A sponsored blog by Veradigm®
Nonalcoholic fatty liver disease (NAFLD), including the more aggressive NASH (nonalcoholic steatohepatitis), has become a major public health concern, affecting about one in four Americans (Younossi 2018). Its increasing prevalence parallels rising rates of obesity and type 2 diabetes.
Real-world studies have evaluated diagnostic gaps, risk factors, clinical predictors, and long-term outcomes for NAFLD.The real-world evidence (RWE) obtained from these studies demonstrates an ongoing need for identifying individuals with NAFLD, stratifying risk, and referring patients to specialists.
This blog presents a retrospective cohort study using de-identified real-world data sourced from a cloud-based, U.S. electronic health record (EHR) dataset offered by Veradigm® as part of its Health Insights database. The dataset includes ambulatory patients seen by primary care and specialty healthcare providers (practice size ranging from one to four clinicians).
The objectives of the current analysis were to:
- Establish the prevalence of NAFLD over a five-year period;
- Characterize the all-inclusive NAFLD cohort and NAFL and NASH subgroups; and,
- Identify which provider types are seeing NAFLD patients.
Entry criteria included:
- Any diagnosis code recorded during the five-year period between June 15, 2014 – June 15, 2019 inclusive (Intake), with at least one office visit with a healthcare provider (HCP) >6 months prior to the most recent diagnosis;
- A diagnosis code for NASH, NAFL, or NAFLD recorded during intake (index);
- No diagnoses related to alcohol abuse, alcoholic steatosis, or alcoholic hepatitis;
- Eighteen (18) years of age or older at index.
The complete research approach is described in a White Paper. Please reach out to undine@scientist.com for white paper content.
Five-year Period Prevelance of NAFLD/NASH
Of nearly 12 million individuals meeting entry criteria for any diagnosis code and an HCP office visit, fewer than 1% met additional criteria for a NAFLD, NAFL, or NASH diagnosis; alcohol use/code restrictions; and age. Of these patients, 6% had diagnosis codes for NASH, the more progressive inflammatory form of NAFLD.
The lower than expected prevalence of NAFLD and NASH (Younossi 2018) likely reflects an underreporting of NAFLD-related diagnoses in primary care and may be due, in part, to placement of diagnoses in unstructured clinical notes within the EHR. Such data could be identified, extracted, and presented in structured format using natural language processing (NLP). Veradigm has used NLP to extract and supplement HbA1c levels (by 41%), left ventricular ejection fraction (by 100%), and DXA scores (by 81%) for patients with type 2 diabetes, heart failure, and osteoporosis, respectively.
Cardiometabolic Risk and Clinical Implications
Higher percentages of NASH than NAFL patients were obese (based on BMI), had hypertension (based on systolic blood pressure and diagnosis codes), diabetes (based on HbA1c levels and diagnosis codes), and had metabolic syndrome (based on three or more risk factors and diagnosis codes).
Compared with the NAFL subgroups, individuals in the NASH subgroup were significantly more likely to have BMI ≥30 (58.7% vs 60.8%) and systolic blood pressure ≥130 mmHg (42.5% vs 45.4%) (both P≤0.001). A higher percentage of patients in the NAFL than in the NASH subgroup had HbA1c levels suggestive of prediabetes (Figure 1). More patients in the NASH subgroup had HbA1c levels ≥6.5%, the recommended cutoff for diagnosing type 2 diabetes, and fasting blood glucose levels >100mg/dL (27.7% vs 23.5%. respectively).
Most NAFLD patients had additional diagnoses for one or more of six cardiometabolic conditions, as shown in Figure 2 below.
Cardiometabolic risk contributes substantially to the overall clinical burden of NAFLD. Our analysis suggests that not all patients are being managed for cardiometabolic conditions. Overall, nearly two-thirds of NAFLD patients had a diagnosis for dyslipidemia. Of these patients, more than one-third did not have prescriptions for or documented use of antilipidemic medications. Approximately one-third of patients had a diagnosis of type 2 diabetes, and nearly 20% did not have prescriptions for or documented use of antidiabetic medications.
