Fatty Liver Disease in Germany: What the FLAG Study Reveals About Diagnosis, Treatment Gaps, and the Path Forward

Table of Contents

• Key Points
• Why This Research Matters
• How the Study Was Conducted
• Key Finding #1: How Common Is Advanced Liver Fibrosis?
• Key Finding #2: Who Is Most Affected by Advanced Fibrosis?
• Key Finding #3: Lifestyle Factors and Interventions
• Key Finding #4: Use of Potentially Helpful Medications
• Key Finding #5: Identifying Patients Who May Need Future Treatments
• Key Finding #6: What Happened After 12 Months
• What These Findings Mean for Patients
• Study Limitations
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• In the FLAG study of 507 German fatty liver patients, 10% had advanced liver fibrosis at first specialist visit.
• Patients with advanced fibrosis had markedly higher rates of diabetes (58%) and hypertension (79%).
• Only 6% of advanced fibrosis patients exercised more than twice weekly; about half did no exercise.
• Weight loss improved liver enzymes: ALT dropped 20% after losing weight, while weight gain increased ALT 26%.
• Potentially helpful medications, like GLP-1 agonists, were rarely used even in diabetic patients in this cohort.

Why This Research Matters

Non-alcoholic fatty liver disease (NAFLD) is a condition in which fat builds up in the liver of people who drink little or no alcohol. Globally, it is the most common liver disease, with an estimated prevalence of 24%. It is not a single condition but rather a spectrum that ranges from harmless fat accumulation (steatosis) to a more dangerous inflammatory form called non-alcoholic steatohepatitis (NASH), which can progress to liver cirrhosis, liver failure, and even liver cancer (hepatocellular carcinoma).

For patients, NAFLD means more than just a lab value. It is linked to impaired quality of life and carries the risk of developing end-stage liver disease with serious complications. At the societal level, the disease generates high economic and healthcare costs. In 2013, end-stage liver disease related to NAFLD was the second most common reason for liver transplantation in the United States.

Despite this heavy burden, there is a surprising lack of information about how NAFLD patients are actually cared for outside of clinical trials. Regulatory trials test new drugs in highly selected patients whose disease stage is confirmed by liver biopsy. But real-world data—what actually happens in doctors' offices and clinics—are largely unknown. This gap matters because new NASH medications are on the horizon. In fact, a first trial studying obeticholic acid versus placebo reported a positive interim analysis showing fibrosis regression after 18 months of treatment. Knowing who could benefit from these treatments requires understanding the real patient population.

Previous German data gave conflicting pictures: a tertiary care cohort using liver biopsy found a 15.6% prevalence of advanced non-cirrhotic fibrosis (F3), while population-based estimates suggested only 400,000 F3 cases nationwide. The FLAG study was designed to help close this knowledge gap.

How the Study Was Conducted

The FLAG study is a prospective (forward-looking) observational real-world cohort study initiated by the Association of Gastroenterologists in Private Practice (Berufsverband Niedergelassener Gastroenterologen Deutschlands) in cooperation with academic medical centers and the German Liver Foundation. It covers both secondary care (specialist practices) and tertiary care (university hospitals) levels.

Data were collected from 13 sites across Germany: 9 office-based gastroenterology practices and 4 academic outpatient clinics. Patient baseline data were recorded between May 2017 and October 2019, with a mean recruitment rate of 17 patients per month. Data collection used electronic case report forms, and data quality was verified by plausibility checks and off-site monitoring. The study protocol was approved by the local ethics committee (Ärztekammer Berlin, Germany; protocol number 51/16), and all patients gave written informed consent.

