adverse events is similar in emergency department (ED) patients with presyncope and syncope. However, admission rates for presyncope are lower, which could imply clinicians underestimate its risk. We sought to compare physician risk estimates and the 30-day rate of serious cardiac outcomes between patients with syncope and presyncope. METHODS:We conducted a secondary analysis of a prospective, observational, multicenter study of patients aged ≥40 years presenting to ED with presyncope or syncope. Patients with serious ED diagnoses were excluded. Descriptive statistics and multivariable regression analyses were used to compare the physician-estimated risk, ED disposition, and 30-day rate of adverse outcomes. RESULTS:Of the 1,263 patients analyzed, 721 (57%) had syncope and 542 (43%) had presyncope. Baseline characteristics were similar between groups. At 30 days, 34 (4.7%) syncope patients and 28 (5.2%) presyncope patients experienced a serious cardiac outcome; logistic regression showed no difference in the odds (odds ratio 1.13; 95% confidence interval 0.66 to 1.79) of serious cardiac outcomes between syncope and presyncope patients. The mean physician-estimated risk of serious cardiac outcomes was 7.6% in syncope, versus 5.3% in presyncope (risk difference 2.3% [0.89%, 3.7%]); this difference remained significant after adjustment for clinical characteristics. Admission rate was lower in presyncope, 38.2% versus 49.5% (risk difference 11.3% [1.2%, 21.5%]). CONCLUSION:Patients with unexplained presyncope and syncope had similar rates of 30-day serious cardiac outcomes after ED visit. Patients with presyncope were less likely to be admitted and had a lower mean physician-estimated risk of adverse outcomes.

We examine how prior specification affects the Bayesian Dirichlet Auto-Regressive Moving Average (B-DARMA) model for compositional time series. Through three simulation scenarios—correct specification, overfitting, and underfitting—we compare five priors: informative, horseshoe, Laplace, mixture of normals, and hierarchical. Under correct model specification, all priors perform similarly, although the horseshoe and hierarchical priors produce slightly lower bias. When the model overfits, strong shrinkage—particularly from the horseshoe prior—proves advantageous. However, none of the priors can compensate for model misspecification if key VAR/VMA terms are omitted. We apply B-DARMA to daily S&P 500 sector trading data, using a large-lag model to demonstrate overparameterization risks. Shrinkage priors effectively mitigate spurious complexity, whereas weakly informative priors inflate errors in volatile sectors. These findings highlight the critical role of carefully selecting priors and managing model complexity in compositional time-series analysis, particularly in high-dimensional settings.

This study compares pre-exposure prophylaxis (PrEP) use among men who have sex with men (MSM) with a recent history of incarceration across various factors known to contribute to HIV transmission risk, including sexual identity, race/ethnicity, sexual activity, incarceration history, injection drug use, and internalized homophobia. We analyzed baseline lifetime PrEP use (yes or no) of 170 male-identifying participants enrolled in a randomized-controlled trial in Los Angeles County between 2019 and 2022. Using logistic regression, we assessed the association of PrEP with sexual identity, socio-demographics, and potential confounders. Compared to gay/same-gender loving-identified participants, straight/heterosexual-identified (aOR = 0.10, CI = 0.02-0.49) and bi-pansexual-identified (0.39, 0.16-0.95) participants had reduced odds of PrEP use. Black/African American-identified participants had lower odds (0.15, 0.03-0.78) of PrEP use than White-identified participants. Participants reporting 3+ years cumulative lifetime incarceration had lower odds (0.28, 0.09-0.87) of PrEP use than participants reporting <6 months. Controlling for internalized homophobia rendered differences among sexual identity groups non-significant. A similar effect was not observed for race/ethnicity and lifetime incarceration. Internalized homophobia was an important driver of PrEP use differences among MSM along the lines of sexual identity but not along the lines of race/ethnicity or cumulative incarceration.

