Staging and clean room: Constructs designed to facilitate transparency and reduce bias in comparative analyses of real-world data.

PURPOSE: We describe constructs designed to protect the integrity of the results from comparative analyses using real-world data (RWD): staging and clean room. METHODS: Staging involves performing sequential preliminary analyses and evaluating the population size available and potential bias before conducting comparative analyses. A clean room involves restricted access to data and preliminary results, policies governing exploratory analyses and protocol deviations, and audit trail. These constructs are intended to allow decisions about protocol deviations, such as changes to design or model specification, to be made without knowledge of how they might affect subsequent analyses. We describe an example for implementing staging with a clean room. RESULTS: Stage 1 may involve selecting a data source, developing and registering a protocol, establishing a clean room, and applying inclusion/exclusion criteria. Stage 2 may involve attempting to achieve covariate balance, often through propensity score models. Stage 3 may involve evaluating the presence of residual confounding using negative control outcomes. After each stage, check points may be implemented when a team of statisticians, epidemiologists and clinicians masked to how their decisions may affect study outcomes, reviews the results. This review team may be tasked with making recommendations for protocol deviations to address study precision or bias. They may recommend proceeding to the next stage, conducting additional analyses to address bias, or terminating the study. Stage 4 may involve conducting the comparative analyses. CONCLUSIONS: The staging and clean room constructs are intended to protect the integrity and enhance confidence in the results of analyses of RWD.

Ancestral diversity in lipoprotein(a) studies helps address evidence gaps.

INTRODUCTION: The independent and causal cardiovascular disease risk factor lipoprotein(a) (Lp(a)) is elevated in >1.5 billion individuals worldwide, but studies have prioritised European populations. METHODS: Here, we examined how ancestrally diverse studies could clarify Lp(a)'s genetic architecture, inform efforts examining application of Lp(a) polygenic risk scores (PRS), enable causal inference and identify unexpected Lp(a) phenotypic effects using data from African (n=25 208), East Asian (n=2895), European (n=362 558), South Asian (n=8192) and Hispanic/Latino (n=8946) populations. RESULTS: Fourteen genome-wide significant loci with numerous population specific signals of large effect were identified that enabled construction of Lp(a) PRS of moderate (R2=15% in East Asians) to high (R2=50% in Europeans) accuracy. For all populations, PRS showed promise as a 'rule out' for elevated Lp(a) because certainty of assignment to the low-risk threshold was high (88.0%-99.9%) across PRS thresholds (80th-99th percentile). Causal effects of increased Lp(a) with increased glycated haemoglobin were estimated for Europeans (p value =1.4×10-6), although inverse effects in Africans and East Asians suggested the potential for heterogeneous causal effects. Finally, Hispanic/Latinos were the only population in which known associations with coronary atherosclerosis and ischaemic heart disease were identified in external testing of Lp(a) PRS phenotypic effects. CONCLUSIONS: Our results emphasise the merits of prioritising ancestral diversity when addressing Lp(a) evidence gaps.

Retrospective comparison of traditional and artificial intelligence-based heart failure phenotyping in a US health system to enable real-world evidence.

OBJECTIVE: Quantitatively evaluate the quality of data underlying real-world evidence (RWE) in heart failure (HF). DESIGN: Retrospective comparison of accuracy in identifying patients with HF and phenotypic information was made using traditional (ie, structured query language applied to structured electronic health record (EHR) data) and advanced (ie, artificial intelligence (AI) applied to unstructured EHR data) RWE approaches. The performance of each approach was measured by the harmonic mean of precision and recall (F1 score) using manual annotation of medical records as a reference standard. SETTING: EHR data from a large academic healthcare system in North America between 2015 and 2019, with an expected catchment of approximately 5 00 000 patients. POPULATION: 4288 encounters for 1155 patients aged 18-85 years, with 472 patients identified as having HF. OUTCOME MEASURES: HF and associated concepts, such as comorbidities, left ventricular ejection fraction, and selected medications. RESULTS: The average F1 scores across 19 HF-specific concepts were 49.0% and 94.1% for the traditional and advanced approaches, respectively (p<0.001 for all concepts with available data). The absolute difference in F1 score between approaches was 45.1% (98.1% relative increase in F1 score using the advanced approach). The advanced approach achieved superior F1 scores for HF presence, phenotype and associated comorbidities. Some phenotypes, such as HF with preserved ejection fraction, revealed dramatic differences in extraction accuracy based on technology applied, with a 4.9% F1 score when using natural language processing (NLP) alone and a 91.0% F1 score when using NLP plus AI-based inference. CONCLUSIONS: A traditional RWE generation approach resulted in low data quality in patients with HF. While an advanced approach demonstrated high accuracy, the results varied dramatically based on extraction techniques. For future studies, advanced approaches and accuracy measurement may be required to ensure data are fit-for-purpose.

