Abstract
Background:
As metabolic dysfunction-associated steatotic liver disease (MASLD) management extends into primary care, little is known about patterns of specialty referral for affected patients. We determined the proportion of primary care patients with MASLD that received a gastroenterology (GI) consultation and compared advanced fibrosis risk between patients with and without a referral.
Methods:
This retrospective study of electronic health record data from a primary care clinic included patients with MASLD, no competing chronic liver disease diagnoses, and no history of cirrhosis. Referral to GI for evaluation and management (E/M) any time after MASLD ascertainment was the outcome. Fibrosis-4 Index (FIB-4) scores were calculated, categorized by advanced fibrosis risk, and compared by receipt of a GI E/M referral. Logistic regression models were developed to determine the association of FIB-4 risk with receipt of a GI referral.
Results:
The cohort included 652 patients of which 12% had FIB-4 scores (≥ 2.67) at high-risk for advanced fibrosis. Overall, 31% of cohort patients received a GI referral for E/M. There was no difference in the proportion of patients with high (12% vs. 12%, p=0.952) risk FIB-4 scores by receipt of a GI E/M referral. In adjusted logistic regression models, high-risk FIB-4 scores (OR 1.01; 95% CI 0.59 – 1.71) were not associated with receipt of a referral.
Conclusions:
Only 30% of patients in this primary care MASLD cohort received a GI E/M referral during the study period, and those patients with a referral did not differ by FIB-4 advanced fibrosis risk.
Keywords: nonalcoholic fatty liver disease, NAFLD, NASH, fibrosis, metabolic syndrome
Introduction
As public health efforts to address the metabolic dysfunction-associated steatotic liver disease (MASLD, formerly nonalcoholic fatty liver disease [NAFLD]) epidemic have emerged, comprehensive public health approaches include unifying themes of raising civic awareness of MASLD and empowering primary care providers (PCP) to better recognize, diagnose, and manage MASLD in the primary care setting.1–4 Migrating MASLD care from the specialty Gastroenterology (GI) and Hepatology setting, where it has historically resided, to primary care will require an understanding of current MASLD management practices and a needs assessment for successful integration of ongoing MASLD care delivery into an already overwhelming primary care workload.5 One critical element of the MASLD care transition will be the practice of specialty referral to GI and the determination of which patients are referred to GI and when this referral should occur.6
Recently, the American Gastroenterological Association (AGA) and the American Association for the Study of Liver Diseases (AASLD) published guidelines to assist PCPs in providing care to patients with MASLD.7,8 In these guidance statements, a high-risk assessment of advanced fibrosis provides the key indicator for specialty referral since advanced fibrosis (Metavir F3+) is the single best predictor of progression to cirrhosis and hepatocellular cancer outcomes.9–11 One of the primary tools for assessing advanced fibrosis risk is the Fibrosis-4 Index (FIB-4), a non-invasive, serologic risk score calculated from laboratory inputs (alanine aminotransferase [ALT], aspartate aminotransferase [AST], and platelet count). FIB-4 is easily calculated (FIB-4=[Age x AST] / [Platelets x √ALT] ), inexpensive, and the inputs are readily available in primary care.12,13 FIB-4 was first studied in Canada as a potential key cog in the referral pathway, with an initial concern that as the MASLD epidemic expands too many patients would be referred to GI and overwhelm the health system.14 An initial FIB-4 fibrosis risk assessment strategy accurately identified advanced fibrosis and reduced potential specialty referrals by 87%.14 Though FIB-4 holds a lot of promise for non-invasively detecting advanced fibrosis and optimizing specialty referrals to focus on patients at greatest risk for future poor liver outcomes, few PCPs are familiar with FIB-4 and even fewer calculate these scores in practice.15
To successfully implement MASLD primary care pathways with advanced fibrosis risk as a signal for linking patients to specialty care, a better understanding of historical MASLD referral practices, and how they differ by advanced fibrosis risk, is needed. Using a cohort of primary care patients with MASLD, we aimed to determine the proportion of primary care patients with MASLD that received a referral to GI for evaluation and management (E/M) services.16 Further, we wanted to know how advanced fibrosis risk compared between patients with and without a GI referral. We hypothesized that most of the primary care MASLD patients would not be referred to GI and that a greater proportion of those referred would be at high-risk for advanced fibrosis compared to patients not referred.