Purpose: To investigate global longitudinal structure–function (SF) relationships between macular ganglion cell complex (GCC) thickness and central visual field (VF) mean deviation (MD) rates of change using a Bayesian joint bivariate longitudinal model. Design: Prospective cohort study. Participants: One hundred seventeen eyes from 117 patients with glaucoma with central damage or moderate to advanced glaucoma were included. Eligible patients had at least 4 visits over a follow-up period of 2 years or longer. Methods: Longitudinal GCC thickness was assessed using optical coherence tomography, and central VF MD was measured with 10-2 standard automated perimetry. A Bayesian joint bivariate longitudinal model was used to estimate random intercepts, slopes, and residual standard deviations (SDs) for structural and functional measures and their correlations. A simulation study compared the Bayesian model (BM)'s performance against simple linear regression (SLR) in estimating these correlations. Main Outcome Measures: Correlations between GCC and MD intercepts, slopes, and residual errors. Results: The mean baseline MD was −8.3 (SD: 5.2) decibels, with an average follow-up period of 5.0 (SD: 0.9) years. The mean correlation was 0.47 (95% credible interval: 0.32 to 0.61) for GCC-MD intercepts (baseline values), 0.29 (95% credible interval: 0.04 to 0.52) for GCC-MD slopes (rates of change), 0.20 (95% credible interval: −0.06 to 0.44) for GCC-MD log residual SDs, and 0.060 (95% credible interval: −0.013, 0.132) for the observation-level GCC-MD residual correlation. The BM consistently demonstrated a smaller root mean squared error than SLR in estimating GCC-MD slope correlations in all simulated scenarios where GCC-MD residual correlation differed from GCC-MD slope correlation. Conclusions: The Bayesian joint model improved accuracy and reduced uncertainty in estimating SF relationships compared to SLR. Correlations between global SF rates of change were significantly positive, although weaker than random intercept correlations. This model represents a key step towards developing local longitudinal SF models to enhance glaucoma progression monitoring. Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Purpose:Individual visual field (VF) sensitivities become unreliable at threshold sensitivities of 19 dB or less, limiting glaucoma monitoring. We evaluated longitudinal variability of central 10° VF measurements based on baseline sensitivity using a Bayesian hierarchical model. Methods:We included 124 glaucoma patients (124 eyes) with central or moderate-to-advanced VF damage, more than 2 years follow-up, and more than 4 central 10-2 VF tests. A Bayesian linear model estimated pointwise change rates, compared with simple linear regression (SLR). Simulations modeled average (-0.21 dB/year) and benchmark (-0.5 dB/year) slopes with residual standard deviations (SD) of 2, 4, 7, or 10 dB. Outcomes included pointwise residual SDs and proportions of significant slopes in cohort and simulations. Results:The average baseline 10-2 VF mean deviation, follow-up time, and median VF tests were 8.4 ± 5.4 dB, 4.6 ± 0.8 years, and 9 VF tests (range, 4-12 VF tests), respectively. The mean global slopes for Bayesian and SLR models were -0.21 and -0.36 dB/year. Residual SDs were markedly higher when baseline threshold sensitivities was 5 to 20 dB compared with 25 dB or greater. The Bayesian model identified more significant negative slopes, particularly at points with residual SD of less than 4 dB, relative to SLR. Conclusions:When baseline pointwise sensitivity is 5 to 20 dB, residual variability is very large, substantially reducing the ability to detect glaucoma progression. Translational Relevance:Visual field locations with sensitivity near or less than 20 dB demonstrate markedly greater variability over time; thus, excluding these points from visual field algorithms or analytical models could improve efficiency in detecting perimetric progression.

