Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Nneka Ufere

Liver Transplant · MGH

View

Dr. Vinay Jahagirdar

Liver Transplant · Medical College of Virginia

View

Dr. Lauren Feld

Liver Transplant · UMass Memorial

View

Dr. Sonali Paul

Liver Transplant · UCMC

View

Dr. Pierre Gholam

Liver Transplant · CWRU

View

Dr. Elliot Tapper

Liver Transplant · UM

View

Dr. Saurabh Sethi

General GI · Oja Health

View

Dr. Sara Horst

IBD · Vanderbilt

View

Dr. Wendi LeBrett

General GI · ModernGut

View

Dr. Treta Purohit

General GI · Oshi Health

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    A nationwide study of primary biliary cholangitis prevalence, geographic distribution, and health care providers.

    Apr 14, 2025DOI: 10.1097/hc9.0000000000000677Journal ArticleHepatology Communications

    Written by: Cynthia LevyKeri-Ann Buchanan-PeartJoanna P MacEwanAlina LevineRadhika NairDarren WheelerLeona BessonovaAparna GoelRobert G GishAlan Bonder

    Find these authors on Roon

    AB
    Dr. Alan Bonder
    Liver Transplant · BIDMC

    INTRODUCTION

    Primary biliary cholangitis (PBC) is a rare chronic, autoimmune, cholestatic liver disease characterized by the destruction of intrahepatic bile ducts that predominantly affects women.1,2 PBC commonly presents with nonspecific symptoms such as fatigue and pruritus, and without treatment, the disease slowly progresses to cirrhosis approximately 10–20 years after diagnosis.3,4 However, the clinical presentation and natural course of PBC are highly variable among patients.4 PBC progression confers economic burden to payers and patients, including risk of hospitalizations, progression to cirrhosis and its complications, and liver transplantation.5 Ursodeoxycholic acid (UDCA), the first-line treatment for PBC, has been approved by the US Food and Drug Administration since 1997.6 For patients with inadequate response or intolerance to UDCA, the US Food and Drug Administration conditionally approved obeticholic acid (OCA) in 20167 and, more recently, seladelpar and elafibranor in 2024 as second-line therapies.8,9 Off-label use of fibrates may also be considered in patients with inadequate response to UDCA.2,3While significant progress has been made in understanding the natural history and overall management of PBC, better knowledge of epidemiology and geographic variation in the United States is needed to improve access to care.10 To date, the global prevalence of PBC has largely been extrapolated from regional databases with relatively small patient cohorts.11–19 The prevalence estimates of PBC vary substantially, ranging from 1.91 to 40.2 per 100,000 persons, with discrepancies likely due to differing study methods assessing unique populations and geographic areas.20–22 Additionally, it has been shown that global PBC prevalence has increased over time, perhaps contributing further to the wide range of reported prevalence estimates.23One of the largest and most recent epidemiologic studies of PBC in the United States identified a 12-year period (2003–2014) prevalence of 29.3 per 100,000 persons.10 This overall prevalence varied by geographic region and patient demographics, such as age, gender, and race. Using the same clinical database (the Fibrotic Liver Disease [FOLD] Consortium) of 11 health systems spanning 4 US Census Bureau–defined regions, the prevalence of PBC was found to have increased annually from 2006 (21.7 per 100,000 persons) to 2014 (39.2 per 100,000 persons).24 This reported prevalence of 39.2 per 100,000 persons in 2014 is the most comprehensive assessment to date of PBC prevalence in the United States.Since 2014, PBC disease nomenclature has changed,25 and the management and available therapeutic options for PBC have evolved.2,3 Furthermore, in 2015, the International Classification of Diseases, Ninth Revision (ICD-9) transitioned to ICD-10, resulting in a more specific diagnosis code for PBC.26 A contemporary description of PBC prevalence in the United States using a robust national database is needed to better reflect improved disease awareness, as well as improved coding and capture in medical claims data. Additionally, no study has performed a granular analysis of the geographic distribution of PBC in the United States, which is particularly important given the evolution of understanding about the sociodemographic and potential environmental risk factors for PBC.27
    The objectives of this study were to estimate the overall prevalence of PBC in the United States and assess the regional variability by 3-digit ZIP Code tabulation area (ZCTA3) using a large national administrative claims database, as well as to describe the health care providers (HCPs) and their prescribing practices for patients with PBC.