Methods
Study design and Setting
The work described has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving human subjects. The Institutional Review Board at MUSC approved this study. This retrospective study of a primary care MASLD cohort used electronic health record (EHR) data from 2012 to 2018 in a primary care clinic at the Medical University of South Carolina (MUSC) in Charleston, SC.
Participants
The cohort included patients with (i) radiographic reports of liver steatosis identified by natural language processing (abdominal ultrasound, CT, or MRI) and (ii) inputs for FIB-4 calculation (aminotransferases and platelet count) within 1 year of imaging.16 Aminotransferases had to be < 350 IU/L, so as not to represent an acute episode of liver pathology not related to advanced fibrosis risk. Patients with competing, non-MASLD chronic liver disease diagnoses (International Classification of Diseases [ICD]-9/10) were excluded (Supplement, Table S1).16 This exclusion extended to heavy alcohol use, as patients with a recorded alcohol history during the study period that exceeded 21 drinks per week in males and 14 drinks per week in females were excluded.17 We also excluded patients with a diagnosis of cirrhosis, complications of cirrhosis, hepatocellular carcinoma, or liver transplantation, identified by a composite of previously validated ICD-9/10 codes, prior to the follow-up period (Supplement).18–20
Outcomes
Referral to Gastroenterology or Hepatology (GI) for evaluation and management (E/M) any time after the imaging results with steatosis was the primary outcome of interest. GI referrals were categorized by the reason for referral and included E/M or a procedural need (e.g. colonoscopy). E/M referrals were further classified as for MASLD, abnormal liver tests, steatosis, or other. For all patients referred, whether the patient attended a specialty visit was recorded.
Primary Predictor Variable
Advanced fibrosis risk assessed by FIB-4 was the primary independent variable of interest. FIB-4 was calculated by study team for each patient in the MASLD cohort using the aminotransferases (AST and ALT) at the time of, or most proximal to (within 1 year), the first abdominal image noting hepatic steatosis. AST and ALT had to both be less than 350 IU/L for inclusion in the calculation. Age at the time of the selected AST/ALT results was used, along with a platelet count from the time of the AST/ALT results, or within the previous 6 months. Each patient had 1 FIB-4 score and FIB-4 was categorized by advanced fibrosis risk: low- (FIB-4 < 1.3), indeterminate- (1.3 ≤ FIB-4 <−2.67), and high-risk (FIB-4 ≥ 2.67).
Covariates
Demographic, vital sign, laboratory, and comorbidity variables were assessed. Demographic variables included in the study were age (continuous), gender (Female / Male), race (Black / non-Black). Clinical variables included body mass index (BMI) as a categorical variable (≥ 30 kg/m2) and bilirubin, AST, ALT, alkaline phosphatase, platelet count, and albumin as continuous variables (not normally distributed). Patterns of liver chemistry abnormality were categorical variables based on “abnormal” thresholds set in the MUSC EHR (AST > 35 IU/L, ALT > 45 IU/L). Categories of abnormality included Elevated AST, Elevated ALT, Elevated AST or ALT, and Elevated AST and ALT. Categorical variables for comorbidities of diabetes mellitus, hypertension and hyperlipidemia were included and identified using Elixhauser ICD-9/10 coding algorithms.21,22 Other categorical clinical variables were evidence of negative viral Hepatitis B testing, negative viral Hepatitis C testing, and alcohol use history (Yes, below exclusion threshold / None / Not recorded).
Statistical Analysis
Patient characteristics were reported overall and by FIB-4 risk category. Age was reported as a mean, categorical variables were reported as proportions, and laboratory variables were reported as medians. Age was compared across FIB-4 risk categories by one-way ANOVA tests, lab variables were compared by Kruskal Wallis tests, and categorical variables were compared with Chi square tests. Proportions of GI specialty referral were calculated by indication and by attendance. Univariate analysis of the cohort variables by outcome of receiving a GI referral was performed, comparing the proportion of patients with high-risk FIB-4 scores by referral status. Chi square tests were used to compare categorical variables, a two-sample t-test was used to compare age, and Mann-Whitney U tests were used to compare lab variables by referral status. Logistic regression models were developed for the binary outcome of receiving a referral to GI for E/M services (Yes / No). An unadjusted model evaluating the association between FIB-4 risk and referral status was developed, as was a model adjusting for the other covariates, chosen a priori. Other than age, AST, and ALT value (which were components of the FIB-4 calculation), all other covariates were considered for inclusion in the adjusted model. We tested multicollinearity in the final models and removed variables that were highly correlated. Statistical analyses were performed using SAS version 9.4.
