The Oncology QCARD Initiative: Fostering efficient evaluation of initial real-world data proposals.

PURPOSE: The oncology quality, characterization, and assessment of real-world data (Oncology QCARD) Initiative was formed to develop a set of minimum study design and data elements needed to evaluate the fitness of the real-world data (RWD) source(s) proposed in an initial study concept as part of early interaction with scientific reviewers. METHODS: A multidisciplinary executive committee (EC) was established to guide the Oncology QCARD Initiative. The EC conducted a landscape review of published literature, guidances, and guidelines to evaluate relevant dimensions of data quality measurement. Guided by the review and informed by expert feedback, the Oncology QCARD Initial Protocol Characterization (IPC) provides a summary of minimum elements needed to adequately describe an initial clinical study concept that involves RWD and is intended to support decision-making. RESULTS: Fit-for-use data and fit-for-purpose design emerged as themes from the landscape analysis. Data that are fit-for-use are both relevant (sufficiently capturing exposure, outcomes, and covariates) and reliable (understanding data accrual and quality control and whether the data represent the underlying concepts they are intended to represent) to answer a specific research question. A fit-for-purpose design takes appropriate steps to ensure internal and external validity and allows for transparency in reporting. The QCARD-IPC focuses on high-level characteristics of RWD sources and study design domains including data temporality, population, medical product exposure, comparators, and covariates, endpoints, statistical analysis, and data quality assurance plans. CONCLUSIONS: Evaluation of studies including RWD requires understanding the data source, study design, and potential biases to preliminarily evaluate whether selected RWD are fit-for-use for the research question. The Oncology QCARD-IPC provides a structured, transparent approach to facilitate early review and enhanced communication between study sponsors and scientific reviewers of initial study proposals including RWD.

Emulations of Oncology Trials Using Real-World Data: A Systematic Literature Review.

By evaluating published emulations of oncology RCTs studies in which both the active and comparator groups are sourced from RWD and target trial results are available for benchmarking, this systematic review aims to gain insight into factors related to emulation performance. Thirteen oncology emulation studies using various types of RWD were identified through an online database search of PubMed through 2022. Based on the ROBINS-I tool, most studies (N=8) had a serious risk of overall bias driven by risk of bias from confounding. Approximately half of the studies (N=6) fully proxied the RCT entry criteria. Of 11 RWD studies that provided sufficient detail to quantify emulation performance, the emulation HR estimate fell within the 95% CI of the trial estimate in 9 of the studies. There were no clear trends between risk of bias or degree to which the entry criteria were proxied and emulation performance. Findings may have been influenced by publication bias and researcher degrees of freedom, as only one emulation study pre-registered its protocol. Tools for comprehensively characterizing factors that affect emulation performance, including the real-world clinical context as it relates to the RCT research question, are needed to evaluate the feasibility of a RCT emulation.

Real-world performance of SARS-Cov-2 serology tests in the United States, 2020.

BACKGROUND: Real-world performance of COVID-19 diagnostic tests under Emergency Use Authorization (EUA) must be assessed. We describe overall trends in the performance of serology tests in the context of real-world implementation. METHODS: Six health systems estimated the odds of seropositivity and positive percent agreement (PPA) of serology test among people with confirmed SARS-CoV-2 infection by molecular test. In each dataset, we present the odds ratio and PPA, overall and by key clinical, demographic, and practice parameters. RESULTS: A total of 15,615 people were observed to have at least one serology test 14-90 days after a positive molecular test for SARS-CoV-2. We observed higher PPA in Hispanic (PPA range: 79-96%) compared to non-Hispanic (60-89%) patients; in those presenting with at least one COVID-19 related symptom (69-93%) as compared to no such symptoms (63-91%); and in inpatient (70-97%) and emergency department (93-99%) compared to outpatient (63-92%) settings across datasets. PPA was highest in those with diabetes (75-94%) and kidney disease (83-95%); and lowest in those with auto-immune conditions or who are immunocompromised (56-93%). The odds ratios (OR) for seropositivity were higher in Hispanics compared to non-Hispanics (OR range: 2.59-3.86), patients with diabetes (1.49-1.56), and obesity (1.63-2.23); and lower in those with immunocompromised or autoimmune conditions (0.25-0.70), as compared to those without those comorbidities. In a subset of three datasets with robust information on serology test name, seven tests were used, two of which were used in multiple settings and met the EUA requirement of PPA ≥87%. Tests performed similarly across datasets. CONCLUSION: Although the EUA requirement was not consistently met, more investigation is needed to understand how serology and molecular tests are used, including indication and protocol fidelity. Improved data interoperability of test and clinical/demographic data are needed to enable rapid assessment of the real-world performance of in vitro diagnostic tests.