Lipoprotein(a) and the Risk for Recurrent Atherosclerotic Cardiovascular Events Among Adults With CKD: The Chronic Renal Insufficiency Cohort (CRIC) Study.

Rationale & Objective: Many adults with chronic kidney disease (CKD) and atherosclerotic cardiovascular disease (ASCVD) have high lipoprotein(a) levels. It is unclear whether high lipoprotein(a) levels confer an increased risk for recurrent ASCVD events in this population. We estimated the risk for recurrent ASCVD events associated with lipoprotein(a) in adults with CKD and prevalent ASCVD. Study Design: Observational cohort study. Setting & Participants: We included 1,439 adults with CKD and prevalent ASCVD not on dialysis enrolled in the Chronic Renal Insufficiency Cohort study between 2003 and 2008. Exposure: Baseline lipoprotein(a) mass concentration, measured using a latex-enhanced immunoturbidimetric assay. Outcomes: Recurrent ASCVD events (primary outcome), kidney failure, and death (exploratory outcomes) through 2019. Analytical Approach: We used Cox proportional-hazards regression models to estimate adjusted HR (aHRs) and 95% CIs. Results: Among participants included in the current analysis (mean age 61.6 years, median lipoprotein(a) 29.4 mg/dL [25th-75th percentiles 9.9-70.9 mg/dL]), 641 had a recurrent ASCVD event, 510 developed kidney failure, and 845 died over a median follow-up of 6.6 years. The aHR for ASCVD events associated with 1 standard deviation (SD) higher log-transformed lipoprotein(a) was 1.04 (95% CI, 0.95-1.15). In subgroup analyses, 1 SD higher log-lipoprotein(a) was associated with an increased risk for ASCVD events in participants without diabetes (aHR, 1.23; 95% CI, 1.02-1.48), but there was no evidence of an association among those with diabetes (aHR, 0.99; 95% CI, 0.88-1.10, P comparing aHRs = 0.031). The aHR associated with 1 SD higher log-lipoprotein(a) in the overall study population was 1.16 (95% CI, 1.04-1.28) for kidney failure and 1.02 (95% CI, 0.94-1.11) for death. Limitations: Lipoprotein(a) was not available in molar concentration. Conclusions: Lipoprotein(a) was not associated with the risk for recurrent ASCVD events in adults with CKD, although it was associated with a risk for kidney failure.

Lipoprotein(a) and the Risk for Coronary Heart Disease and Ischemic Stroke Events Among Black and White Adults With Cardiovascular Disease.

Background It is unclear whether lipoprotein(a) is associated with coronary heart disease (CHD) and ischemic stroke events in White and Black adults with atherosclerotic cardiovascular disease (ASCVD). Methods and Results We conducted a case-cohort analysis, including Black and White REGARDS (Reasons for Geographic and Racial Differences in Stroke) study participants ≥45 years of age with prevalent ASCVD (ie, CHD or stroke) at baseline between 2003 and 2007. Baseline lipoprotein(a) molar concentration was measured in participants with ASCVD who experienced a CHD event by December 2017 (n=1166) or an ischemic stroke by September 2019 (n=492) and in a random subcohort of participants with prevalent ASCVD (n=1948). The hazard ratio (HR) for CHD events per 1 SD (1.5 units) higher log-transformed lipoprotein(a) was 1.26 (95% CI, 1.02-1.56) among Black participants and 1.16 (95% CI, 1.02-1.31) among White participants (P value comparing HRs, 0.485). The HR for CHD events per 1 SD higher log-lipoprotein(a) within subgroups with hs-CRP (high-sensitivity C-reactive protein) ≥2 and <2 mg/L was 1.31 (95% CI, 0.99-1.73) and 1.23 (95% CI, 0.85-1.80), respectively (P value comparing HRs, 0.836), among Black participants, and 1.07 (95% CI, 0.91-1.27) and 1.36 (95% CI, 1.10-1.70), respectively (P value comparing HRs, 0.088), among White participants. There was no evidence that the association between lipoprotein(a) and CHD events differed by statin use. There was no evidence of an association between lipoprotein(a) and ischemic stroke events among Black or White participants. Conclusions Higher lipoprotein(a) levels were associated with an increased risk for CHD events in Black and White adults with ASCVD.

Race/ethnic and sex differences in the initiation of non-statin lipid-lowering medication following myocardial infarction.