Results
The cohort included 652 patients that were 64.1% female, 35.9% Black, and had a mean age of 54.7 (SD ± 14.1) years (Table 1, Supplement Figure S1). Of included patients, 73.3% had a BMI ≥ 30 kg/m2, 46.0% (300) had diabetes, 77.8% (507) had hypertension, and 67.8% (442) were diagnosed with hyperlipidemia. Median (IQR) AST and ALT values for the cohort were 26 IU/L (IQR: 20–39) and 28 IU/L (IQR: 19–49), respectively. Of the sample, 37.6% had an elevated AST or ALT. FIB-4 scores were high-risk for advanced fibrosis (FIB-4 ≥ 2.67) in 11.5% (75) of patients, indeterminate-risk (1.3 ≤ FIB-4 < 2.67) for 31.1% (203) of patients, and low-risk (FIB-4 < 1.3) for 57.4% (374) of the sample.
Table 1.
MASLD cohort characteristics for the overall sample and by Fibrosis-4 Index (FIB-4) advanced fibrosis risk category.
| FIB-4 Advanced Fibrosis Risk | |||||
|---|---|---|---|---|---|
|
| |||||
| Overall | Low | Indeterminate | High | p-value | |
| < 1.3 | 1.3 – 2.66 | ≥ 2.67 | |||
| n=652 | n=374 | n=203 | n=75 | ||
| Demographics | |||||
| Age, Mean, years (SD) | 54.7 (± 14.1) | 49.1 (± 13.0) | 62.7 (± 11.6) | 60.9 (± 12.8) | <0.001* |
| Gender % (n) | 0.052† | ||||
| Male | 35.9% (234) | 32.1% (120) | 39.9% (81) | 44.0% (33) | |
| Female | 64.1% (418) | 67.9% (254) | 60.1% (122) | 56.0% (42) | |
| Race % (n) | 0.921† | ||||
| Black | 35.9% (234) | 35.6% (133) | 37.0% (75) | 34.7% (26) | |
| Non-Black | 64.1% (418) | 64.4% (241) | 63.1% (128) | 65.3% (49) | |
|
| |||||
| Clinical Variables, Median (IQR) | |||||
| BMI ≥ 30 kg/m2 % (n) | 73.3% (478) | 75.1% (281) | 74.9% (152) | 60.0% (45) | 0.022† |
| Bili, mg/dL | 0.5 (0.4, 0.8) | 0.5 (0.4, 0.7) | 0.6 (0.4, 0.9) | 0.8 (0.5, 1.2) | <0.001‡ |
| AST, U/L | 26 (20, 39) | 23 (19, 30) | 30 (23, 44) | 67 (36, 118) | <0.001‡ |
| ALT, U/L | 28 (19, 49) | 28 (19, 46) | 27 (19, 50) | 44 (23, 82) | <0.001‡ |
| ALP, U/L | 82 (66, 105) | 79 (64, 100) | 84 (67, 103) | 98 (76, 180) | <0.001‡ |
| Platelets, x109/L | 241 (200, 293) | 273 (232, 318) | 212 (181, 247) | 141 (103,202) | <0.001‡ |
| Albumin g/dL | 3.7 (3.4, 4.0) | 3.8 (3.5, 4.1) | 3.7 (3.2, 4.0) | 3.6 (2.9, 3.9) | 0.003‡ |
|
| |||||
| Liver Chemistry Abnormality % (n) | |||||
| Elevated AST | 31.9% (208) | 19.3% (72) | 37.9% (77) | 78.7% (59) | <0.001† |
| Elevated ALT | 28.7% (187) | 25.4% (95) | 27.1% (55) | 49.3% (37) | <0.001† |
| Elevated AST or ALT | 37.6% (245) | 29.1% (109) | 37.9% (77) | 78.7% (59) | <0.001† |
| Elevated AST and ALT | 23.0% (150) | 15.5% (58) | 27.1% (55) | 49.3% (37) | <0.001† |
|
| |||||
| Comorbidities % (n) | |||||
| Diabetes | 46.0% (300) | 43.6% (163) | 49.3% (100) | 49.3% (37) | 0.353† |
| Hypertension | 77.8% (507) | 72.5% (271) | 85.2% (173) | 84.0% (63) | <0.001† |
| Hyperlipidemia | 67.8% (442) | 62.3% (233) | 77.3% (157) | 69.3% (52) | 0.001† |