Real-world utilization of SARS-CoV-2 serological testing in RNA positive patients across the United States.

BACKGROUND: As diagnostic tests for COVID-19 were broadly deployed under Emergency Use Authorization, there emerged a need to understand the real-world utilization and performance of serological testing across the United States. METHODS: Six health systems contributed electronic health records and/or claims data, jointly developed a master protocol, and used it to execute the analysis in parallel. We used descriptive statistics to examine demographic, clinical, and geographic characteristics of serology testing among patients with RNA positive for SARS-CoV-2. RESULTS: Across datasets, we observed 930,669 individuals with positive RNA for SARS-CoV-2. Of these, 35,806 (4%) were serotested within 90 days; 15% of which occurred <14 days from the RNA positive test. The proportion of people with a history of cardiovascular disease, obesity, chronic lung, or kidney disease; or presenting with shortness of breath or pneumonia appeared higher among those serotested compared to those who were not. Even in a population of people with active infection, race/ethnicity data were largely missing (>30%) in some datasets-limiting our ability to examine differences in serological testing by race. In datasets where race/ethnicity information was available, we observed a greater distribution of White individuals among those serotested; however, the time between RNA and serology tests appeared shorter in Black compared to White individuals. Test manufacturer data was available in half of the datasets contributing to the analysis. CONCLUSION: Our results inform the underlying context of serotesting during the first year of the COVID-19 pandemic and differences observed between claims and EHR data sources-a critical first step to understanding the real-world accuracy of serological tests. Incomplete reporting of race/ethnicity data and a limited ability to link test manufacturer data, lab results, and clinical data challenge the ability to assess the real-world performance of SARS-CoV-2 tests in different contexts and the overall U.S. response to current and future disease pandemics.

The Aetion Coalition to Advance Real-World Evidence through Randomized Controlled Trial Emulation Initiative: Oncology.

Legislative and technological advancements over the past decade have given rise to the proliferation of healthcare data generated from routine clinical practice, often referred to as real-world data (RWD). These data have piqued the interest of healthcare stakeholders due to their potential utility in generating evidence to support clinical and regulatory decision making. In the oncology setting, studies leveraging RWD offer distinct advantages that are complementary to randomized controlled trials (RCTs). They also permit the conduct of investigations that may not be possible through prospective designs due to ethics or feasibility. Despite its promise, the use of RWD for the generation of clinical evidence remains controversial due to concerns of unmeasured confounding and other sources of bias that must be carefully addressed in the study design and analysis. To facilitate a better understanding of when RWD can provide reliable conclusions on drug effectiveness, we seek to conduct 10 RWD-based studies that emulate RCTs in oncology using a systematic, protocol-driven approach described herein. Results of this investigation will help inform clinical, scientific, and regulatory stakeholders on the applications of RWD in the context of product labeling expansion, drug safety, and comparative effectiveness in oncology.

Development and Evaluation of the Algorithm CErtaInty Tool (ACE-IT) to Assess Electronic Medical Record and Claims-based Algorithms' Fit for Purpose for Safety Outcomes.