Metabolic syndrome appears to be underdiagnosed or underrecognized. While 3,708 (3.6%) patients in the NAFLD cohort had a diagnosis for metabolic syndrome, 44,045 NAFLD patients had evidence of metabolic syndrome based on individual risk factors (i.e., diagnosis codes for three or more of the following: prediabetes/insulin resistance, type 2 diabetes, overweight/obesity, dyslipidemia, and hypertension).
Fibrosis Scores
Up to 500 NAFL patients in our analysis may be at risk of progressing to decompensated liver disease absent timely confirmation of advancing liver fibrosis or intervention.
A higher percentage of patients in the NAFL subgroup than in the NASH subgroup had calculated FIB‑4 scores corresponding to “low risk” (65.0% vs 52.9%). The FIB-4 Index is a non-invasive, validated test for detecting advanced fibrosis (Dyson 2014; Chalasani 2018). Conversely, higher percentages of patients in the NASH subgroup than in the NAFL subgroup had calculated FIB-4 scores corresponding to “intermediate risk” (32.4% vs 28.9%) and “high risk” (14.7% vs 6.2%).
FIB-4 scores indicated that over 380 patients with NASH and over 2,500 patients with NAFL were at high risk of fibrosis. Nevertheless, these NAFL patients had no recorded diagnoses codes for NASH.
Percentagewise, significantly more NASH than NAFL patients had recorded diagnoses for cirrhosis (11.1% vs 1.8%) and hepatocellular carcinoma (0.4% vs 0.1%) (P>0.001).
Provider Specialty
Most patients with diagnoses for NAFLD, NAFL, or NASH were seeing primary care practitioners, with fewer than 10% of patients in each of these groups having documented visits with gastroenterologists and hepatologists.
Past research has shown that less than one-half of primary care providers and non-gastroenterologic/non-hepatologic subspecialty providers are comfortable managing NAFLD. Just one in three were comfortable referring patients with suspected NAFLD to gastroenterologists or hepatologists (Wieland 2013).
What We Learned: Practical Applications
- Fibrosis scores and other clinical data from EHRs, patient registries, and transactional claims may be leveraged by life sciences stakeholders to, for example, develop a deeper understanding of the NAFL-to-NASH-to-cirrhosis progression or to examine resource utilization by NAFLD patients according to cardiometabolic and hepatic risk.
- Availability of best-practice guidelines from within EHRs and registries may increase adoption of evidence-based care plans and, when warranted, support timely referrals from primary care providers to specialists to avoid complications and forestall end-stage liver disease in at-risk individuals.
- Additionally, medical educational materials may be shared by HCPs with their patients at the point of care or through patient-provider portals for the purpose of promoting health literacy, to enable shared decision-making and encourage patients to participate as partners in their own health and well-being.
Future Inquiries
Veradigm is working towards full-scale adoption of artificial intelligence and related technologies for managing RWD. These custom-built analytic tools may be used to extract and de-identify patient data from unstructured clinical notes available in Veradigm ambulatory EHRs and registries.
Veradigm offers data enrichment solutions to healthcare stakeholders, including leaders in HEOR and RWE, population health, and epidemiology, along with others in the life sciences who are interested in gaining timely, in-depth insights into real-world populations, for purposes of addressing challenging research questions, optimizing clinical outcomes, and advancing patient care.
For more information on datasets and analytic services offered by Veradigm, contact us.
References
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 2018;67(1):328-357. doi: 10.1002/hep.29367. Epub 2017 Sep 29.
Dyson JK, McPherson S, Anstee Q. Non-alcoholic fatty liver disease: non-invasive investigation and risk stratification. Postgrad Med J 2014;90:254-266. doi: 10.1136/jclinpath-2013-201620.
Wieland AC, Quallick M, Truesdale A, et al. Identifying practice gaps to optimize medical care for patients with nonalcoholic fatty liver disease. Dig Dis Sci 2013;58(10):2809-2816. doi: 10.1007/s10620-013-2740-8.
Younossi Z, Anstee QM, Marietti M, et al. Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol 2018;15(1):11-20. doi: 10.1038/nrgastro.2017.109.
