The distillation method: A novel approach for analyzing randomized trials when exposure to the intervention is diluted.

OBJECTIVE: To introduce a novel analytical approach for randomized controlled trials that are underpowered because of low participant enrollment or engagement. DATA SOURCES: Reanalysis of data for 805 patients randomized as part of a pilot complex care intervention in 2015-2016 in a large delivery system. In the pilot randomized trial, only 64.6% of patients assigned to the intervention group participated. STUDY DESIGN: A case study and simulation. The "Distillation Method" capitalizes on the frequently observed correlation between the probability of subjects' participation or engagement in the intervention and the magnitude of benefit they experience. The novel method involves three stages: first, it uses baseline covariates to generate predicted probabilities of participation. Next, these are used to produce nested subsamples of the randomized intervention and control groups that are more concentrated with subjects who were likely to participate/engage. Finally, for the outcomes of interest, standard statistical methods are used to re-evaluate intervention effectiveness in these concentrated subsets. DATA EXTRACTION METHODS: We assembled secondary data on patients who were randomized to the pilot intervention for one year prior to randomization and two follow-up years. Data included program enrollment status, membership data, demographics, utilization, costs, and clinical data. PRINCIPAL FINDINGS: Using baseline covariates only, Generalized Boosted Regression Models predicting program enrollment performed well (AUC 0.884). We then distilled the full randomized sample to increasing levels of concentration and reanalyzed program outcomes. We found statistically significant differences in outpatient utilization and emergency department utilization (both follow-up years), and in total costs (follow-up year two only) at select levels of population concentration. CONCLUSIONS: By offering an internally valid analytic framework, the Distillation Method can increase the power to detect effects by redefining the estimand to subpopulations with higher enrollment probabilities and stronger average treatment effects while maintaining the original randomization.

The natural history of large abdominal aortic aneurysms in patients without timely repair.

OBJECTIVE: Contemporary data on the natural history of large abdominal aortic aneurysms (AAAs) in patients undergoing delayed or no repair are lacking. In this study, we examine the impact of large AAA size on the incidence of rupture and mortality. METHODS: From a prospectively maintained aneurysm surveillance registry, patients with an unrepaired, large AAA (≥5.5 cm in men and ≥5.0 cm in women) at baseline (ie, index imaging) or who progressed to a large size from 2003 to 2017 were included, with follow-up through March 2020. Outcomes of interest obtained by manual chart review included rupture (confirmed by imaging/autopsy), probable rupture (timing/findings consistent with rupture without more likely cause of death), repair, reasons for either no or delayed (>1 year after diagnosis of large AAA) repair and total mortality. Cumulative incidence of rupture was calculated using a nonparametric cumulative incidence function, accounting for the competing events of death and aneurysm repair and was stratified by patient sex. RESULTS: Of the 3248 eligible patients (mean age, 83.6 ± 9.1 years; 71.2% male; 78.1% white; and 32.0% current smokers), 1423 (43.8%) had large AAAs at index imaging, and 1825 progressed to large AAAs during the follow-up period, with a mean time to qualifying size of 4.3 ± 3.4 years. In total, 2215 (68%) patients underwent repair, of which 332 were delayed >1 year; 1033 (32%) did not undergo repair. The most common reasons for delayed repair were discrepancy in AAA measurement between surgeon and radiologist (34%) and comorbidity (20%), whereas the most common reasons for no repair were patient preference (48%) and comorbidity (30%). Among patients with delayed repair (mean time to repair, 2.6 ± 1.8 years), nine (2.7%) developed symptomatic aneurysms, and an additional 11 (3.3%) ruptured. Of patients with no repair, 94 (9.1%) ruptured. The 3-year cumulative incidence of rupture was 3.4% for initial AAA size 5.0 to 5.4 cm (women only), 2.2% for 5.5 to 6.0 cm, 6.0% for 6.1 to 7.0 cm, and 18.4% for >7.0 cm. Women with AAA size 6.1 to 7.0 cm had a 3-year cumulative incidence of rupture of 12.8% (95% confidence interval, 7.5%-19.6%) compared with 4.5% (95% confidence interval, 3.0%-6.5%) in men (P = .002). CONCLUSIONS: In this large cohort of AAA registry patients over 17 years, annual rupture rates for large AAAs were lower than previously reported, with possible increased risk in women. Further analyses are ongoing to identify those at increased risk for aneurysm rupture and may provide targeted surveillance regimens and improve patient counseling.

Validation of natural language processing to determine the presence and size of abdominal aortic aneurysms in a large integrated health system.