PURPOSE:To investigate the influence of baseline blood pressure (BP) on retinal nerve fiber layer (RNFL) rates of change (RoCs) in glaucoma patients with central damage or moderate to severe disease. DESIGN:Prospective cohort study. PARTICIPANTS:One hundred ten eyes with ≥ 4 RNFL OCT scans and ≥ 2 years of follow-up. METHODS:Global RNFL RoCs were modeled with a Bayesian hierarchical model with subject- and sector-level random effects. Influence of baseline systolic and diastolic BP measures and their interactions with intraocular pressure (IOP) on global RNFL RoCs was investigated in prognostic models adjusting for relevant baseline demographic and clinical measures. MAIN OUTCOME MEASURES:Magnitude and direction of coefficients for BP, IOP, and their interaction for prediction of global RNFL RoCs. One-sided Bayesian P values denote posterior probability that a regression coefficient is greater than or less than zero, with P < 0.025 or P > 0.975 defining significance. RESULTS:Average (standard deviation) 24-2 visual field mean deviation (MD) at baseline, follow-up time, and number of OCT scans were -8.8 (6.0) dB, 4.3 (0.5) years, and 8.3 (1.4), respectively. In multivariable analyses, female sex, Hispanic ethnicity (vs. White ethnicity), better baseline 24-2 MD, higher contrast sensitivity at 12 cycles per degree, presence of diabetes, and thicker central cornea predicted faster RNFL thinning. Adjusted for covariates, lower diastolic BP combined with higher IOP predicted faster RNFL RoCs. Parallel multivariable models incorporating systolic BP showed similar effects. Among various BP/IOP combinations, eyes with IOP at the 90th percentile and diastolic (systolic) BP at 10th percentile demonstrated the fastest RNFL thinning rates (-0.554 and -0.539 μm/year). CONCLUSIONS:Low BP and higher IOP at baseline predicted faster (worse) RNFL RoCs in glaucoma patients with central damage or moderate to advanced disease. Although there may be potential benefits to BP management in glaucoma patients, the therapeutic value of BP manipulation in glaucoma patients is yet to be established given the proven benefits of tight BP control in reducing cardiovascular morbidity and mortality. FINANCIAL DISCLOSURE(S):Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Prior choice can strongly influence Bayesian Dirichlet ARMA (B-DARMA) inference for compositional time-series. Using simulations with (i) correct lag order, (ii) overfitting, and (iii) underfitting, we assess five priors: weakly-informative, horseshoe, Laplace, mixture-of-normals, and hierarchical. With the true lag order, all priors achieve comparable RMSE, though horseshoe and hierarchical slightly reduce bias. Under overfitting, aggressive shrinkage-especially the horseshoe-suppresses noise and improves forecasts, yet no prior rescues a model that omits essential VAR or VMA terms. We then fit B-DARMA to daily SP 500 sector weights using an intentionally large lag structure. Shrinkage priors curb spurious dynamics, whereas weakly-informative priors magnify errors in volatile sectors. Two lessons emerge: (1) match shrinkage strength to the degree of overparameterization, and (2) prioritize correct lag selection, because no prior repairs structural misspecification. These insights guide prior selection and model complexity management in high-dimensional compositional time-series applications.

HIV disproportionately impacts minoritized individuals, particularly those of intersectional minoritized identities. Incarceration disproportionately impacts minoritized individuals as well, and increases HIV risk, in part due to its disruption to people’s lives, social networks, and access to care. We developed MEPS, a 6-month intervention designed to holistically support HIV prevention in men who have sex with men and transgender women leaving incarceration. We tested MEPS in a 1:1 randomized controlled trial with 208 individuals. All participants received a needs assessment and personalized wellness plan, followed by either standard of care or the MEPS intervention. MEPS integrated support from a Peer Mentor, incentives for engagement in social and health services, and a mobile app. Participants completed baseline assessments and follow-up assessments at 3, 6, and 9 months. We tested for changes in PrEP use using a group-based trajectory model, for changes in HIV and STI testing, frequent substance use and recidivism using logistic mixed models, and for changes in HCV testing and in having a regular place for care using Poisson models. MEPS participants were significantly more likely than control participants to be among those who used PrEP [AOR (95