INTRODUCTION: Electronic health record (EHR) or medical claims-based algorithms (i.e., operational definitions) can be used to define safety outcomes using real-world data. However, existing tools do not allow researchers and decision-makers to adequately appraise whether a particular algorithm is fit for purpose (FFP) to support regulatory decisions on drug safety surveillance. Our objective was to develop a tool to enable regulatory decision-makers and other stakeholders to appraise whether a given algorithm is FFP for a specific decision context. METHODS: We drafted a set of 77 generic items informed by regulatory guidance documents, existing instruments, and publications. The outcome of ischemic stroke served as an exemplar to inform the development of draft items. The items were designed to be outcome independent. We conducted a three-round online Delphi panel to develop and refine the tool and achieve consensus on items (> 70% agreement) among panel participants composed of regulators, researchers from pharmaceutical organizations, academic clinicians, methodologists, pharmacoepidemiologists, and cardiologists. We conducted a qualitative analysis of panel responses. Five pairs of reviewers independently evaluated two ischemic stroke algorithm validation studies to test its application. We developed a user guide, with explanation and elaboration for each item, guidance on essential and additional elements for user responses, and an illustrative example of a complete assessment. Furthermore, we conducted a 2-h online stakeholder panel of 16 participants from regulatory agencies, academic institutions, and industry. We solicited input on key factors for an FFP assessment, their general reaction to the Algorithm CErtaInty Tool (ACE-IT), limitations of the tool, and its potential use. RESULTS: The expert panel reviewed and made changes to the initial list of 77 items. The panel achieved consensus on 38 items, and the final version of the ACE-IT includes 34 items after removal of duplicate items. Applying the tool to two ischemic stroke algorithms demonstrated challenges in its application and identified shared concepts addressed by more than one item. The ACE-IT was viewed positively by the majority of stakeholders. They identified that the tool could serve as an educational resource as well as an information-sharing platform. The time required to complete the assessment was identified as an important limitation. We consolidated items with shared concepts and added a preliminary screen section and a summary assessment box based on their input. The final version of the ACE-IT is a 34-item tool for assessing whether algorithm validation studies on safety outcomes are FFP. It comprises the domains of internal validity (24 items), external validity (seven items), and ethical conduct and reporting of the validation study (three items). The internal validity domain includes sections on objectives, data sources, population, outcomes, design and setting, statistical methods, reference standard, accuracy, and strengths and limitations. The external validity domain includes items that assess the generalizability to a proposed target study. The domain on ethics and transparency includes items on ethical conduct and reporting of the validation study. CONCLUSION: The ACE-IT supports a structured, transparent, and flexible approach for decision-makers to appraise whether electronic health record or medical claims-based algorithms for safety outcomes are FFP for a specific decision context. Reliability and validity testing using a larger sample of participants in other therapeutic areas and further modifications to reduce the time needed to complete the assessment are needed to fully evaluate its utility for regulatory decision-making.

Using the IMEDS distributed database for epidemiological studies in type 2 diabetes mellitus.

INTRODUCTION: This study aimed to assess data relevancy and data quality of the Innovation in Medical Evidence Development and Surveillance System Distributed Database (IMEDS-DD) for diabetes research and to evaluate comparability of its type 2 diabetes cohort to the general type 2 diabetes population. RESEARCH DESIGN AND METHODS: A retrospective study was conducted using the IMEDS-DD. Eligible members were adults with a medical encounter between April 1, 2018 and March 31, 2019 (index period). Type 2 diabetes and co-existing conditions were determined using all data available from April 1, 2016 to the most recent encounter within the index period. Type 2 diabetes patient characteristics, comorbidities and hemoglobin A1c (HbA1c) values were summarized and compared with those reported in national benchmarks and literature. RESULTS: Type 2 diabetes prevalence was 12.6% in the IMEDS-DD. Of 4 14 672 patients with type 2 diabetes, 52.8% were male, and the mean age was 65.0 (SD 13.3) years. Common comorbidities included hypertension (84.5%), hyperlipidemia (82.8%), obesity (45.3%), and cardiovascular disease (44.7%). Moderate-to-severe chronic kidney disease was observed in 20.2% patients. The most commonly used antihyperglycemic agents included metformin (35.7%), sulfonylureas (14.8%), and insulin (9.9%). Less than one-half (48.9%) had an HbA1c value recorded. These findings demonstrated the notable similarity in patient characteristics between type 2 diabetes populations identified within the IMEDS-DD and other large databases. CONCLUSIONS: Despite the limitations related to HbA1c data, our findings indicate that the IMEDS-DD contains robust information on key data elements to conduct pharmacoepidemiological studies in diabetes, including member demographic and clinical characteristics and health services utilization.

The US Food and Drug Administration Sentinel System: a national resource for a learning health system.

The US Food and Drug Administration (FDA) created the Sentinel System in response to a requirement in the FDA Amendments Act of 2007 that the agency establish a system for monitoring risks associated with drug and biologic products using data from disparate sources. The Sentinel System has completed hundreds of analyses, including many that have directly informed regulatory decisions. The Sentinel System also was designed to support a national infrastructure for a learning health system. Sentinel governance and guiding principles were designed to facilitate Sentinel's role as a national resource. The Sentinel System infrastructure now supports multiple non-FDA projects for stakeholders ranging from regulated industry to other federal agencies, international regulators, and academics. The Sentinel System is a working example of a learning health system that is expanding with the potential to create a global learning health system that can support medical product safety assessments and other research.

Patient characteristics, pain treatment patterns, and incidence of total joint replacement in a US population with osteoarthritis.