BACKGROUND: Adults with atherosclerotic cardiovascular disease (ASCVD) at very high-risk for recurrent events who have low-density lipoprotein cholesterol ≥ 70 mg/dL despite maximally-tolerated statin therapy are recommended to initiate ezetimibe or a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor. OBJECTIVE: Compare the initiation of ezetimibe and a PCSK9 inhibitor after a myocardial infarction (MI) among very high-risk ASCVD patients by race/ethnicity and sex. METHODS: We analyzed data from 374,786 adults ≥ 66 years of age with Medicare fee-for-service coverage who had an MI between July 1, 2015 and December 31, 2018, were not taking ezetimibe or a PCSK9 inhibitor, and had very high-risk ASCVD defined by the 2018 American Heart Association/American College of Cardiology multi-society cholesterol guideline. Pharmacy claims through December 31, 2018 were used to determine ezetimibe and PCSK9 inhibitor initiation. RESULTS: Overall, 6980 (1.9%) beneficiaries initiated ezetimibe, and 1433 (0.4%) initiated a PCSK9 inhibitor. Adjusted hazard ratios (aHR) for ezetimibe initiation among non-Hispanic Black, Hispanic, and Asian versus non-Hispanic White beneficiaries were 0.77 (95% confidence interval [95%CI]: 0.70-0.86), 0.92 (95%CI: 0.76-1.11) and 0.73 (95%CI: 0.59-0.89), respectively. Compared to non-Hispanic White beneficiaries, the aHRs for PCSK9 inhibitor initiation were 0.63 (95%CI: 0.48-0.81) among non-Hispanic Black, 0.70 (95%CI: 0.43-1.13) among Hispanic, and 0.93 (95%CI: 0.62-1.39) among Asian beneficiaries. The aHRs for ezetimibe and PCSK9 inhibitor initiation comparing women to men were 1.11 (95%CI: 1.06-1.17) and 1.13 (95%CI: 1.01-1.25), respectively. CONCLUSION: There are race/ethnic and sex disparities in the initiation of ezetimibe and a PCSK9 inhibitor following MI among very high-risk ASCVD patients.

Factors Associated With PCSK9 Inhibitor Initiation Among US Veterans.

Background Few adults at high risk for atherosclerotic cardiovascular disease events use a PCSK9i (proprotein convertase subtilisin/kexin type 9 inhibitor). Methods and Results Using data from the US Veterans Health Administration, we identified veterans who initiated a PCSK9i between January 2018 and December 2019, matched 1:4 to veterans who did not initiate this medication over this time period (case-cohort study). Two cohorts of veterans were analyzed: (1) atherosclerotic cardiovascular disease, with a most recent low-density lipoprotein cholesterol (LDL-C) ≥70 mg/dL; and (2) severe hypercholesterolemia (ie, familial hypercholesterolemia or any prior LDL-C ≥190 mg/dL, with most recent LDL-C ≥100 mg/dL). Conditional logistic regression was used to analyze factors associated with PCSK9i initiation, adjusting for all factors, simultaneously. There were 2394 initiators and 9576 noninitiators in the atherosclerotic cardiovascular disease cohort (median LDL-C, 141 and 96 mg/dL, respectively; P<0.001). Factors associated with a higher likelihood of PCSK9i initiation included age 65 to <75 versus <65 years, highest versus lowest quartile of median area-level income, familial hypercholesterolemia, former statin use, and current ezetimibe use. PCSK9i initiation was lower among veterans of a race/ethnicity other than non-Hispanic White. There were 245 initiators and 980 noninitiators in the severe hypercholesterolemia cohort (median LDL-C, 183 and 151 mg/dL, respectively; P<0.001). Age ≥75 versus <65 years, history of chronic kidney disease, former statin use, and current ezetimibe use were associated with a higher likelihood of PCSK9i initiation. Conclusions Several patient-level factors, including age, sex, and race/ethnicity, were significantly associated with PCSK9i initiation, suggesting an unmet treatment need in several patient groups.

Long-Term Levels of LDL-C and Cognitive Function: The CARDIA Study.