To be included, patients had to be men and women aged 18 years or older with a NAFLD diagnosis based on:

• Hepatic steatosis (fatty liver) confirmed by ultrasound or controlled attenuation parameter (CAP) measurements
• Availability of clinical, technical, and laboratory data needed to calculate non-invasive fibrosis scores, including the fibrosis-4 (FIB-4) index, the AST-to-platelet ratio index (APRI) score, the NAFLD fibrosis score, and liver stiffness measurements (LSM)
• Data available to assess components of the metabolic syndrome
• Alcohol consumption below 30 g/day for men and below 20 g/day for women

Patients were excluded if they had other chronic liver diseases, such as alcoholic liver disease, chronic hepatitis B or C infections, autoimmune or cholestatic liver disease, hemochromatosis, alpha-1-antitrypsin deficiency, or Wilson disease. Patients with a history of hepatotoxic medication (methotrexate, amiodarone, long-term NSAIDs) or malignant diseases within 12 months before enrollment were also excluded.

Laboratory testing included liver enzymes (AST, ALT, GGT), blood cell counts, lipid profiles, serum ferritin, and HbA1c measurements. The presence of metabolic syndrome components was assessed using waist circumference, body mass index (BMI), arterial hypertension, and type 2 diabetes mellitus (T2DM). Former cardiovascular events (CVEs) were also recorded.

Non-invasive fibrosis scores were used to classify patients into three groups: no significant fibrosis, indeterminate (intermediate) stage, and advanced fibrosis. The FIB-4 index used lower cut-off <1.45 and higher cut-off >2.67 (or alternatively >3.25). The APRI score used cut-offs of <0.50 and >1.50. The NAFLD fibrosis score used cut-offs of <−1.455 and >0.676. Liver stiffness measurements (LSMs) used cut-offs of 8.2 kPa and 9.6 kPa. FibroScan® liver stiffness measurements were available in 251 patients (50%), and CAP measurements in 107 patients (21%).

Statistical analysis used descriptive statistics, non-parametric chi-square tests for categorical variables, and Kruskal-Wallis or Wilcoxon tests for continuous variables. A p-value of less than 0.05 was considered statistically significant.

Key Finding #1: How Common Is Advanced Liver Fibrosis?

The study included 507 patients with a mean age of 53 years; 268 were men (53%) and 239 were women (47%). More than two-thirds of patients (n = 360; 71%) were recruited at office-based practices, with the remaining 147 (29%) at academic sites. Patients were predominantly Caucasian (89%).

Using the FIB-4 index with the cut-off of >2.67 for advanced fibrosis (a threshold shown to have increased sensitivity for detecting advanced fibrosis in a recent meta-analysis), the cohort broke down as follows:

• 64% (n = 324) had no significant fibrosis
• 26% (n = 130) had an indeterminate (intermediate) stage
• 10% (n = 53) had advanced fibrosis

That means every 10th patient arriving at a German gastroenterology practice already had advanced liver fibrosis. When the original higher cut-off of >3.25 was used, the results were similar: 64% no significant fibrosis, 28% indeterminate, and 8% advanced fibrosis.

Different scoring tools gave different pictures, illustrating why multiple tests are used together. The NAFLD fibrosis score (calculated in 366 patients, since albumin data was incomplete in some) classified 37% as no significant fibrosis, 30% as indeterminate, and 33% as advanced fibrosis. The APRI score classified far fewer patients as advanced (1%), while liver stiffness measurement classified 27% as advanced fibrosis. Notably, there were no significant differences in fibrosis stage between patients recruited at office-based practices and those at academic sites.

Key Finding #2: Who Is Most Affected by Advanced Fibrosis?

Patients with advanced fibrosis had a distinct profile. The following differences between the three groups (no significant fibrosis, indeterminate, advanced fibrosis) were statistically significant:

• Age: 48 years vs. 61 years vs. 65 years (p <0.001)
• Waist circumference: 103 cm vs. 108 cm vs. 108 cm (p = 0.007)
• AST (aspartate aminotransferase): 39 U/L vs. 51 U/L vs. 66 U/L (p <0.001)
• GGT (gamma-glutamyltransferase): 89 U/L vs. 122 U/L vs. 231 U/L (p <0.001)
• Platelets: 267 vs. 207 vs. 133 g/dl (p <0.001)—lower platelets are a sign of advancing liver disease
• Ferritin: 223 vs. 305 vs. 352 mg/dl (p = 0.002)
• Liver stiffness measurement: 7.5 kPa vs. 10.7 kPa vs. 24.1 kPa (p <0.001)