PURPOSE: Demonstrate that a novel Bayesian hierarchical spatial longitudinal (HSL) model identifies macular superpixels with rapidly deteriorating ganglion cell complex (GCC) thickness more efficiently than simple linear regression (SLR). center dot DESIGN: Prospective cohort study. center dot SETTING: Tertiary Glaucoma Center. center dot SUBJECTS: One hundred e leven eyes (111 patients) with moderate to severe glaucoma at baseline and >4 macular optical coherence tomography scans and >2 years of follow-up. center dot OBSERVATION PROCEDURE: Superpixel-patient-specific GCC slopes and their posterior variances in 49 superpixels were derived from our latest Bayesian HSL model and Bayesian SLR. A simulation cohort was created with known intercepts, slopes, and residual variances in individual superpixels. center dot MAIN OUTCOME MEASURES: We compared HSL and SLR in the fastest progressing deciles on (1) proportion of superpixels identified as significantly progressing in the simulation study and compared to SLR slopes in cohort data; (2) root mean square error (RMSE), and SLR/HSL RMSE ratios. center dot RESULTS: Cohort- In the fastest decile of slopes per SLR, 77% and 80% of superpixels progressed significantly according to SLR and HSL, respectively. The SLR/HSL posterior SD ratio had a median of 1.83, with 90% of ratios favoring HSL. Simulation- HSL identified 89% significant negative slopes in the fastest progressing decile vs 64% for SLR. SLR/HSL RMSE ratio was 1.36 for the fastest decile of slopes, with 83% of RMSE ratios favoring HSL. center dot CONCLUSION: The Bayesian HSL model improves the estimation efficiency of local GCC rates of change regardless of underlying true rates of change, particularly in fast progressors. (Am J Ophthalmol 2024;261: 8594. (c) 2024 The Authors.

Purpose:Demonstrate that a novel Bayesian hierarchical spatial longitudinal (HSL) model improves estimation of local macular ganglion cell complex (GCC) rates of change compared to simple linear regression (SLR) and a conditional autoregressive (CAR) model. Methods:We analyzed GCC thickness measurements within 49 macular superpixels in 111 eyes (111 patients) with four or more macular optical coherence tomography scans and two or more years of follow-up. We compared superpixel-patient-specific estimates and their posterior variances derived from the latest version of a recently developed Bayesian HSL model, CAR, and SLR. We performed a simulation study to compare the accuracy of intercept and slope estimates in individual superpixels. Results:HSL identified a significantly higher proportion of significant negative slopes in 13/49 superpixels and a significantly lower proportion of significant positive slopes in 21/49 superpixels than SLR. In the simulation study, the median (tenth, ninetieth percentile) ratio of mean squared error of SLR [CAR] over HSL for intercepts and slopes were 1.91 (1.23, 2.75) [1.51 (1.05, 2.20)] and 3.25 (1.40, 10.14) [2.36 (1.17, 5.56)], respectively. Conclusions:A novel Bayesian HSL model improves estimation accuracy of patient-specific local GCC rates of change. The proposed model is more than twice as efficient as SLR for estimating superpixel-patient slopes and identifies a higher proportion of deteriorating superpixels than SLR while minimizing false-positive detection rates. Translational Relevance:The proposed HSL model can be used to model macular structural measurements to detect individual glaucoma progression earlier and more efficiently in clinical and research settings.

Lead time data is compositional data found frequently in the hospitality industry. Hospitality businesses earn fees each day, however these fees cannot be recognized until later. For business purposes, it is important to understand and forecast the distribution of future fees for the allocation of resources, for business planning, and for staffing. Motivated by 5 years of daily fees data, we propose a new class of Bayesian time series models, a Bayesian Dirichlet Auto-Regressive Moving Average (B-DARMA) model for compositional time series, modeling the proportion of future fees that will be recognized in 11 consecutive 30 day windows and 1 last consecutive 35 day window. Each day's compositional datum is modeled as Dirichlet distributed given the mean and a scale parameter. The mean is modeled with a Vector Autoregressive Moving Average process after transforming with an additive log ratio link function and depends on previous compositional data, previous compositional parameters and daily covariates. The B-DARMA model offers solutions to data analyses of large compositional vectors and short or long time series, offers efficiency gains through choice of priors, provides interpretable parameters for inference, and makes reasonable forecasts.