OBJECTIVE: Previous studies of the natural history of abdominal aortic aneurysms (AAAs) have been limited by small cohort sizes or heterogeneous analyses of pooled data. By quickly and efficiently extracting imaging data from the health records, natural language processing (NLP) has the potential to substantially improve how we study and care for patients with AAAs. The aim of the present study was to test the ability of an NLP tool to accurately identify the presence or absence of AAAs and detect the maximal abdominal aortic diameter in a large dataset of imaging study reports. METHODS: Relevant imaging study reports (n = 230,660) from 2003 to 2017 were obtained for 32,778 patients followed up in a prospective aneurysm surveillance registry within a large, diverse, integrated healthcare system. A commercially available NLP algorithm was used to assess the presence of AAAs, confirm the absence of AAAs, and extract the maximal diameter of the abdominal aorta, if stated. A blinded expert manual review of 18,000 randomly selected imaging reports was used as the reference standard. The positive predictive value (PPV or precision), sensitivity (recall), and the kappa statistics were calculated. RESULTS: Of the randomly selected 18,000 studies that underwent expert manual review, 48.7% were positive for AAAs. In confirming the presence of an AAA, the interrater reliability of the NLP compared with the expert review showed a kappa value of 0.84 (95% confidence interval [CI], 0.83-0.85), with a PPV of 95% and sensitivity of 88.5%. The NLP algorithm showed similar results for confirming the absence of an AAA, with a kappa of 0.79 (95% CI, 0.799-0.80), PPV of 77.7%, and sensitivity of 91.9%. The kappa, PPV, and sensitivity of the NLP for correctly identifying the maximal aortic diameter was 0.88 (95% CI, 0.87-0.89), 88.8%, and 88.2% respectively. CONCLUSIONS: The use of NLP software can accurately analyze large volumes of radiology report data to detect AAA disease and assemble a contemporary aortic diameter-based cohort of patients for longitudinal analysis to guide surveillance, medical management, and operative decision making. It can also potentially be used to identify from the electronic medical records pre- and postoperative AAA patients "lost to follow-up," leverage human resources engaged in the ongoing surveillance of patients with AAAs, and facilitate the construction and implementation of AAA screening programs.

Preoperative Vision and Surgeon Volume as Predictors of Visual Outcomes after Cataract Surgery.

PURPOSE: To evaluate the relationship between preoperative vision and surgeon volume with visual outcomes after cataract surgery. DESIGN: Retrospective cohort study. PARTICIPANTS: Patients aged ≥18 years old enrolled in the Kaiser Permanente Southern California Health Plan who underwent cataract surgery by nontrainee surgeons. METHODS: Patients who underwent cataract surgery between January 1, 2013 and December 31, 2015, were included. A multivariate analysis using Generalized Additive Mixed Models was performed to determine the relationship between surgeon volume and postoperative visual acuity after controlling for patient age, preoperative visual acuity, history of diabetes, and history of diabetic retinopathy. Modeling was done for the relationship between preoperative vision and visual outcomes while controlling for surgeon volume, patient age, history of diabetes, and history of diabetic retinopathy. MAIN OUTCOME MEASURE: Absolute letter change and percentage of patients to achieve ≥5 Early Treatment Diabetic Retinopathy Study (ETDRS) letter gain postoperatively. RESULTS: There were 103 920 cataract surgeries performed by 136 surgeons included in this analysis. Patients whose surgeons performed <91.0 surgeries/year (95% confidence interval [CI], 61.1-139; P < 0.05) gained fewer letters postoperatively than the overall average, whereas those whose surgeons performed >91 but <227 surgeries/year (95% CI, 169-∞; P < 0.05) gained more letters than average. Although statistically significant, the difference between the lowest and highest performing surgeons was approximately 1.25 letters. Surgeons who performed <110 surgeries/year (95% CI, 81.7-149; P < 0.05) had fewer patients who gained ≥5 letters. Surgeons who performed >110 but <293 surgeries/year (95% CI, 232-∞; P < 0.05) were approximately 15% more likely to have patients who gained ≥5 letters. Patients with preoperative vision <74.7 letters (95% CI, 74.7-74.8; P < 0.05) and <75.8 letters (95% CI, 75.8-75.9; P < 0.05) gained more letters and were more likely to gain ≥5 letters postoperatively, respectively. CONCLUSIONS: Patients whose vision is approximately 20/32 or worse are more likely to have significant visual gains after cataract surgery. Although statistically significant differences exist in postoperative vision based on surgeon volume, these do not appear to be clinically meaningful. Overall, visual outcomes are functionally comparable across a wide range of surgeon volumes.

Synthesis and structures of Pb3O2(CH3COO)2·0.5H2O and Pb2O(HCOO)2: two corrosion products revisited.

Reactions of carboxylic acids with lead play an important role in the atmospheric corrosion of lead and lead-tin alloys. This is of particular concern for the preservation of lead-based cultural objects, including historic lead-tin alloy organ pipes. Two initial corrosion products, Pb(3)O(2)(CH(3)COO)(2)·0.5H(2)O (1) and Pb(2)O(HCOO)(2) (2), had been identified through powder diffraction fingerprints in the Powder Diffraction File, but their structures had never been determined. We have crystallized both compounds using hydrothermal solution conditions, and structures were determined using laboratory and synchrotron single-crystal X-ray diffraction data. Compound 1 crystallizes in P1, and 2 in Cccm. These compounds may be viewed as inorganic-organic networks containing single and double chains of edge-sharing Pb(4)O tetrahedra and have structural similarities to inorganic basic lead compounds. Bond valence sum analysis has been applied to the hemidirected lead coordination environments in each compound. Atmospheric exposure experiments contribute to understanding of the potential for conversion of these short-term corrosion products to hydrocerussite, Pb(3)(CO(3))(2)(OH)(2), previously identified as a long-term corrosion product on lead-rich objects. Each compound was also characterized by elemental analysis, thermogravimetric analysis and differential scanning calorimetry (TGA-DSC), and Raman spectroscopy.