OBJECTIVES: It is uncertain if long-term levels of low-density lipoprotein-cholesterol (LDL-C) affect cognition in middle age. We examined the association of LDL-C levels over 25 years with cognitive function in a prospective cohort of black and white US adults. METHODS: Lipids were measured at baseline (1985-1986; age: 18-30 years) and at serial examinations conducted over 25 years. Time-averaged cumulative LDL-C was calculated using the area under the curve for 3,328 participants with ≥3 LDL-C measurements and a cognitive function assessment. Cognitive function was assessed at the Year 25 examination with the Digit Symbol Substitution Test [DSST], Rey Auditory Visual Learning Test [RAVLT], and Stroop Test. A brain magnetic resonance imaging (MRI) sub-study (N = 707) was also completed at Year 25 to assess abnormal white matter tissue volume (AWMV) and gray matter cerebral blood flow volume (GM-CBFV) as secondary outcomes. RESULTS: There were 15.6%, 32.9%, 28.9%, and 22.6% participants with time-averaged cumulative LDL-C <100 mg/dL, 101-129 mg/dL, 130-159 mg/dL, and ≥160 mg/dL, respectively. Standardized differences in all cognitive function test scores ranged from 0.16 SD lower to 0.09 SD higher across time-averaged LDL-C categories in comparison to those with LDL-C < 100 mg/dL. After covariate adjustment, participants with higher versus lower time-averaged LDL-C had a lower RAVLT score (p-trend = 0.02) but no differences were present for DSST, Stroop Test, AWMV, or GM-CBFV. CONCLUSION: Cumulative LDL-C was associated with small differences in memory, as assessed by RAVLT scores, but not other cognitive or brain MRI measures over 25 years of follow-up.

Death, long-term nursing home placement, and impoverishment after recurrent myocardial infarction.

Study objective: To determine whether recurrent myocardial infarction (MI) is associated with increased risk of mortality, long-term nursing home placement, and impoverishment. Design: Retrospective cohort study. Setting: United States Medicare program. Participants: Individuals age > 65 years with recurrent MI hospitalizations (n = 228,826) between January 1, 2007 and June 30, 2017 and controls with initial but not recurrent MI (n = 915,304). Main outcome measures: Death, nursing home placement, and impoverishment (Medicaid enrollment or subsidies for low-income and -resource individuals) through December 31, 2017. Results: In the recurrent MI and control cohorts, 47% and 41% of individuals were age > 80 years, respectively, and 56% of both cohorts were women. After 1 year, 48% of the recurrent MI cohort and 16% of the control cohort died, 9% and 7% experienced nursing home placement, and 4% and 2% experienced impoverishment. Multivariable-adjusted hazard ratios (95% confidence intervals) comparing the recurrent MI and control cohorts were 2.04 (2.03-2.06) for death, 0.89 (0.88-0.91) for nursing home placement, and 1.32 (1.28-1.36) for impoverishment. Conclusions: Older US adults with recurrent MI had higher risk of death and impoverishment than controls who had experienced an initial MI. Unadjusted, recurrent MI was associated with higher risk of nursing home placement; however, after adjusting for sociodemographic characteristics and comorbidities, individuals with recurrent MI had slightly lower risk of nursing home placement. Preventing recurrent MI may also reduce the risk of death and impoverishment among older US adults.

Atherosclerotic Risk and Statin Use Among Patients With Peripheral Artery Disease.

BACKGROUND: Peripheral artery disease (PAD) is associated with increased risk for atherosclerotic cardiovascular disease (ASCVD) events. OBJECTIVES: The goal of this study was to compare the risk for ASCVD events and the use of statins among patients with PAD versus those with coronary heart disease (CHD) or cerebrovascular disease. METHODS: The authors conducted a retrospective cohort study of adults age ≥19 years with commercial or Medicare health insurance who had a history of PAD, CHD, or cerebrovascular disease on December 31, 2014. Patients were followed for ASCVD events comprising CHD, cerebrovascular disease, and PAD events until December 31, 2017. RESULTS: Among 943,232 patients included in the analysis, the age-standardized ASCVD event rate per 1,000 person-years for those with a history of 1, 2, and 3 conditions including PAD, CHD, and cerebrovascular disease was 40.8 (95% confidence interval [CI]: 40.3 to 41.3), 68.9 (95% CI: 67.9 to 70.0), and 119.5 (95% CI: 117.0 to 122.0), respectively. The ASCVD event rate among patients with PAD only, CHD only, and cerebrovascular disease only was 34.7 (95% CI: 33.2 to 36.2), 42.2 (95% CI: 41.5 to 42.8), and 38.9 (95% CI: 37.6 to 40.1), respectively. Among patients with PAD and CHD, with PAD and cerebrovascular disease, and with CHD and cerebrovascular disease, the ASCVD event rate was 72.8 (95% CI: 71.0 to 74.7), 63.9 (95% CI: 60.6 to 67.4), and 67.9 (95% CI: 66.4 to 69.3), respectively. Statin use was lower in patients with PAD only (33.9%) versus those with cerebrovascular disease only (43.0%) or CHD only (51.7%). CONCLUSIONS: Despite having high risk for ASCVD events, patients with PAD were less likely to be taking a statin versus those with CHD or cerebrovascular disease. ASCVD risk-reduction interventions including statin therapy in patients with PAD are warranted.