|
| |||||
| Other Clinical Variables % (n) | |||||
| Negative HBV Testing | 35.3% (230) | 31.0% (116) | 32.0% (65) | 65.3% (49) | <0.001† |
| Negative HCV Testing | 46.2% (301) | 42.3% (158) | 44.8% (91) | 69.3% (52) | |
| Alcohol Use History | 0.329† | ||||
| Yes, below threshold§ | 37.9% (247) | 34.5% (129) | 42.4% (86) | 42.7% (32) | |
| None | 52.6% (343) | 55.1% (206) | 49.8% (101) | 48.0% (36) | |
| Not recorded | 9.5% (62) | 10.4% (39) | 7.9% (16) | 9.3% (7) | |
One-way ANOVA.
Chi square test.
Kruskal Wallis test. FIB-4=Fibrosis-4 Index. SD=standard deviation. IQR=interquartile range. BMI=body mass index. Bili=bilirubin. AST=aspartate aminotransferase.
ALT=alanine aminotransferase. ALP=alkaline phosphatase.
Overall, 45.7% (298) of patients received a referral to GI, with 100 (15.3% overall) of these referrals being for colonoscopy and 198 (31.1%) being for E/M services (Table 2). Of the cohort, only 8 (1.2%), 13 (2.0%), and 22 (3.4%) were referred for MASLD, abnormal liver tests, or hepatic steatosis, respectively. Of patients referred for evaluation and management, 75.4% (153/198) of patients attended the appointment.
Table 2.
Gastroenterology / Hepatology referrals in the cohort by reason for referral.
| n (%) | |||
|---|---|---|---|
| Overall Sample | 652 | ||
| GI/Hep Referrals* | 298 (45.7%) | ||
| Colonoscopy | 100 (15.3%) | ||
| Evaluation and Management (E/M) | 198 (31.1%) | ||
| MASLD | 8 (1.2%) | ||
| Abnormal liver tests | 13 (2.0%) | ||
| Hepatic steatosis on imaging | 22 (3.4%) | ||
| Other | 155 (23.8%) | ||
| Attended E/M Visit | 153 (23.5%) | ||
| MASLD | 4 (0.6%) | ||
| Abnormal liver tests | 12 (1.8%) | ||
| Hepatic steatosis on imaging | 20 (3.1%) | ||
| Other | 112 (17.2%) | ||
No reason for referral was provided for 5 referral orders. GI/Hep=Gastroenterology/Hepatology. E/M=evaluation and management. MASLD=metabolic dysfunction-associated steatotic liver disease.
Univariate analyses demonstrated no significant difference in the proportion of patients with a high-risk FIB-4 between patients with (11.6%) and without (11.5%, p=0.952) a referral to GI (Figure 1, Table 3). Of the 75 patients with a FIB-4 score at high-risk for advanced fibrosis, 30.7% (23) received a GI referral during the study period. A higher proportion of patients referred to GI had negative viral Hepatitis B testing (41.9%) compared to those not referred (32.4%, p=0.019). Otherwise, there were no significant differences in the other demographic, laboratory, and comorbidity variables by GI referral.
Figure 1.
Proportions of patients in the cohort, referred to gastroenterology / hepatology for evaluation and management, and attending the specialty visit with indeterminate- (FIB-4 ≥ 1.3) and high-risk (FIB-4 ≥ 2.67) FIB-4 scores for advanced fibrosis.
Table 3.