The frequency of co-morbidities increased sharply with fibrosis stage:

• Type 2 diabetes: 21% (no fibrosis) vs. 41% (indeterminate) vs. 58% (advanced fibrosis) (p <0.001)
• High blood pressure (hypertension): 44% vs. 61% vs. 79% (p <0.001)
• Previous cardiovascular events: 3% vs. 10% vs. 11% (p = 0.001)

Obesity was common across all groups (56%, 62%, and 64%, respectively), and although it trended upward with fibrosis stage, this did not reach statistical significance (p = 0.302). Interestingly, CAP measurements (a measure of liver fat) did not differ significantly between the groups (p = 0.202), confirming that it is the fibrosis (scarring), not the fat itself, that matters most for clinical outcomes.

Key Finding #3: Lifestyle Factors and Interventions

Because lifestyle modification is the cornerstone of NAFLD management, researchers assessed smoking, alcohol, exercise, and nutritional counseling. There were no significant differences between fibrosis groups regarding smoking and alcohol consumption.

Alcohol consumption: Between 31% and 38% of patients reported drinking no alcohol; roughly half reported occasional drinking; and 11–15% reported regular drinking. (By protocol, no patient exceeded 30 g/day for men or 20 g/day for women.)

Smoking: Current smokers made up 23% of the no-fibrosis group, 12% of the indeterminate group, and 16% of the advanced fibrosis group.

Physical exercise: This is where the study found one of its most concerning results. Approximately half of all patients in each group reported no exercise at all (48%, 54%, and 57%). Another 27–37% exercised less than twice per week. The widely recommended level of more than 2 exercise sessions per week was reported by:

• 17% of patients with no significant fibrosis
• 19% of patients with indeterminate stage
• Only 6% of patients with advanced fibrosis

Nutritional counseling: Approximately 25% of patients had never received nutritional counseling (25% in the no-fibrosis group, 23% in the indeterminate group, and 27% in the advanced fibrosis group). This means that even patients with the most advanced disease often missed out on dietary support—despite diet being a central pillar of NAFLD care.

Key Finding #4: Use of Potentially Helpful Medications

While no medication is currently approved specifically for NAFLD, several drugs used for other conditions have shown potential benefit in NAFLD patients. The study examined how often these were prescribed, and the overall picture is one of very low usage.

In patients with type 2 diabetes (n = 153):

• GLP-1 agonists (a class of diabetes drugs with anti-fibrotic effects in trials): only 6 patients (3.9%)
• Statins: 59 patients (38.6%)
• Acetylsalicylic acid (aspirin): 31 patients (20.3%)
• Vitamin E: 1 patient (0.7%)
• Vitamin D: 31 patients (20.3%)
• Silymarin (milk thistle): 1 patient (0.7%)
• Ursodeoxycholic acid: 4 patients (2.6%)
• Metformin: 88 patients (57%)
• Gliptins: 25 patients (16%)
• Insulin: 42 patients (27%)

In patients without diabetes (n = 354):

• GLP-1 agonists: only 1 patient (0.3%)
• Statins: 40 patients (11.5%)
• Acetylsalicylic acid: 22 patients (6.3%)
• Vitamin E: 4 patients (1.2%)
• Vitamin D: 50 patients (14.5%)
• Silymarin: 11 patients (3.2%)
• Ursodeoxycholic acid: 5 patients (1.4%)

Notably, GLP-1 agonists—which have shown anti-fibrotic effects in phase II randomized controlled trials in NAFLD patients and are available as a weight-lowering medication even in non-diabetic patients—were rarely used. Vitamin E, which has been studied together with pioglitazone in a randomized controlled trial for histologically proven non-diabetic NASH, was recorded in only 5 patients total (1 with diabetes, 4 without). The researchers noted that statins, aspirin, ursodeoxycholic acid, and vitamin D were more frequently used in the diabetes group, while silymarin was more often prescribed in patients without diabetes.