Designing, Conducting, Monitoring, and Analyzing Data from Pragmatic Randomized Clinical Trials: Proceedings from a Multi-stakeholder Think Tank Meeting.

In late 2018, the Food and Drug Administration (FDA) outlined a framework for evaluating the possible use of real-world evidence (RWE) to support regulatory decision-making. This framework was created to facilitate studies that would generate high-quality RWE, including pragmatic clinical trials (PCTs), which are randomized trials designed to inform clinical or policy decisions by assessing the real-world effectiveness of an intervention. There is general agreement among experts that the use of existing healthcare and patient-generated data holds promise for making randomized trials more efficient, less costly, and more generalizable. Yet the benefits of relying on real-world data sources must be weighed against difficulties with ensuring data integrity and completeness. Additionally, appropriately monitoring patient safety in randomized trials of new drugs using healthcare system data that might not be available in real time can be quite difficult. Recognizing that these and other concerns are critical to the development and acceptability of PCTs, a group of stakeholders from academia, industry, professional organizations, regulatory bodies, government agencies, and patient advocates discussed a path forward for PCT growth and sustainability at a think tank meeting entitled "Monitoring and Analyzing Data from Pragmatic Streamlined Randomized Clinical Trials," which took place in January 2019 (Washington, DC). The goals of this meeting were to: (1) evaluate study design and methodological options specific to PCTs that have the potential to yield high-quality evidence; (2) discuss best practices to ensure data quality in PCTs; and (3) identify appropriate methods for study monitoring. Proceedings from the think tank meeting are summarized in this manuscript.

Use of Low-Density Lipoprotein-Lowering Therapies Before and After PCSK9 Inhibitor Initiation.

Background Proprotein convertase subtilisin/kexin type 9 inhibitors (PCSK9i) are used to reduce low-density lipoprotein (LDL) cholesterol. PCSK9i use after initiation, as well as persistence with or alterations to other LDL-lowering therapy after PCSK9i initiation, is not well understood. Methods and Results We conducted a retrospective study of alirocumab or evolocumab (PCSK9i) new users from July 2015 to December 2017 in the MarketScan Early View database of US commercial insurance beneficiaries. We determined the prevalence of PCSK9i interruption (≥30-day gap in supply) and LDL-lowering therapy use in the year after PCSK9i initiation. The average age of 6151 patients initiating PCSK9i therapy was 63 years, 44.4% were women, and 76.8% had atherosclerotic cardiovascular disease. Overall, 52.2% (95% CI, 50.8%-53.7%) of patients had an interruption in PCSK9i therapy in the first year after treatment initiation and 62.5% remained on PCSK9i therapy at 1-year postinitiation. Also, 27.7% of patients were taking a statin at the time of PCSK9i initiation, with only 22.4% on statin therapy at 1 year after PCSK9i initiation. Ezetimibe use decreased from 20.9% at the time of PCSK9i initiation to 12.0% a year later. By 1 year after PCSK9i initiation, 44.0% of patients had experienced an interruption in all LDL-lowering therapies, and 26.6% were no longer on any LDL-lowering therapies. Conclusions After PCSK9i initiation, statins were often discontinued, whereas more than half of patients experienced an interruption in PCSK9i therapy. These results suggest that many new PCSK9i users may remain at high risk for cardiovascular events because of interruptions in LDL-lowering therapy.

Residual risk for coronary heart disease events and mortality despite intensive medical management after myocardial infarction.

BACKGROUND: High-intensity statins, beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and antiplatelet agents (ie, intensive medical management) reduce coronary heart disease (CHD) risk after myocardial infarction (MI). OBJECTIVE: The objective of the study was to determine the risk of CHD events or death despite receiving intensive medical management after MI. METHODS: We studied 16,853 United States adults with health insurance in the MarketScan and Medicare databases who underwent percutaneous coronary intervention while hospitalized for MI between January 1, 2014 and June 30, 2015 and received intensive medical management within 30 days after hospital discharge. MI, CHD, and all-cause mortality rates from 30 days after hospital discharge through December 31, 2015 were compared with 67,412 individuals in each of three groups: (1) the general MarketScan and Medicare populations, (2) with diabetes, and (3) with a CHD history. RESULTS: Among beneficiaries intensively medically managed after their MI, recurrent MI, CHD events, and all-cause mortality rates were 47.1, 72.0, and 57.5 per 1000 person-years, respectively. The multivariable-adjusted hazard ratio (95% CI) comparing intensively medically managed beneficiaries after MI to the general population, those with diabetes, and those with a history of CHD were 8.54 (7.52-9.70), 7.40 (6.61-8.28), and 5.45 (4.92-6.05), respectively, for recurrent MI; 7.82 (7.07-8.64), 6.27 (5.74-6.86), and 4.45 (4.10-4.82), respectively, for CHD events; and 1.15 (1.05-1.25), 1.05 (0.97-1.14), and 1.06 (0.97-1.15), respectively, for all-cause mortality. CONCLUSION: Substantial residual risk for MI and CHD events remains despite intensive medical management after MI.