    METHODS

    Komodo's Healthcare Map (Komodo Health), which includes administrative claims data on approximately 330 million patients and 15 million new clinical encounters daily, was used for this study.28 Patients were included in the prevalence numerator if they had either ≥1 inpatient or ≥2 outpatient claims at least a day apart with a diagnosis code for PBC (ICD-9 code [571.6] or ICD-10 code [K74.3])29 between January 1, 2014, and December 31, 2021, and they were 18 years of age or above at the time of initial claim with a PBC diagnosis code. The prevalence denominator comprised individuals 18 years of age or above in 2021. Patients were excluded from the study if they died on or before January 1, 2021, per the Komodo Health mortality table, had missing information on their 3-digit ZIP Code and/or gender in 2021 Komodo Health enrollment periods, or there was suspicion of patient duplication in Komodo Health.
    The 3-digit ZIP Codes were mapped to ZCTA3s, with certain ZCTA3s combined into groups to allow for 1-to-1 mapping. ZCTAs, as defined by the US Census Bureau, are “generalized areal representations of the geographic extent and distribution of the point-based ZIP Codes built using 2020 Census tabulation blocks.”30 ZCTAs, rather than ZIP Codes, were used for this analysis because the Census Bureau produces population data at the ZCTA3 level.
    PBC prevalence per 100,000 adult population was adjusted by age and gender at the ZCTA3 level, which was derived from the 2021 American Community Survey.31 Adjustment via direct standardization accounted for differences in coverage and representativeness of Komodo Health, as well as the age and gender distribution of patients in Komodo Health by ZCTA3. In the age, gender, and ZCTA3 groups where the Komodo Health population was larger than the Census population, the Census population was used as the denominator.
    The absolute number of PBC cases, as well as the unadjusted, age-adjusted, and gender-adjusted PBC prevalence per 100,000 adults, were estimated. Descriptive analyses were used to assess the demographic characteristics of individuals with PBC. Choropleth maps of the United States were generated to display the geographic distribution of patients with PBC by ZCTA3, and 4 Census Bureau–defined regions (Northeast [Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont], Midwest [Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin], West [Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming], and South [Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia]) were used to further characterize regional prevalence.32
    In an exploratory analysis, HCPs were identified using patients’ latest medical claim with a primary diagnosis of PBC or pharmacy claims for UDCA, OCA, or fenofibrate with a nonmissing National Provider Identifier (NPI) number. Additionally, all PBC-related claims occurring during the study period (2014–2021) were assessed in a separate analysis. Komodo Health’s provider summary table supplied the HCP’s primary health care organization and HCP primary type/specialty. HCPs were categorized as specialists (gastroenterologists and hepatologists), general medicine physicians (comprising general practice, family practice, internal medicine, and geriatric medicine), advanced practice providers (APPs; including specialist APPs as specialty designation was not available for this category), or other. Komodo Health’s Health Care Organization master table was then used to determine if the HCP’s primary health care organization was an academic institution/center. If a claim date had multiple claims with multiple primary NPI numbers or prescriber NPI numbers (for pharmacy claims) with different specialties and/or primary affiliations, then data from all primary HCPs/prescribing HCPs (ie, all claims) were included in the analyses summarizing HCP specialties and academic affiliation.
    Finally, in a separate exploratory analysis, the number of patients who filled prescriptions for UDCA and/or second-line therapy (ie, OCA or fenofibrate) before 2022 was assessed. In this analysis, patients were required to have ≥30 days of pharmacy enrollment at any point on or after their initial PBC diagnosis date up until December 31, 2021. The specialty/type of prescribers was also examined, where available. As patients may have been prescribed UDCA and/or second-line therapy by different HCPs over time (eg, a gastroenterologist, an APP, and a general medicine physician), prescriptions from all HCPs were included in the analysis.
    Only deidentified patient data from Komodo Health were used in this study and may be considered exempt. The study was not submitted to an Institutional Review Board for review.