BACKGROUND: Currently available medications for chronic osteoarthritis pain are only moderately effective, and their use is limited in many patients because of serious adverse effects and contraindications. The primary surgical option for osteoarthritis is total joint replacement (TJR). The objectives of this study were to describe the treatment history of patients with osteoarthritis receiving prescription pain medications and/or intra-articular corticosteroid injections, and to estimate the incidence of TJR in these patients. METHODS: This retrospective, multicenter, cohort study utilized health plan administrative claims data (January 1, 2013, through December 31, 2019) of adult patients with osteoarthritis in the Innovation in Medical Evidence Development and Surveillance Distributed Database, a subset of the US FDA Sentinel Distributed Database. Patients were analyzed in two cohorts: those with prevalent use of "any pain medication" (prescription non-steroidal anti-inflammatory drugs [NSAIDs], opioids, and/or intra-articular corticosteroid injections) using only the first qualifying dispensing (index date); and those with prevalent use of "each specific pain medication class" with all qualifying treatment episodes identified. RESULTS: Among 1 992 670 prevalent users of "any pain medication", pain medications prescribed on the index date were NSAIDs (596 624 [29.9%] patients), opioids (1 161 806 [58.3%]), and intra-articular corticosteroids (323 459 [16.2%]). Further, 92 026 patients received multiple pain medications on the index date, including 71 632 (3.6%) receiving both NSAIDs and opioids. Altogether, 20.6% of patients used an NSAID at any time following an opioid index dispensing and 17.2% used an opioid following an NSAID index dispensing. The TJR incidence rates per 100 person-years (95% confidence interval [CI]) were 3.21 (95% CI: 3.20-3.23) in the "any pain medication" user cohort, and among those receiving "each specific pain medication class" were NSAIDs, 4.63 (95% CI: 4.58-4.67); opioids, 7.45 (95% CI: 7.40-7.49); and intra-articular corticosteroids, 8.05 (95% CI: 7.97-8.13). CONCLUSIONS: In patients treated with prescription medications for osteoarthritis pain, opioids were more commonly prescribed at index than NSAIDs and intra-articular corticosteroid injections. Of the pain medication classes examined, the incidence of TJR was highest in patients receiving intra-articular corticosteroids and lowest in patients receiving NSAIDs.

Longitudinal assessment of liver stiffness by transient elastography for chronic hepatitis C patients.

BACKGROUND: Liver fibrosis is a common pathway of liver injury and is a feature of most chronic liver diseases. Fibrosis progression varies markedly in patients with hepatitis C virus (HCV). Liver stiffness has been recommended as a parameter of fibrosis progression/regression in patients with HCV. AIM: To investigate changes in liver stiffness measured by transient elastography (TE) in a large, racially diverse cohort of United States patients with chronic hepatitis C (CHC). METHODS: We evaluated the differences in liver stiffness between patients treated with direct-acting antiviral (DAA) therapy and untreated patients. Patients had ≥ 2 TE measurements and no prior DAA exposure. We used linear regression to measure the change in liver stiffness between first and last TE in response to treatment, controlling for age, sex, race, diabetes, smoking status, human immunodeficiency virus status, baseline alanine aminotransferase, and baseline liver stiffness. Separate regression models analyzed the change in liver stiffness as measured by kPa, stratified by cirrhosis status. RESULTS: Of 813 patients, 419 (52%) initiated DAA treatment. Baseline liver stiffness was 12 kPa in 127 (16%). Median time between first and last TE was 11.7 and 12.7 mo among treated and untreated patients, respectively. There was no significant change in liver stiffness observed over time in either the group initiating DAA treatment (0.016 kPa/month; CI: -0.051, 0.084) or in the untreated group (0.001 kPa/mo; CI: -0.090, 0.092), controlling for covariates. A higher baseline kPa score was independently associated with decreased liver stiffness. CONCLUSION: DAA treatment was not associated with a differential change in liver stiffness over time in patients with CHC compared to untreated patients.

Value of Triage Treatment Strategies to Distribute Hepatitis C Direct-Acting Antiviral Agents in an Integrated Healthcare System: A Cost-Effectiveness Analysis.