Excess Risk for Atherosclerotic Cardiovascular Outcomes Among US Adults With HIV in the Current Era.

Background In the 2000s, adults with HIV had a higher risk for atherosclerotic cardiovascular disease (ASCVD) compared with those without HIV. There is uncertainty if this excess risk still exists in the United States given changes in antiretroviral therapies and increased statin use. Methods and Results We compared the risk for ASCVD events between US adults aged ≥19 years with and without HIV who had commercial or supplemental Medicare health insurance between January 1, 2011, and December 31, 2016. Beneficiaries with HIV (n=82 426) were frequency matched 1:4 on age, sex, and calendar year to those without HIV (n=329 704). Beneficiaries with and without HIV were followed up through December 31, 2016, for ASCVD events, including myocardial infarction, stroke, and lower extremity artery disease hospitalizations. Most beneficiaries were aged <55 years (79%) and men (84%). Over a median follow-up of 1.6 years (maximum, 6 years), there were 3287 ASCVD events, 2190 myocardial infarctions, 891 strokes, and 322 lower extremity artery disease events. The rate per 1000 person-years among beneficiaries with and without HIV was 5.53 and 3.49 for ASCVD, respectively, 3.58 and 2.34 for myocardial infarction, respectively, 1.49 and 0.94 for stroke, respectively, and 0.65 and 0.31 for lower extremity artery disease hospitalizations, respectively. The multivariable-adjusted hazard ratio (95% CI) for ASCVD, myocardial infarction, stroke, and lower extremity artery disease hospitalizations comparing beneficiaries with versus without HIV was 1.29 (1.18-1.40), 1.26 (1.13-1.39), 1.30 (1.11-1.52), and 1.46 (1.11-1.92), respectively. Conclusions Adults with HIV in the United States continue to have a higher ASCVD risk compared with their counterparts without HIV.

Trends in statin utilization among adults with severe peripheral artery disease including critical limb ischemia in an integrated healthcare delivery system.

Evidence suggests that statin therapy in patients with peripheral artery disease (PAD) is beneficial yet use remains suboptimal. We examined trends in statin use, intensity, and discontinuation among adults aged ⩾ 40 years with incident severe PAD and a subset with critical limb ischemia (CLI) between 2002 and 2015 within an integrated healthcare delivery system. Discontinuation of statin therapy was defined as the first 90-day gap in treatment within 1 year following PAD diagnosis. We identified 11,059 patients with incident severe PAD: 31.1% (n = 3442) with CLI and 68.9% (n = 7617) without CLI. Mean (SD) age was 68.6 (11.3) years, 60.5% were male, 54.2% white, 23.2% Hispanic, and 16.2% black. Statin use in the year before diagnosis increased from 50.4% in 2002 to 66.0% in 2015 (CLI: 43.7% to 68.0%; without CLI: 53.1% to 64.2%, respectively). The proportion of patients on high-intensity statins increased from 7.3% in 2002 to 41.9% in 2015 (CLI: 7.2% to 39.4%; without CLI: 7.4% to 44.2%, respectively). Of the 40.5% (n = 4481) who were not on a statin in the year before diagnosis, 13.5% (n = 607) newly initiated therapy within 1 month (CLI: 10.1% (n = 150); without CLI: 15.3% (n = 457)). Following diagnosis, 12.5% (n = 660) discontinued statin therapy within 1 year (CLI: 15.5% (n = 202); without CLI: 11.5% (n = 458)). Although use of statins increased from 2002 to 2015, a substantial proportion of the overall PAD and CLI subpopulation remained untreated with statins, representing a significant treatment gap in a population at high risk for cardiovascular events and adverse limb outcomes.

PCSK9 loss-of-function variants and Lp(a) phenotypes among black US adults.