    RESULTS

    A total of 41,426 of the 106.8 million adults in Komodo Health met all study selection criteria. Most of the patients were female (83.0%), with a mean (SD) age in 2021 of 61.9 (13.2) years (Table 1). Of the 31,806 (76.8% of total) patients with race/ethnicity information available, 67.6% were White, 16.9% were Hispanic/Latino, 7.6% were Black/African American, 3.5% were Asian or Pacific Islander, and 4.3% were classified as “other.” Commercial and Medicare were the most common types of insurance coverage (45.8% and 31.6% of patients, respectively).
    Demographics of patients with primary biliary cholangitis in 2021 (N=41,426)
    Race/ethnicity was missing in 23.2% (n=9620) of the total sample.
    Patients were categorized as having 1 type of insurance, with a hierarchy of Medicare > Medicaid > commercial > other. Insurance coverage was missing in 1.4% (n=560) of the total sample.
    Includes self-insured, exchange (marketplace), or any other insurance besides Medicare, Medicaid, and commercial.
    The adjusted prevalence of adult patients with PBC in 2021 was 40.9 per 100,000 persons, for an estimated total of 105,506 adults with PBC in the United States (including Puerto Rico) in 2021 (Table 2). This national prevalence of PBC in the adult population in 2021 was then reported by region, with prevalence per 100,000 adults in the Northeast, Midwest, West, and South of 43.8 (19,881 estimated total cases), 43.0 (23,038 estimated total cases), 39.6 (23,839 estimated total cases), and 39.5 (37,989 estimated total cases), respectively. A sensitivity analysis, including only patients with a PBC diagnosis based on the ICD-10 diagnosis code, resulted in an adjusted prevalence of 38.3 per 100,000 adult population with similar regional prevalence and estimated total patient counts (Table 2). Whereas the highest estimated absolute number of patients with PBC occurred in dense urban areas (Supplemental Figure S1, http://links.lww.com/HC9/B937), the prevalence per 100,000 adult population was highest in some rural areas located in Texas, North Dakota, Colorado, Wyoming, Idaho, Washington, Vermont, and North Carolina (Figure 1).
    2021 adjusted primary biliary cholangitis prevalence in adults and estimated total patient count by region in the United Statesa
    Including Puerto Rico.
    Abbreviation: ICD, International Classification of Diseases.
    2021 prevalence of primary biliary cholangitis per 100,000 adult population by 3-digit ZIP Code tabulation area. Rate is defined as number of patients per 100,000 adult population.
    This study also quantified the type/specialty of HCPs that manage patients with PBC and their primary affiliation (academic vs. nonacademic), where available. Among the 36,700 patients with clinician primary specialty information available, approximately half (53.5%) of patients with PBC, irrespective of therapy use, were treated by a gastroenterologist or hepatologist as reported in their most recent medical or pharmacy claim for PBC (Table 3). Among patients treated by a specialist per their latest claim, 42.5% were treated by an HCP with an academic affiliation. The remainder of patients were treated by an APP (inclusive of specialist physician assistants or nurse practitioners; 19.3%), a general medicine physician (18.1%), or a physician of another specialty (11.7%). Among patients treated by an APP or general medicine physician as reported in their most recent claim, 46.0% and 23.1%, respectively, were treated by an HCP with an academic affiliation.
    Provider primary specialties managing patients with primary biliary cholangitis
    Note: Among encounters with a nonmissing primary specialty.
    The latest PBC-related medical or pharmacy claim occurred in 2021.
    The additional all-claims analysis included all PBC-related medical and pharmacy claims that occurred from 2014 to 2021.
    Percentages do not add up to 100% because patients could have multiple primary NPI numbers for providers with different specialties and/or affiliations on their last claim date. Only the primary specialty of the provider was considered.
    Analysis only included providers who were able to be classified as either academic or nonacademic in affiliation. The total n’s for this subpopulation were 19,137; 5982; 6626; and 3919 for GI/hepatology, general medicine, APP, and other, respectively.
    Percentages do not add up to 100% owing to considerable overlap as HCP primary specialty categories were not mutually exclusive. For each claim, only the primary specialty of the provider was considered.
    Analysis only included providers who were able to be classified as either academic or nonacademic in affiliation. The total n’s for this subpopulation were 29,873, 14,768, 14,556, and 15,500 for GI/hepatology, general medicine, APP, and other, respectively.
    Includes general practice, family practice, internal medicine, and geriatric medicine.
    Abbreviations: APP, advanced practice provider; GI, gastroenterology; NPI, National Provider Identifier; PBC, primary biliary cholangitis.
    In the analysis of all PBC-related medical and pharmacy claims occurring during the study period (n=36,837 with clinician primary specialty information available), 83.2% of patients were treated by a specialist; 51.1% of these patients were treated by a specialist with an academic affiliation (Table 3). The remainder of patients were treated by a general medicine physician (43.9%), an APP (43.3%), or a physician of another specialty (45.5%). A minority (29.1%) of patients were treated by a general medicine physician with an academic affiliation, whereas approximately half (51.1%) of patients were treated by an APP with an academic affiliation.
    Among the 37,929 patients with at least a month of pharmacy coverage at any time during the study period on or after their first PBC diagnosis per Komodo Health, 28,869 (76.1%) patients had a claim for UDCA between their first PBC diagnosis and 2021 (Table 4). Most (73.6%) patients prescribed UDCA had ≥1 prescription from a specialist (gastroenterologist or hepatologist), whereas 36.9% and 27.4% of patients had ≥1 prescription from an APP or general medicine physician, respectively. For PBC therapy overall, 73.8% of patients had ≥1 prescription from a specialist, and 37.7% had ≥1 prescription from an APP (inclusive of specialist APPs).
    Ursodeoxycholic acid and/or second-line therapy prescriptions during the study period (2014–2021) by provider specialty
    Note: Percentages among different provider types do not add up to 100% as patients may have been prescribed UDCA and/or second-line therapy by different providers over time; prescriptions from all providers were included in the analysis. The denominators for the percentage calculations among different provider types are the number of patients prescribed UDCA (28,869) or patients prescribed UDCA and/or second-line therapy (29,196).
    Includes OCA or fenofibrate.
    Includes general practice, family practice, internal medicine, and geriatric medicine.
    Abbreviations: APP, advanced practice provider; GI, gastroenterology; OCA, obeticholic acid; PBC, primary biliary cholangitis; UDCA, ursodeoxycholic acid.