OBJECTIVES: This study aimed to assess the cost-effectiveness of fibrosis-based direct-acting antiviral treatment policies for patients with chronic hepatitis C virus at the Kaiser Permanente Mid-Atlantic States health system. METHODS: We used a Markov model to compare the lifetime costs and effects of treating patients with chronic hepatitis C virus at different stages of disease severity, or all stages simultaneously, based on a fibrosis score from the US healthcare sector perspective and societal perspective. The initial distribution of patients across fibrosis scores, the effectiveness of direct-acting antiviral therapy, and follow-up and monitoring protocols were specific to the Kaiser Permanente Mid-Atlantic States health system. Direct and indirect costs, transition probabilities, and utilities were derived from the literature. Deterministic and probabilistic sensitivity analyses were performed to assess the robustness of our results. RESULTS: The "Treat All" option was dominant from both the societal and healthcare sector perspectives. The conclusion was robust in deterministic sensitivity analysis. The range of incremental costs between the less restrictive policies was small-the difference between the "Treat F1+" and the "Treat All" option was only $111 per person. Probabilistic sensitivity analyses showed, at both the $100 000/quality-adjusted life-year and $150 000/quality-adjusted life-year thresholds, there was a 70% chance that the "Treat All" option was more cost-effective than the "Treat F1+" option. CONCLUSIONS: We found that expanded treatment access is cost-effective and, in many cases, cost saving. Although our results are primarily applicable to a regional integrated healthcare system, it offers some direction to any healthcare setting faced with resource constraints in the face of highly priced drugs.

Dynamic Risk Prediction of Response to Ursodeoxycholic Acid Among Patients with Primary Biliary Cholangitis in the USA.

BACKGROUND: Ursodeoxycholic acid (UDCA) remains the first-line therapy for primary biliary cholangitis (PBC); however, inadequate treatment response (ITR) is common. The UK-PBC Consortium developed the modified UDCA Response Score (m-URS) to predict ITR (using alkaline phosphatase [ALP] > 1.67 times the upper limit of normal [*ULN]) at 12 months post-UDCA initiation). Using data from the US-based Fibrotic Liver Disease Consortium, we assessed the m-URS in our multi-racial cohort. We then used a dynamic modeling approach to improve prediction accuracy. METHODS: Using data collected at the time of UDCA initiation, we assessed the m-URS using the original formula; then, by calibrating coefficients to our data, we also assessed whether it remained accurate when using Paris II criteria for ITR. Next, we developed and validated a dynamic risk prediction model that included post-UDCA initiation laboratory data. RESULTS: Among 1578 patients (13% men; 8% African American, 9% Asian American/American Indian/Pacific Islander; 25% Hispanic), the rate of ITR was 27% using ALP > 1.67*ULN and 45% using Paris II criteria. M-URS accuracy was "very good" (AUROC = 0.87, sensitivity = 0.62, and specificity = 0.82) for ALP > 1.67*ULN and "moderate" (AUROC = 0.74, sensitivity = 0.57, and specificity = 0.70) for Paris II. Our dynamic model significantly improved accuracy for both definitions of ITR (ALP > 1.67*ULN: AUROC = 0.91; Paris II: AUROC = 0.81); specificity approached 100%. Roughly 9% of patients in our cohort were at the highest risk of ITR. CONCLUSIONS: Early identification of patients who will not respond to UDCA treatment using a dynamic prediction model based on longitudinal, repeated risk factor measurements may facilitate earlier introduction of adjuvant treatment.

The use of narrative electronic prescribing instructions in pharmacoepidemiology: A scoping review for the International Society for Pharmacoepidemiology.

Narrative electronic prescribing instructions (NEPIs) are text that convey information on the administration or co-administration of a drug as directed by a prescriber. For researchers, NEPIs have the potential to advance our understanding of the risks and benefits of medications in populations; however, due to their unstructured nature, they are not often utilized. The goal of this scoping review was to evaluate how NEPIs are currently employed in research, identify opportunities and challenges for their broader application, and provide recommendations on their future use. The scoping review comprised a comprehensive literature review and a survey of key stakeholders. From the literature review, we identified 33 primary articles that described the use of NEPIs. The majority of articles (n = 19) identified issues with the quality of information in NEPIs compared with structured prescribing information; nine articles described the development of novel algorithms that performed well in extracting information from NEPIs, and five described the used of manual or simpler algorithms to extract prescribing information from NEPIs. A survey of 19 stakeholders indicated concerns for the quality of information in NEPIs and called for standardization of NEPIs to reduce data variability/errors. Nevertheless, stakeholders believed NEPIs present an opportunity to identify prescriber's intent for the prescription and to study temporal treatment patterns. In summary, NEPIs hold much promise for advancing the field of pharmacoepidemiology. Researchers should take advantage of addressing important questions that can be uniquely answered with NEPIs, but exercise caution when using this information and carefully consider the quality of the data.