Key Finding #5: Identifying Patients Who May Need Future Treatments

As NASH pharmacotherapies are expected to arrive on the market, identifying which patients would benefit most becomes critical. The FLAG study used two additional tools to estimate this "population to treat."

The FibroScan-AST (FAST) score—which combines liver stiffness measurement (kPa), CAP (dB/m), and AST (U/L)—was designed to identify patients with NASH, significant inflammatory activity, and fibrosis who may be candidates for new therapies. In the 107 patients for whom the FAST score could be calculated:

• 48.6% had a score below 0.35 ('rule out NASH')
• 34.6% fell into the "grey zone" (uncertain)
• 16.8% had a score above 0.67 ('rule in NASH'), meaning they likely had progressive NASH

Breaking this down by fibrosis stage, 63.6% of patients with advanced fibrosis (by FIB-4) had FAST scores indicating progressive NASH, compared to 23.3% of the indeterminate group and only 6.1% of the no-significant-fibrosis group.

Using an alternative approach—a liver stiffness measurement cut-off of >9.1 kPa proposed to identify significant fibrosis (≥F2)—29.3% of the 251 patients with available LSM data could be considered a "population to treat." This proportion rose dramatically with fibrosis stage: 20.1% for no significant fibrosis, 30.0% for indeterminate, and an overwhelming 84.0% for advanced fibrosis.

More than half of the patients for whom the FAST score was available were recruited at office-based practices (60 out of 107), suggesting that this kind of risk stratification is feasible in routine outpatient care.

Key Finding #6: What Happened After 12 Months

A subset of patients—117 out of 507 (23%)—completed a follow-up visit at year 1, with a mean follow-up time of 12.4 (1.6) months. Most of these patients (104 out of 117) were recruited at office-based practices.

Over the follow-up period, body weight changed as follows:

• Weight remained stable: 14 patients (12%)
• Weight increased: 48 patients (41%)
• Weight decreased: 55 patients (47%)

Of those who lost weight, 17% lost more than 5% of their body weight—a threshold that is generally considered clinically meaningful in NAFLD. Baseline BMI differed significantly among those who would go on to remain stable, gain weight, or lose weight (p = 0.009).

The liver enzyme changes were striking. In patients who lost weight:

• AST decreased by 11%
• ALT decreased by 20%
• GGT decreased by 14%

In patients who gained weight, the opposite occurred:

• AST increased by 24%
• ALT increased by 26%
• GGT increased by 22%

These findings reinforce that even modest weight changes have a direct and measurable impact on liver health, and that weight gain actively harms the liver. Only a few new cardiovascular events or co-morbidities occurred in this relatively small follow-up subgroup.

What These Findings Mean for Patients

This is the first report on NAFLD from a secondary-care real-world cohort in Germany, and its implications are significant for patients and healthcare systems alike.

First, the disease is already advanced when patients reach specialists. One in 10 patients had advanced fibrosis at baseline, and among patients with advanced fibrosis, the rates of diabetes (58%) and hypertension (79%) were extremely high. This suggests that NAFLD is not being detected early enough in primary care, and that opportunities to intervene before cirrhosis develops are being missed.

Second, the "standard of care" is not meeting the need. Despite guidelines recommending lifestyle modification, only about a quarter of patients had received nutritional counseling, and only a small minority achieved the recommended exercise level. Even more concerning, the patients who needed lifestyle changes most—those with advanced fibrosis—had the lowest exercise rates (only 6% exercised more than twice per week).

Third, pharmacotherapy options are underused. Given that GLP-1 agonists have shown anti-fibrotic effects in trials and can aid weight loss, very few patients in this cohort received them (3.9% of those with diabetes; 0.3% of those without). This suggests a window of opportunity exists for more proactive medication management.