    DISCUSSION

    This analysis provides the most comprehensive, population-based estimate of PBC prevalence in the United States to date. Komodo Health allows for a contemporary, robust, and inclusive estimation of the national prevalence of PBC in the United States.
    The demographics of the PBC population, predominantly composed of White women approximately 60 years of age, are in line with prior studies in the United States.10,15 Notably, 17% of patients in this study were male, a relatively high percentage compared with other US-based studies,15,33 but similar to that from the FOLD database (18%).10 Additionally, nearly 17% of the study population with information available on race/ethnicity were Hispanic/Latino, and 8% were Black/African American. These are also comparable percentages with those from the FOLD database10 but higher than those seen in other US-based studies.15,33 Finding diverse and inclusive study cohorts, such as this one, is an important means of working toward health equity among patients with PBC. Black patients are significantly less likely to receive treatment with first-line UDCA than White patients10 and have increased odds of mortality among PBC-related hospitalizations.34 Compared with White patients, Hispanic patients have a higher PBC hospitalization rate34 and lower rates of response to UDCA,35 and both Black and Hispanic patients are more likely to have advanced PBC.33 Developing a better understanding of PBC epidemiology in the United States, especially among the non-White population, is a critical step in providing more equitable care.
    The adjusted 2021 adult prevalence of 40.9 per 100,000 persons in this study is similar to the highest previous estimate of PBC prevalence (40.2 per 100,000 persons) from Olmsted County, Minnesota.15 Compared with the small community analysis from Olmsted County performed over 25 years ago,15 the present study provides an updated and population-adjusted estimation of overall PBC prevalence in the United States. These data also support the observed increase in PBC prevalence in the United States from 2006 (21.7 per 100,000 persons) to 2014 (39.2 per 100,000 persons).24 This increase in prevalence may be attributed to increased disease awareness, better disease reporting, more precise diagnosis codes, timely patient identification, improved patient survival associated with PBC treatment, and access to larger databases capturing the population with PBC.23,24
    These data are the first to highlight the geographic diversity of PBC prevalence in the United States. As PBC has been associated with certain infections, cigarette smoking, pollutants, groundwater contamination, Superfund toxic waste sites, and coal mining,27,36–38 the regional spikes in prevalence observed in this study may be behaviorally and/or environmentally related. A link between the genome and the environment in the form of epigenetic changes may also contribute to differences in disease presentation, progression, and treatment response, all of which are highly variable among patients.39 The observed localized concentrations of higher PBC prevalence in some rural areas suggest the need to further explore potential environmental, behavioral, and/or genetic risk factors.
    These data, in an overall population diagnosed with PBC with patients in varying stages of the disease, also show that the vast majority (83.2%) of patients, irrespective of therapy use, received care from a specialist (gastroenterologist or hepatologist) at some point during the study period. However, only about half (53.5%) of patients were seen by a specialist at their most recent visit with a PBC-related claim, suggesting that there may still be barriers to specialist care for some patients in the United States. It is also possible that these patients were seen by specialist APPs practicing in a gastroenterology/hepatology setting. Most patients were prescribed PBC therapies by a gastroenterologist, hepatologist, or APP. These data highlight the need for continuous access to specialist care once patients are diagnosed with PBC so that specialists who are familiar with PBC can provide PBC-specific care. Access to specialists is important for patients with PBC, particularly as the therapeutic landscape for PBC continues to evolve. Additional studies are required to assess potential differences in PBC care by region, population setting (eg, urban vs. rural), and care setting (eg, larger academic centers vs. community care).
    Approximately 75% of patients in this analysis filled a UDCA prescription, which is in line with rates from prior PBC studies.10,24,40 Further, <10% of patients filled a prescription for second-line therapy, leaving a meaningful proportion of patients who were not being treated with available first-line or second-line therapies during the study period. Although claims databases do not describe why patients may not have filled their prescriptions, the substantial proportion of patients not on therapy emphasizes a need for greater awareness of PBC and its management. Additional analyses may help to better elucidate these gaps in treatment among an overall patient population with PBC in varying stages of the disease.