We model longitudinal macular thickness measurements to monitor the course of glaucoma and prevent vision loss due to disease progression. The macular thickness varies over a 6 × 6 grid of locations on the retina, with additional variability arising from the imaging process at each visit. currently, ophthalmologists estimate slopes using repeated simple linear regression for each subject and location. To estimate slopes more precisely, we develop a novel Bayesian hierarchical model for multiple subjects with spatially varying population-level and subject-level coefficients, borrowing information over subjects and measurement locations. We augment the model with visit effects to account for observed spatially correlated visit-specific errors. We model spatially varying: (a) intercepts, (b) slopes, and (c) log-residual standard deviations (SD) with multivariate Gaussian process priors with Matérn cross-covariance functions. Each marginal process assumes an exponential kernel with its own SD and spatial correlation matrix. We develop our models for and apply them to data from the Advanced Glaucoma Progression Study. We show that including visit effects in the model reduces error in predicting future thickness measurements and greatly improves model fit.

HIV and substance abuse are common among young men, associated with a cluster of risk behaviors. Yet, most services addressing these challenges are delivered in setting underutilized by men and are often inconsistent with male identity. This cluster randomized controlled trial aimed to reduce multiple risk behaviors found among young men township areas on the outskirts of Cape Town, South Africa. Young men aged 18–29 years (N = 1193) across 27 neighborhoods were randomized by area to receive HIV-related skills training during either: (1) a 12-month soccer league (SL) intervention; (2) 6-month SL followed by 6 months of vocational training (VT) intervention (SL/VT, n = 9); or 3) a control condition (CC). Bayesian longitudinal mixture models were used to evaluate behaviors over time. Because we targeted multiple outcomes as our primary outcome, we analyzed if the number of significantly different outcomes between conditions exceeded chance for 13 measures over 18 months (with 83

Purpose To compare ganglion cell complex (GCC) and retinal nerve fiber layer (RNFL) rates of change (RoC) in eyes with central or moderate to advanced glaucoma. Design Prospective cohort study. Participants 918 matched macular and RNFL OCT scan pairs from 109 eyes (109 patients) enrolled in the Advanced Glaucoma Progression Study with ≥2 years of follow-up and ≥4 OCT scans. Methods We exported GCC and RNFL thickness measurements in 49 central macular superpixels and 12 RNFL clock-hour sectors, respectively. We applied our latest Bayesian hierarchical longitudinal model to estimate population and subject-specific baseline thickness (intercepts) and rates of change (RoC) in macular superpixels and RNFL sectors. Global RNFL and GCC RoC were analyzed in a single bivariate longitudinal model to properly compare them accounting for the correlation between their RoC. Main Outcome Measures Proportion of significant negative (deteriorating) and positive (improving) RoC expressed in μm/year. Standardized RoC were calculated by dividing RoC by the corresponding population SD. Analyses were repeated in eyes with visual field mean deviation (MD) ≤–6 and >–6 dB. Results Average (SD) 24-2 visual field MD and follow-up length were –8.6 (6.3) dB and 4.2 (0.5) years, respectively. Global RNFL RoC (–0.70 µm/year) were faster than GCC (–0.44 µm/year) (p<.001); corresponding normalized RoC were not significantly different (p=0.052). In bivariate analysis, patients with a significant negative global RNFL RoC (n=63, 57%) or GCC (n=56, 51%) frequently did so for both outcomes (n=49, 45%). The average proportion of significantly decreasing RNFL sectors within an eye was 30.7% in eyes with MD >–6 dB compared to 20.5% in those with MD ≤–6 dB (p=0.014); the proportions for GCC superpixels were 21.1% vs. 18.7%, respectively (p=0.63). Conclusions Both GCC and RNFL measures can detect structural progression in glaucoma patients with central damage or moderate to advanced glaucoma. The clinical utility of RNFL imaging decreases with worsening severity of glaucoma.