The pharmacologic inhibition of proprotein convertase subtilisin-kexin type 9 (PCSK9) lowers lipoprotein (a) [Lp(a)] concentrations. However, the impact of genetic PCSK9 loss-of-function variants (LOFVs) on Lp(a) is uncertain. We determined the association of PCSK9 LOFVs with Lp(a) measures among black adults. Genotyping for PCSK9 LOFVs was conducted in 10,196 black Reasons for Geographic and Racial Differences in Stroke study participants. Among 241 participants with and 723 randomly selected participants without PCSK9 LOFVs, Lp(a) concentations, apo(a) kringle IV (KIV) repeats (a proxy for isoform size), and oxidized phospholipid (OxPL) apoB levels were measured using validated methods. Median Lp(a) concentrations among participants with and without PCSK9 LOFVs were 63.2 and 80.4 nmol/l, respectively (P = 0.016). After adjusting for age, sex, estimated glomerular filtration rate, LDL cholesterol, and statin use, participants with versus without a PCSK9 LOFV had a lower median Lp(a) concentration [Δ = -18.8 nmol/l (95% CI: -34.2, -3.3)]. Median apo(a) isoform sizes were 24 and 23 KIV repeats (P = 0.12) among participants with and without PCSK9 LOFVs, respectively [Δ = 1.1 (95% CI: 0.2, 2.0) after adjustment]. Median OxPL-apoB levels among participants with and without PCSK9 LOFVs were 3.4 and 4.1 nM (P = 0.20), respectively [Δ = -1.2 nM (95% CI -2.4, -0.04) after adjustment]. Among black adults, PCSK9 LOFVs were associated with lower Lp(a) concentration and OxPL-apoB levels.

Lipid-Lowering Prescription Patterns in Patients With Diabetes Mellitus or Cardiovascular Disease.

The purpose of this study is to describe lipid-lowering therapy (LLT) prescriptions and low-density lipoprotein cholesterol (LDL-C) monitoring in patients with diabetes mellitus (DM) with or without concomitant cardiovascular disease (CVD). Olmsted County, Minnesota residents with a first-ever diagnosis of DM or CVD (ischemic stroke/transient ischemic attack, myocardial infarction, unstable angina pectoris, or revascularization procedure) between 2005 and 2012 were classified as having DM only, CVD only, or CVD + DM. All LLT prescriptions and LDL-C measurements were obtained for 2 years after diagnosis. A total of 4,186, 2,368, and 724 patients had DM, CVD, and CVD + DM, respectively. Rates of LDL-C measurement were 1.31, 1.66, and 1.88 per person-year and 14%, 32%, and 42% of LDL-C measurements were <70 mg/dl in those with DM, CVD, and CVD + DM. Within 3 months after diagnosis, 47%, 71%, and 78% of patients with DM, CVD, and CVD + DM were prescribed LLT. Most prescriptions were for moderate-intensity statins. Under one-fifth of patients with CVD and CVD + DM were prescribed high-intensity statins. Predictors of high-intensity statin prescriptions included male sex, having CVD or CVD + DM, increasing LDL-C, and LDL-C measured more recently (2012 to 2014 vs before 2012). In conclusion, a large proportion of patients at high CVD risk are not adequately treated with LLT. Despite often being considered a risk equivalent, patients with DM have substantially lower rates of LLT prescriptions and lesser controlled LDL-C than those with CVD or CVD + DM.

Real world evidence in cardiovascular medicine: ensuring data validity in electronic health record-based studies.

OBJECTIVE: With growing availability of digital health data and technology, health-related studies are increasingly augmented or implemented using real world data (RWD). Recent federal initiatives promote the use of RWD to make clinical assertions that influence regulatory decision-making. Our objective was to determine whether traditional real world evidence (RWE) techniques in cardiovascular medicine achieve accuracy sufficient for credible clinical assertions, also known as "regulatory-grade" RWE. DESIGN: Retrospective observational study using electronic health records (EHR), 2010-2016. METHODS: A predefined set of clinical concepts was extracted from EHR structured (EHR-S) and unstructured (EHR-U) data using traditional query techniques and artificial intelligence (AI) technologies, respectively. Performance was evaluated against manually annotated cohorts using standard metrics. Accuracy was compared to pre-defined criteria for regulatory-grade. Differences in accuracy were compared using Chi-square test. RESULTS: The dataset included 10 840 clinical notes. Individual concept occurrence ranged from 194 for coronary artery bypass graft to 4502 for diabetes mellitus. In EHR-S, average recall and precision were 51.7% and 98.3%, respectively and 95.5% and 95.3% in EHR-U, respectively. For each clinical concept, EHR-S accuracy was below regulatory-grade, while EHR-U met or exceeded criteria, with the exception of medications. CONCLUSIONS: Identifying an appropriate RWE approach is dependent on cohorts studied and accuracy required. In this study, recall varied greatly between EHR-S and EHR-U. Overall, EHR-S did not meet regulatory grade criteria, while EHR-U did. These results suggest that recall should be routinely measured in EHR-based studes intended for regulatory use. Furthermore, advanced data and technologies may be required to achieve regulatory grade results.