    There are several limitations to this study. As claims data are collected for billing purposes, undercoding and overcoding are possible. Since Komodo Health includes individuals with health insurance and a health encounter, this study’s prevalence estimates may be conservative; uninsured and undiagnosed individuals may not be included in this data set, and not all patients enrolled in health insurance may be part of Komodo Health. Additionally, patients with either an ICD-9 code of 571.6 (nonspecific for PBC) or an ICD-10 code of K74.3 (specific for PBC) were eligible to be included in the study. While the ICD-9 code for PBC is not as specific as the ICD-10 code, a sensitivity analysis performed with only patients diagnosed with PBC based on the PBC-specific ICD-10 code showed similar results to the original study analysis, demonstrating that overestimation of prevalence due to the use of the ICD-9 code for PBC was unlikely.
    Komodo Health may have limitations inherent to secondary databases, including missing information, variables, and subgroups of interest.41 For example, 14.1% of patients had a last claim date without a nonmissing HCP, in which case their most recent claim with a nonmissing NPI number was chosen for the last claim analysis. As this claim only reflects a specific point in time of a patient’s treatment journey, it is possible that this claim did not capture the HCP primarily responsible for the patients’ PBC management. All PBC-related claims occurring during the study period were also examined to provide a broader overview of PBC providers during a multiyear period. The all-claims analysis, reflecting different lengths of available data during which the HCP category could be captured, yielded overlapping (ie, not mutually exclusive) HCP categories. The all-claims analysis indicates that while there is a record of gastroenterology/hepatology care in the majority (83.2%) of patients with PBC at one time, all patients may not have had equal or recent access to specialist care throughout their journey. As both methods (last claim and all claims) of assessing HCPs had limitations, a study assessing the patient journey is needed to better understand the different HCPs involved in patients’ care.
    A limitation of NPI-based classification of HCPs’ primary specialties is the selection of a single primary taxonomy code, even if HCPs practice in multiple specialties. It is possible that HCPs’ practice information and/or primary taxonomy was not accurate or up to date. Additionally, while details about HCP practices and work settings are not readily available in a claims database, this analysis did include the designation for academic setting from Komodo Health, where available, to provide insight into where patients with PBC are obtaining care. However, the representation of academic vs. private practices in Komodo Health has not been assessed; hence, this cannot be ruled out as a potential source of bias. A meaningful proportion of patients in this analysis was not prescribed PBC therapy, which may be owing to clinicians practicing in isolation or in nonacademic settings with more restricted access to medical meetings, up-to-date management information, and specialist support for multidisciplinary care. Further analysis would be required to determine differences in patient care, if any, between academic and nonacademic HCPs for patients with PBC, but this is beyond the scope of the study. Given the lack of completeness of HCP data, the HCP analysis in this study was considered exploratory.
    Data on how PBC was diagnosed in study patients (ie, biochemical evidence of cholestasis, antimitochondrial antibody assay, histopathology) are also not captured in Komodo Health. Lastly, pharmacy claims for fenofibrate, which is more traditionally used for lipid disorders, were used in the determination of HCP specialty and affiliation. However, fenofibrate comprised only 1.7% of pharmacy claims in this analysis.
    In summary, this study provides the most comprehensive population-based estimate to date of the prevalence of PBC in the United States using a large national claims database. The prevalence estimates of PBC and regional variability, especially in some rural areas, highlight the need to further examine potential risk factors for PBC. Additionally, most patients with PBC, irrespective of therapy use, were treated by a specialist (gastroenterologist or hepatologist) at some point during the study period, but only approximately half received specialist care in their most recent claim. Greater access to specialist care once patients are diagnosed to support awareness of PBC and its management is needed.
    ABDr. Alan Bonder

    Dr. Alan Bonder

    Gastroenterology · Hepatology & Liver Transplantation
    Beth Israel Deaconess Medical Center
    alan