MASLD cohort characteristics by the outcome of referral to Gastroenterology / Hepatology (GI) for evaluation and management (E/M).
| GI Referral for E/M | |||
|---|---|---|---|
|
| |||
| Yes | No | p-value | |
| n=198 | n=454 | ||
| FIB-4 Fibrosis Risk % (n) | |||
| High | 11.6% (23) | 11.5% (52) | 0.952* |
| Indeterminate or High | 45.0% (89) | 41.6% (189) | 0.431* |
|
| |||
| Demographics | |||
| Age, Mean, years (SD) | 54.1 (± 14.5) | 54.9 (± 13.9) | 0.473† |
| Gender % (n) | 0.369* | ||
| Male | 33.3% (66) | 37.0% (168) | |
| Female | 66.7% (132) | 63.0% (286) | |
| Race % (n) | 0.074* | ||
| Black | 30.8% (61) | 38.1% (173) | |
| Non-Black | 69.2% (137) | 61.9% (281) | |
|
| |||
| Clinical Variables, Median (IQR) | |||
| BMI ≥ 30 kg/m2 % (n) | 73.2% (145) | 73.4% (333) | 0.976* |
| Bili, mg/dL | 0.5 (0.4, 0.8) | 0.6 (0.4, 0.8) | 0.477‡ |
| AST, U/L | 26 (21, 39) | 26 (20, 39) | 0.762‡ |
| ALT, U/L | 28 (19, 47) | 29 (20, 49) | 0.361‡ |
| ALP, U/L | 81 (65, 103) | 83 (66, 105) | 0.521‡ |
| Platelets, x109/L | 240 (188, 307) | 241 (206, 290) | 0.748‡ |
| Albumin g/dL | 3.8 (3.4, 4.0) | 3.7 (3.4, 4.0) | 0.561‡ |
|
| |||
| Liver Chemistry Abnormality % (n) | |||
| Elevated AST | 33.8% (67) | 31.1% (141) | 0.484* |
| Elevated ALT | 26.8% (53) | 29.5% (134) | 0.476* |
| Elevated AST or ALT | 39.4% (78) | 36.8% (167) | 0.527* |
| Elevated AST and ALT | 21.2% (42) | 23.8% (108) | 0.472* |
|
| |||
| Comorbidities % (n) | |||
| Diabetes | 50.0% (99) | 44.3% (201) | 0.177* |
| Hypertension | 76.8% (152) | 78.2% (355) | 0.687* |
| Hyperlipidemia | 67.7% (134) | 67.8% (308) | 0.967* |
| Other Clinical Variables | |||
| Negative HBV Testing | 41.9% (83) | 32.4% (147) | 0.019* |
| Negative HCV Testing | 50.0% (99) | 44.5% (202) | 0.195* |
| Alcohol Use History | 0.262* | ||
| Yes, below threshold§ | 42.4% (84) | 35.9% (163) | |
| None | 49.5% (98) | 54.0% (245) | |
| Not recorded | 8.1% (16) | 10.1% (46) | |
Chi square test.
Student’s two sample t-test.
Mann Whitney U test. SD=standard deviation. BMI=body mass index. Bili=bilirubin.
AST=aspartate aminotransferase. ALT=alanine aminotransferase.
ALP=alkaline phosphatase.
The unadjusted logistic regression model demonstrated no significant association between high-risk FIB-4 and receipt of a referral (OR 1.02; 95%CI 0.60 – 1.71; Table 4). After adjustment for other potentially confounding covariates, high-risk FIB-4 was not associated with receiving a referral to GI (OR 1.01; 95% CI 0.59 – 1.71).
Table 4.
Estimated odds ratios and 95% confidence intervals from multivariable logistic regression models for the outcome of Gastroenterology/Hepatology referral for evaluation and management.
| Logistic Regression Models | |||
|---|---|---|---|
| Unadjusted | Adjusted Model 1 | ||
| OR 95% CI | OR 95% CI | ||
| AUC=0.501 | AUC=0.579 | ||
| FIB-4 Fibrosis Risk | |||
| High-Risk (>2.67) | 1.02 0.60 – 1.71 |
1.01 0.59 – 1.71 |
|
| Demographics | |||
| Male gender | 0.79 0.55 – 1.14 |
||
| Black | 0.63 0.43 – 0.92 |
||
| Comorbidities | |||
| BMI ≥ 30 kg/m2 | 0.91 0.61 – 1.36 |
||
| Diabetes | 1.48 1.02 – 2.12 |
||
| Hypertension | 0.91 0.59 – 1.40 |
||
Bilirubin, alkaline phosphatase, albumin, and AST/ALT abnormality were found to be highly correlated (>0.20) with other predictors in the model and excluded from the final adjusted model. Hyperlipidemia was highly correlated with hypertension and excluded as well.