Association of Region and Hospital and Patient Characteristics With Use of High-Intensity Statins After Myocardial Infarction Among Medicare Beneficiaries.

Importance: High-intensity statin use after myocardial infarction (MI) varies by patient characteristics, but little is known about differences in use by hospital or region. Objective: To explore the relative strength of associations of region and hospital and patient characteristics with high-intensity statin use after MI. Design, Setting, and Participants: This retrospective cohort analysis used Medicare administrative claims and enrollment data to evaluate fee-for-service Medicare beneficiaries 66 years or older who were hospitalized for MI from January 1, 2011, through June 30, 2015, with a statin prescription claim within 30 days of discharge. Data were analyzed from January 4, 2017, through May 12, 2019. Exposures: Beneficiary characteristics were abstracted from Medicare data. Hospital characteristics were obtained from the 2014 American Hospital Association Survey and Hospital Compare quality metrics. Nine regions were defined according to the US Census. Main Outcomes and Measures: Intensity of the first statin claim after discharge characterized as high (atorvastatin calcium, 40-80 mg, or rosuvastatin calcium, 20-40 mg/d) vs low to moderate (all other statin types and doses). Trends in high-intensity statins were examined from 2011 through 2015. Associations of region and beneficiary and hospital characteristics with high-intensity statin use from January 1, 2014, to June 15, 2015, were examined using Poisson distribution mixed models. Results: Among the 139 643 fee-for-service beneficiaries included (69 968 men [50.1%] and 69 675 women [49.9%]; mean [SD] age, 76.7 [7.5] years), high-intensity statin use overall increased from 23.4% in 2011 to 55.6% in 2015, but treatment gaps persisted across regions. In models considering region and beneficiary and hospital characteristics, region was the strongest correlate of high-intensity statin use, with 66% higher use in New England than in the West South Central region (risk ratio [RR], 1.66; 95% CI, 1.47-1.87). Hospital size of at least 500 beds (RR, 1.15; 95% CI, 1.07-1.23), medical school affiliation (RR, 1.11; 95% CI, 1.05-1.17), male sex (RR, 1.10; 95% CI, 1.07-1.13), and patient receipt of a stent (RR, 1.35; 95% CI, 1.31-1.39) were associated with greater high-intensity statin use. For-profit hospital ownership, patient age older than 75 years, prior coronary disease, and other comorbidities were associated with lower use. Conclusions and Relevance: This study's findings suggest that geographic region is the strongest correlate of high-intensity statin use after MI, leading to large treatment disparities.

Familial Hypercholesterolemia Among Young Adults With Myocardial Infarction.

BACKGROUND: There are limited data on the prevalence and treatment of familial hypercholesterolemia (FH) among U.S. adults who experience a myocardial infarction (MI) at a young age. OBJECTIVES: This study aimed to evaluate the prevalence of clinically defined FH and examine the rates of statin utilization and low-density lipoprotein cholesterol (LDL-C) achieved 1-year post MI. METHODS: The YOUNG-MI registry is a retrospective cohort study that includes patients who experience an MI at or below age 50 years between 2000 and 2016 at 2 academic centers. Probable or definite FH was defined by the Dutch Lipid Clinic criteria. Outcomes included the proportion of patients classified as probable or definite FH, use of lipid-lowering therapy, and LDL-C achieved 1-year post MI. RESULTS: The cohort consisted of 1,996 adults with a median age of 45 years; 19% were women, and 54% had ST-segment elevation MI. Probable/definite FH was present in 180 (9%) of whom 42.8% were not on statins prior to their MI. Of the 1,966 patients surviving until hospital discharge, 89.4% of FH patients and 89.9% of non-FH patients were discharged on statin therapy (p = 0.82). Among FH patients, 63.3% were discharged on high-intensity statin compared with 48.4% for non-FH patients (p < 0.001). At 1-year follow-up, the percent reduction in LDL-C among FH patients was -44.4% compared with -34.5% (p = 0.006) in non-FH patients. The proportion of patients with LDL-C ≥70 mg/dl was higher among FH patients (82.2%) compared with non-FH patients (64.5%; p < 0.001). CONCLUSIONS: Clinically defined FH was present in nearly 1 of 10 patients with MI at a young age. Only two-thirds of FH patients were discharged on high-intensity statin therapy, and the vast majority had elevated LDL-C at 1 year. These findings reinforce the need for more aggressive lipid-lowering therapy in young FH and non-FH patients post-MI.