Yiyue Lou

PURPOSE: To compare performance of a deep-learning enhanced algorithm for automated detection of diabetic retinopathy (DR), to the previously published performance of that algorithm, the Iowa Detection Program (IDP)-without deep learning components-on the same publicly available set of fundus images and previously reported consensus reference standard set, by three US Board certified retinal specialists. METHODS: We used the previously reported consensus reference standard of referable DR (rDR), defined as International Clinical Classification of Diabetic Retinopathy moderate, severe nonproliferative (NPDR), proliferative DR, and/or macular edema (ME). Neither Messidor-2 images, nor the three retinal specialists setting the Messidor-2 reference standard were used for training IDx-DR version X2.1. Sensitivity, specificity, negative predictive value, area under the curve (AUC), and their confidence intervals (CIs) were calculated. RESULTS: Sensitivity was 96.8% (95% CI: 93.3%-98.8%), specificity was 87.0% (95% CI: 84.2%-89.4%), with 6/874 false negatives, resulting in a negative predictive value of 99.0% (95% CI: 97.8%-99.6%). No cases of severe NPDR, PDR, or ME were missed. The AUC was 0.980 (95% CI: 0.968-0.992). Sensitivity was not statistically different from published IDP sensitivity, which had a CI of 94.4% to 99.3%, but specificity was significantly better than the published IDP specificity CI of 55.7% to 63.0%. CONCLUSIONS: A deep-learning enhanced algorithm for the automated detection of DR, achieves significantly better performance than a previously reported, otherwise essentially identical, algorithm that does not employ deep learning. Deep learning enhanced algorithms have the potential to improve the efficiency of DR screening, and thereby to prevent visual loss and blindness from this devastating disease.

Elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA) was shown to be safe and efficacious in people with cystic fibrosis (CF) with ≥ 1 F508del-CFTR allele in Phase 3 clinical trials. ELX/TEZ/IVA treatment led to improved lung function, with increases in percent predicted forced expiratory volume in 1 second (ppFEV1) and Cystic Fibrosis Questionnaire-Revised respiratory domain score. Here, we evaluated the impact of ELX/TEZ/IVA on the rate of lung function decline over time by comparing changes in ppFEV1 in participants from the Phase 3 trials with a matched group of people with CF from the US Cystic Fibrosis Foundation Patient Registry not eligible for cystic fibrosis transmembrane conductance regulator (CFTR) modulator therapy. Participants treated with ELX/TEZ/IVA had on average no loss of pulmonary function over a 2-year period (mean annualized rate of change in ppFEV1, +0.39 percentage points [95% CI, −0.06 to 0.85]) compared with a 1.92 percentage point annual decline (95% CI, −2.16 to −1.69) in ppFEV1 in untreated controls. ELX/TEZ/IVA is the first CFTR modulator therapy shown to halt lung function decline over an extended time period.

BACKGROUND: Concordance between survey reports and claims data is not well established. We compared them for disease histories, preventative, and other health services use in a large, nationally representative sample of older Medicare beneficiaries with special attention given to evaluating age, aging, memory, and respondent status effects. METHODS: Baseline (1993) and biennial follow-up data (through 2010) from the Survey on Assets and Health Dynamics among the Oldest-Old were linked to Medicare claims from 1991 to 2010, for 4910 participants yielding 19,556 person-periods. Concordance was measured by simple, weighted, and prevalence and bias-adjusted κ, and Lin's concordance statistics. Generalized estimating equation negative binomial models were used to predict the summary counts of concordant reports, survey underreports, and survey overreports. RESULTS: Concordance was highly variable overall, unacceptably low for arthritis and physician visits, and less than substantial for angina, heart disease, hypertension, and outpatient surgery. Generalized estimating equation negative binomial models revealed reductions in reporting accuracy (more underreporting and overreporting) associated with both age (interindividual) and aging (intraindividual) effects, countervailing memory effects on concordance due to less underreporting but more overreporting, and countervailing proxy-respondent effects on concordance due to less underreporting but more overreporting. CONCLUSIONS: Further research should explore whether these findings are time or cohort bound, address the potential heterogeneity of the proxy-respondent effects based on the reason for and relationship of the proxy to the target person, and evaluate the effects of a broader spectrum of performance-based cognitive abilities. In the interim, the significant predictors identified here should be included in future studies.

In clinical endpoint bioequivalence (BE) studies, the primary analysis for assessing equivalence between a generic and an innovator product is based on the observed per-protocol (PP) population (usually completers and compliers). However, missing data and noncompliance are post-randomization intercurrent events and may introduce selection bias. Therefore, PP analysis is generally not causal. The FDA Missing Data Working Group recommended using "causal estimands of primary interest." In this paper, we propose a principal stratification causal framework and co-primary causal estimands to test equivalence, which was also recommended by the recently published ICH E9 (R1) addendum to address intercurrent events. We identify three conditions under which the current PP estimator is unbiased for one of the proposed co-primary causal estimands - the "Survivor Average Causal Effect" (SACE) estimand. Simulation shows that when these three conditions are not met, the PP estimator is biased and may inflate Type 1 error and/or change power. We also propose a tipping point sensitivity analysis to evaluate the robustness of the current PP estimator in testing equivalence when the sensitivity parameters deviate from the three identified conditions, but stay within a clinically meaningful range. Our work is the first causal equivalence assessment in equivalence studies with intercurrent events.