Discussion
In this retrospective cohort study of patients with MASLD in primary care, only 31.1% of patients received a referral to GI for E/M services. Of those patients referred for evaluation and management, 11.6% had FIB-4 scores at high-risk for advanced fibrosis similar to the proportion of patients with a high-risk FIB-4 (11.5%) not receiving a referral. For the entire cohort, only 30.7% (23/75) of patients with a high-risk FIB-4 were referred to GI for evaluation and management. In the logistic regression model, after adjusting for other potentially confounding covariates, high-risk FIB-4 scores were not associated with GI referral.
Fewer than one-third of the patients in this primary care MASLD cohort were referred to GI for evaluation and management during the study period, and for many of those that were referred (78%, 155/198), it is not entirely clear that the reason for referral concerned ongoing MASLD management. These findings suggest that strategies to increase MASLD recognition and detection in primary care may increase referral to GI and Hepatology specialty services. Though current guidelines seek to align referral decisions with advanced fibrosis risk in hopes of optimizing the process (and reducing unnecessary consults), this baseline low rate of referral means that any concerted specialty consultation approach is likely to increase the referral volume. If this is the case, health systems and specialty practices will need to prepare for these changes to meet patient needs and preserve the accessibility of specialty clinicians.
No significant difference in the proportion of patients with high-risk advanced fibrosis assessments by referral status suggests that MASLD disease severity does not play a role in current referral patterns. This finding can most easily be explained by unawareness of measures for non-invasive risk assessment, specifically FIB-4 calculation, in primary care both during the study period and currently.15 Education and dissemination of advanced fibrosis risk assessment measures will play a key role in managing MASLD in primary care and identifying the patients most likely to benefit from specialty referral.7,8 However, the current recommendations call for confirmatory testing in all patients with MASLD and a FIB-4 score of indeterminate-advanced fibrosis risk and higher (FIB-4 ≥ 1.3).7,8 Confirmation can be done with liver stiffness measurements using ultrasound vibration-controlled elastography (VCTE) or magnetic resonance imaging with elastography. Non-invasive confirmatory fibrosis risk assessment can also be completed using proprietary serologic tests like the Enhanced Liver Fibrosis (ELF™) test.8,23,24 Unfortunately, these confirmatory studies are unfamiliar to many primary care clinicians and are not readily available. These results can inform clinicians and health systems on how to approach making one of these confirmatory measures available in primary care and help stakeholders decide when (before or after seeking specialty GI consultation) and where (radiology, specialty care, primary care) confirmatory testing should occur. The specialty society guidelines recommend confirmatory testing before referring to GI, which seems appropriate given that 43% (278/652) of this sample had indeterminate-risk or greater (≥ 1.3) FIB-4 scores in this sample.7,8,14,23 Finding advanced fibrosis in primary care patients with MASLD matters more than ever before, as evidence continues to grow for the benefits of weight loss and exercise on halting fibrosis progression and reducing liver fat.25,26 Additionally, in recent ground-breaking news, resmetirom (a thyroid hormone receptor-β agonist) became the first United States Food and Drug Administration (FDA)-approved pharmacotherapy for MASLD with moderate to advanced liver fibrosis after demonstrating improvements in steatohepatitis and fibrosis stage in a clinical trial.27
The area under the curve (AUC=0.579) for the adjusted logistic regression model suggests that none of the variables chosen in this study adequately explain patterns of GI referral in patients with MASLD. Specialty referral is a complex topic that has only recently begun to draw significant attention in health services research.6,28 Disease-severity driven guidelines can help to clarify this process for patients, primary care, specialty services, and health systems. Work to better understand this process is necessary to improve care delivery and address healthcare disparities. In our adjusted logistic regression model, when controlling for the other covariates in the final model, patients identified as Black had a lower odds of receiving a referral to GI (OR 0.63; 95%CI 0.43 – 0.92). This finding may demonstrate an ongoing disparity in care that requires urgent attention. Creating clear, disease severity-driven referral processes with inputs from all stakeholders (patients, primary care providers, GI specialists, and health systems) will hopefully optimize the referral process and improve health equity.