Fourth, weight matters—measurably. The follow-up data confirm that losing weight reduces liver enzymes (ALT down 20% in weight losers), while gaining weight increases them (ALT up 26% in weight gainers). This provides real-world evidence that patients can influence their liver disease progression through weight management.

Looking at the broader German context: a mathematical model estimated the overall prevalence of NAFLD in Germany at 23%, including only 3.3% with F3–F4 fibrosis (about 600,000 cases). By contrast, an academic care cohort found F3 fibrosis alone represented 16% of all cases. The FLAG cohort, with 10% advanced fibrosis, sits between these two extremes and likely reflects the reality of specialist care—a biased sample towards more advanced disease compared to the general population, but not as severely selected as tertiary referral centers.

Study Limitations

As with any study, the FLAG cohort has limitations that should be acknowledged:

• No liver biopsies: Fibrosis stage was determined using non-invasive surrogate scores rather than liver histology (biopsy), which is the gold standard. While these scores are well-validated, they can misclassify some patients.
• Moderate sample size: The cohort of 507 patients is relatively small compared to population-based registries, and the age range did not show broad variation, so the researchers did not adjust FIB-4 calculations by age as some other studies have done.
• Incomplete data for some scores: The NAFLD fibrosis score could only be calculated in 366 of 507 patients due to missing albumin data, and liver stiffness measurements were only available in 251 patients (50%), with CAP in just 107 (21%).
• Referral bias: Because the referral rate of NAFLD patients from primary care physicians to specialists is largely unknown in Germany, this cohort may be biased towards more advanced disease. The results describe patients who actually reached specialist care, not all NAFLD patients in the population.
• Low follow-up rate: Only 23% of the cohort (117 patients) completed the year-1 follow-up visit, limiting the strength of the longitudinal conclusions.
• Self-reported lifestyle data: Exercise and alcohol consumption were assessed semi-quantitatively and relied on patient self-reporting, which may be subject to recall bias.
• Observational design: As an observational real-world study, this cannot prove cause-and-effect relationships—for example, that weight loss directly caused the enzyme improvements, though the biological plausibility is strong.

Recommendations for Patients

Based on the FLAG study findings and current medical guidelines, here is actionable advice for patients living with fatty liver disease:

1. Aim for weight loss of at least 5–10%. The study showed that even modest weight changes improved liver enzymes (ALT reduced by 20% in those who lost weight). Weight loss of more than 5% body weight is considered clinically meaningful, and greater weight loss generally produces greater liver benefits.
2. Exercise more than twice per week. This was the recommended threshold in the study, yet only 6% of advanced fibrosis patients achieved it. Regular aerobic and resistance exercise helps reduce liver fat independent of weight loss.
3. Seek nutritional counseling. Only about 25% of patients in this study had received nutritional counseling. A structured diet plan—typically emphasizing reduced calories, limited refined carbohydrates and sugars, and increased vegetables and lean proteins—can make a meaningful difference.
4. Limit alcohol strictly. The study excluded anyone drinking more than 30 g/day (men) or 20 g/day (women), but even "regular" drinking within these limits may be worth discussing with your doctor, as any alcohol can stress a fatty liver.
5. Ask your doctor about your fibrosis stage. If you have NAFLD, ask whether non-invasive tests like FIB-4, liver stiffness measurement (FibroScan), or the FAST score have been done. Knowing your fibrosis stage matters because it determines your risk and monitoring frequency.
6. Manage co-morbidities aggressively. The strong link between NAFLD, diabetes, and hypertension means that controlling blood sugar and blood pressure is essential. If you have diabetes (present in 58% of advanced fibrosis patients in this study), ensure your medications—including options like GLP-1 agonists—are optimized.
7. Advocate for regular follow-up. The FLAG study showed that follow-up care is inconsistent. Ask your doctor when you should return, and make sure liver function tests and fibrosis assessments are repeated to track changes over time.
8. Stay informed about emerging treatments. NASH pharmacotherapies are anticipated in the near future. While lifestyle changes remain the cornerstone of treatment today, new medications may soon offer additional options—particularly for patients with progressive NASH identified by tools like the FAST score.