COVID-19 Evidence Accelerator: A parallel analysis to describe the use of Hydroxychloroquine with or without Azithromycin among hospitalized COVID-19 patients.

BACKGROUND: The COVID-19 pandemic remains a significant global threat. However, despite urgent need, there remains uncertainty surrounding best practices for pharmaceutical interventions to treat COVID-19. In particular, conflicting evidence has emerged surrounding the use of hydroxychloroquine and azithromycin, alone or in combination, for COVID-19. The COVID-19 Evidence Accelerator convened by the Reagan-Udall Foundation for the FDA, in collaboration with Friends of Cancer Research, assembled experts from the health systems research, regulatory science, data science, and epidemiology to participate in a large parallel analysis of different data sets to further explore the effectiveness of these treatments. METHODS: Electronic health record (EHR) and claims data were extracted from seven separate databases. Parallel analyses were undertaken on data extracted from each source. Each analysis examined time to mortality in hospitalized patients treated with hydroxychloroquine, azithromycin, and the two in combination as compared to patients not treated with either drug. Cox proportional hazards models were used, and propensity score methods were undertaken to adjust for confounding. Frequencies of adverse events in each treatment group were also examined. RESULTS: Neither hydroxychloroquine nor azithromycin, alone or in combination, were significantly associated with time to mortality among hospitalized COVID-19 patients. No treatment groups appeared to have an elevated risk of adverse events. CONCLUSION: Administration of hydroxychloroquine, azithromycin, and their combination appeared to have no effect on time to mortality in hospitalized COVID-19 patients. Continued research is needed to clarify best practices surrounding treatment of COVID-19.

Hepatitis C Care Pathway Associated With Increased Screening, Confirmation, and Diagnosis Communication to Patients.

National policy initiatives1-3 and the advent of highly efficacious direct-acting antivirals4 set the stage to increase the identification and care of patients with hepatitis C virus (HCV). We implemented a multifaceted HCV care pathway, inclusive of automated screening alerts for all patients born between 1945 and 1965 as they are registered for appointments, reflex laboratory orders for positive HCV antibody results, and a care coordinator to facilitate diagnosis communication and engagement in follow-up care.5 We report the impact of that pathway on HCV screening, confirmation, diagnosis communication, and co-infection screening.

Validity of an Automated Algorithm to Identify Cirrhosis Using Electronic Health Records in Patients with Primary Biliary Cholangitis.

BACKGROUND: Biopsy remains the gold standard for determining fibrosis stage in patients with primary biliary cholangitis (PBC), but it is unavailable for most patients. We used data from the 11 US health systems in the FibrOtic Liver Disease Consortium to explore a combination of biochemical markers and electronic health record (EHR)-based diagnosis/procedure codes (DPCs) to identify the presence of cirrhosis in PBC patients. METHODS: Histological fibrosis staging data were obtained from liver biopsies. Variables considered for the model included demographics (age, gender, race, ethnicity), total bilirubin, alkaline phosphatase, albumin, aspartate aminotransferase (AST) to platelet ratio index (APRI), Fibrosis 4 (FIB4) index, AST to alanine aminotransferase (ALT) ratio, and >100 DPCs associated with cirrhosis/decompensated cirrhosis, categorized into ten clusters. Using least absolute shrinkage and selection operator regression (LASSO), we derived and validated cutoffs for identifying cirrhosis. RESULTS: Among 4328 PBC patients, 1350 (32%) had biopsy data; 121 (9%) were staged F4 (cirrhosis). DPC clusters (including codes related to cirrhosis and hepatocellular carcinoma diagnoses/procedures), Hispanic ethnicity, ALP, AST/ALT ratio, and total bilirubin were retained in the final model (AUROC=0.86 and 0.83 on learning and testing data, respectively); this model with two cutoffs divided patients into three categories (no cirrhosis, indeterminate, and cirrhosis) with specificities of 81.8% (for no cirrhosis) and 80.3% (for cirrhosis). A model excluding DPCs retained ALP, AST/ALT ratio, total bilirubin, Hispanic ethnicity, and gender (AUROC=0.81 and 0.78 on learning and testing data, respectively). CONCLUSION: An algorithm using laboratory results and DPCs can categorize a majority of PBC patients as cirrhotic or noncirrhotic with high accuracy (with a small remaining group of patients' cirrhosis status indeterminate). In the absence of biopsy data, this EHR-based model can be used to identify cirrhosis in cohorts of PBC patients for research and/or clinical follow-up.