We acknowledge limitations in this study. First, the primary predictor variable is a FIB-4 score which we recognize was unlikely to be calculated in primary care during the study period. However, we feel that this risk score still helps to describe MASLD disease severity and will play a critical role in future referral practices. Also, this cohort was developed using radiographic evidence of hepatic steatosis and it is possible that MASLD was under-recognized in the primary care setting, which would bias results toward under-referral. Additionally, we used ICD-9/10 codes to exclude patients with cirrhosis, complications of cirrhosis, and HCC. While ICD-9/10 codes have limited sensitivity for identifying chronic diseases in administrative health records, we did use a composite of diagnosis codes that have demonstrated good accuracy for cirrhosis detection in previous studies.18–20 Lastly, this data comes from a single clinic which could threaten generalizability.
Conclusion
Few primary care patients with MASLD were referred to GI for evaluation and management. Further, those patients referred do not differ in MASLD advanced fibrosis risk from patients not receiving a GI referral. Implementing MASLD referral practices focused on disease severity may substantially improve ongoing efforts to address the MASLD epidemic.
Supplementary Material
Funding Sources:
National Institute of Diabetes and Digestive and Kidney Diseases (NIH/NIDDK K23DK118200 PI: Schreiner; R03DK129558 PI: Schreiner). This project was also supported by the South Carolina Clinical & Translational Research Institute with an academic home at the Medical University of South Carolina CTSA National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH) under UL1 TR001450.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest: All authors report no conflicts of interest with this work.
References
- 1.Lazarus JV, Anstee QM, Hagstrom H, et al. Defining comprehensive models of care for NAFLD. Nat Rev Gastroenterol Hepatol 2021;18(10):717–729. [DOI] [PubMed] [Google Scholar]
- 2.Lazarus JV, Mark HE, Anstee QM, et al. Advancing the global public health agenda for NAFLD: a consensus statement. Nat Rev Gastroenterol Hepatol 2021. [DOI] [PubMed] [Google Scholar]
- 3.Kanwal F, Shubrook JH, Younossi Z, et al. Preparing for the NASH Epidemic: A Call to Action. Diabetes care 2021. [DOI] [PubMed] [Google Scholar]
- 4.Rinella ME, Lazarus JV, Ratziu V, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature. J Hepatol 2023;79(6):1542–1556. [DOI] [PubMed] [Google Scholar]
- 5.Porter J, Boyd C, Skandari MR, Laiteerapong N. Revisiting the Time Needed to Provide Adult Primary Care. J Gen Intern Med 2023;38(1):147–155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Song Z, Sequist TD, Barnett ML. Patient referrals: a linchpin for increasing the value of care. Jama 2014;312(6):597–598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kanwal F, Shubrook JH, Adams LA, et al. Clinical Care Pathway for the Risk Stratification and Management of Patients With Nonalcoholic Fatty Liver Disease. Gastroenterology. 2021;161(5):1657–1669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD Practice Guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology (Baltimore, Md). 2023;77(5):1797–1835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hagstrom H, Nasr P, Ekstedt M, et al. Fibrosis stage but not NASH predicts mortality and time to development of severe liver disease in biopsy-proven NAFLD. J Hepatol 2017;67(6):1265–1273. [DOI] [PubMed] [Google Scholar]
- 10.Sanyal AJ, Van Natta ML, Clark J, et al. Prospective Study of Outcomes in Adults with Nonalcoholic Fatty Liver Disease. N Engl J Med 2021;385(17):1559–1569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Taylor RS, Taylor RJ, Bayliss S, et al. Association Between Fibrosis Stage and Outcomes of Patients With Nonalcoholic Fatty Liver Disease: A Systematic Review and Meta-Analysis. Gastroenterology. 2020;158(6):1611–1625 e1612. [DOI] [PubMed] [Google Scholar]