In conclusion, the FLAG study provides a clear picture of NAFLD care in Germany: too many patients present late with advanced disease, and too few receive the structured lifestyle interventions and medications that could change their trajectory. The good news is that the study also proves that weight loss works—measurably reducing liver enzymes in just 12 months. The challenge for healthcare systems is to ensure that every NAFLD patient, regardless of where they receive care, gets the early diagnosis, education, and support they deserve.

Frequently Asked Questions

How common was advanced liver fibrosis in the German FLAG study?

In a study of 507 German patients with fatty liver disease, 1 in 10 had advanced liver fibrosis at their first specialist visit. This means about 10% of patients already had significant liver scarring when they arrived for care. Detecting fibrosis early is important because it affects monitoring and treatment decisions.

Which patients were most likely to have advanced fibrosis?

Patients with advanced fibrosis were older (average age 65 years vs. 48 years) and had larger waistlines around 108 cm. They also had much higher rates of type 2 diabetes (58% vs. 21%) and high blood pressure (79% vs. 44%) compared to patients without significant fibrosis. Previous heart events were also more common.

How often did patients follow exercise recommendations?

The study defined recommended exercise as more than two sessions per week. Among patients without significant fibrosis, 17% met this level; for those with indeterminate stage, 19% did so. But only 6% of patients with advanced fibrosis exercised at the recommended level. About half of all groups reported no exercise at all.

What does the FAST score indicate for future treatments?

The FAST score combines liver stiffness, fat, and AST. Among 107 patients where it could be calculated, 16.8% had a score above 0.67, meaning they likely had progressive NASH. Of patients with advanced fibrosis, 63.6% showed progressive NASH needing possible future medication, compared to 6.1% without significant fibrosis.

What follow-up care did patients receive?

After one year, only 23% of the original 507 patients completed a follow-up visit. About 25% of patients had never received nutritional counseling, even among those with advanced fibrosis. The study highlights inconsistent monitoring and lifestyle support, so asking your doctor for structured follow-up and fibrosis checks is advisable.

When should someone with fatty liver disease seek a second opinion about their fibrosis stage and treatment plan?

Seek a second opinion if your fatty liver disease has not been staged with non-invasive tests such as FIB-4, liver stiffness measurement, or the FAST score. The FLAG study found 1 in 10 patients already had advanced fibrosis at their first specialist visit, and many with advanced disease received only general lifestyle advice. If you have diabetes, high blood pressure, or a large waistline, you are at higher risk. A second opinion can confirm whether more intensive monitoring, nutritional counseling, or emerging medication options are appropriate for your case. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original Article Title: The Fatty Liver Assessment in Germany (FLAG) cohort

Authors: Wolf Peter Hofmann, Peter Buggisch, Lisa Schubert, Nektarios Dikopoulos, Jeannette Schwenzer, Marion Muche, Gisela Felten, Renate Heyne, Patrick Ingiliz, Anna Schmidt, Kerstin Stein, Heiner Wedemeyer, Thomas Berg, Johannes Wiegand, Frank Lammert, Stefan Zeuzem, and Jörn M. Schattenberg

Journal: JHEP Reports, 2020, Volume 2. DOI: https://doi.org/10.1016/j.jhepr.2020.100168

Affiliations: The study was conducted across multiple German institutions, including the Association of Gastroenterologists in Private Practice, Charité Campus Benjamin Franklin Berlin, Hannover Medical School, Leipzig University Hospital, Saarland University Hospital, Goethe University Hospital Frankfurt, and the University Medical Centre of the Johannes Gutenberg-University Mainz.

Funding/Access: This is an open access article under the CC BY license, published by Elsevier B.V. on behalf of the European Association for the Study of the Liver (EASL).

Note: This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and does not constitute medical advice. Patients should consult their healthcare providers regarding their individual condition and treatment options.