- 12.Vallet-Pichard A, Mallet V, Nalpas B, et al. FIB-4: an inexpensive and accurate marker of fibrosis in HCV infection. comparison with liver biopsy and fibrotest. Hepatology (Baltimore, Md). 2007;46(1):32–36. [DOI] [PubMed] [Google Scholar]
- 13.Schreiner AD, Zhang J, Durkalski-Mauldin V, et al. Advanced Liver Fibrosis and the Metabolic Syndrome in a Primary Care Setting. Diabetes Metab Res Rev 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Davyduke T, Tandon P, Al-Karaghouli M, Abraldes JG, Ma MM. Impact of Implementing a “FIB-4 First” Strategy on a Pathway for Patients With NAFLD Referred From Primary Care. Hepatol Commun 2019;3(10):1322–1333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Islam KB, Brandman D, Chu JN, Goldman ML, Fox RK. Primary Care Providers and Nonalcoholic Fatty Liver Disease: A Needs Assessment Survey. Dig Dis Sci 2023;68(2):434–438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Nielsen EM, Anderson KP, Marsden J, Zhang J, Schreiner AD. Nonalcoholic Fatty Liver Disease Underdiagnosis in Primary Care: What Are We Missing? J Gen Intern Med 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chalasani N, Younossi Z, Lavine JE, et al. The diagnosis and management of nonalcoholic fatty liver disease: Practice guidance from the American Association for the Study of Liver Diseases. Hepatology (Baltimore, Md). 2018;67(1):328–357. [DOI] [PubMed] [Google Scholar]
- 18.Mapakshi S, Kramer JR, Richardson P, El-Serag HB, Kanwal F. Positive Predictive Value of International Classification of Diseases, 10th Revision, Codes for Cirrhosis and Its Related Complications. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2018;16(10):1677–1678. [DOI] [PubMed] [Google Scholar]
- 19.Nehra MS, Ma Y, Clark C, Amarasingham R, Rockey DC, Singal AG. Use of administrative claims data for identifying patients with cirrhosis. Journal of clinical gastroenterology. 2013;47(5):e50–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hagström H, Adams LA, Allen AM, et al. Administrative Coding in Electronic Health Care Record-Based Research of NAFLD: An Expert Panel Consensus Statement. Hepatology (Baltimore, Md). 2021;74(1):474–482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Quan H, Li B, Saunders LD, et al. Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res 2008;43(4):1424–1441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005;43(11):1130–1139. [DOI] [PubMed] [Google Scholar]
- 23.Srivastava A, Gailer R, Tanwar S, et al. Prospective evaluation of a primary care referral pathway for patients with non-alcoholic fatty liver disease. J Hepatol 2019;71(2):371–378. [DOI] [PubMed] [Google Scholar]
- 24.European Association for the Study of the L, List of panel m, Berzigotti A, et al. Easl Clinical Practice Guidelines (Cpgs) On Non-Invasive Tests For Evaluation Of Liver Disease Severity And Prognosis-2020 Update. J Hepatol 2021. [DOI] [PubMed] [Google Scholar]
- 25.Vilar-Gomez E, Martinez-Perez Y, Calzadilla-Bertot L, et al. Weight Loss Through Lifestyle Modification Significantly Reduces Features of Nonalcoholic Steatohepatitis. Gastroenterology. 2015;149(2):367–378 e365; quiz e314–365. [DOI] [PubMed] [Google Scholar]
- 26.Stine JG, DiJoseph K, Pattison Z, et al. Exercise Training Is Associated With Treatment Response in Liver Fat Content by Magnetic Resonance Imaging Independent of Clinically Significant Body Weight Loss in Patients With Nonalcoholic Fatty Liver Disease: A Systematic Review and Meta-Analysis. Am J Gastroenterol 2023;118(7):1204–1213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Harrison SA, Bedossa P, Guy CD, et al. A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis. N Engl J Med 2024;390(6):497–509. [DOI] [PubMed] [Google Scholar]
- 28.Mehrotra A, Forrest CB, Lin CY. Dropping the baton: specialty referrals in the United States. Milbank Q 2011;89(1):39–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.